Methods for determining surveillance and treatment for diseases

Abstract

Methods, compositions, and devices for use in the early detection of cancer are disclosed herein. A method comprises the steps of sequencing a panel of regions within cell-free nucleic acid molecules and detecting one or more biomarkers that indicate cancer. A method is described herein that comprises the steps of determining the state of biomolecules obtained from a sample derived from a human subject, testing for minimal residual disease (MRD), determining the likelihood of recurrence based on the MRD test, and scheduling one or more additional MRD tests based on the determination of the likelihood of recurrence.

Description

Technical Field

[0001] Cross - Reference to Related Applications This application claims the benefit of U.S. Provisional Patent Application No. 63 / 511,082, filed Jun. 29, 2023, which is hereby incorporated by reference in its entirety.

Background Art

[0002] Background Cancer is a leading cause of disease worldwide. Every year, tens of millions of people are diagnosed with cancer worldwide, and more than half of the patients ultimately die from cancer. In many countries, cancer is the second most common cause of death after cardiovascular disease. For many cancers, early detection leads to improved outcomes.

[0003] To detect cancer, several screening tests are available. General health signs are investigated from physical examinations and medical histories, including confirmation of disease symptoms, such as lumps or other unusual physical symptoms. The patient's health habits and history of past illnesses and treatments are also obtained. Clinical tests are another type of screening test, which may include medical procedures to obtain samples of tissues, blood, urine, or other substances in the body before performing the clinical tests. Imaging procedures screen for cancer by generating visual representations of areas within the body. Genetic tests detect harmful mutations in certain genes associated with some types of cancer. Genetic tests are particularly useful in several diagnostic methods.

Summary of the Invention

Means for Solving the Problems

[0004] Summary of the Invention A method is described herein that includes the steps of determining the state of biomolecules obtained from a sample derived from a human subject, testing for minimal residual disease (MRD), determining the likelihood of recurrence based on the MRD test, and creating a schedule for one or more additional MRD tests based on the determination of the likelihood of recurrence. In other embodiments, the biomolecules are one or more of DNA, methylated DNA, RNA, methylated RNA, proteins, and peptides. In other embodiments, the method includes the steps of preparing a nucleic acid-MBD protein solution by combining a plurality of nucleic acid molecules derived from the subject with a solution containing a certain amount of methyl-binding domain (MBD) protein, and preparing a number of nucleic acid fractions by performing a plurality of washes of the nucleic acid-MBD protein solution with a salt solution, wherein each nucleic acid fraction has a threshold number of methylated cytosines within a plurality of nucleic acid regions having at least a threshold cytosine-guanine content. In other embodiments, one of a plurality of washes is performed using a solution having a certain concentration of sodium chloride (NaCl) to produce a nucleic acid fraction from a number of nucleic acid fractions having a certain range of binding strength to the MBD protein. In other embodiments, the method includes the steps of: determining that a first nucleic acid fraction is associated with a first segment of a plurality of segments of nucleic acid, wherein the first segment corresponds to a first range of binding strength to the MBD protein; attaching a first molecular barcode to the nucleic acid of the first nucleic acid fraction, wherein the first molecular barcode is included in a first set of molecular barcodes associated with the first segment; determining that a second nucleic acid fraction is associated with a second segment of a plurality of segments of nucleic acid, wherein the second segment corresponds to a second range of binding energy to the MBD protein, which is different from a first range of binding strength to the MBD protein; and attaching a second molecular barcode to the nucleic acid of a second nucleic acid fraction, wherein the second molecular barcode is included in a second set of molecular barcodes associated with the second segment.In other embodiments, the method comprises the step of preparing at least a portion of a number of samples to be used to generate sequencing reads by combining at least a portion of the nucleic acid fractions with a certain amount of restriction enzyme that cleaves molecules having one or more unmethylated cytosines, wherein a threshold amount of methylated cytosine corresponds to the minimum frequency of methylated cytosine within a region having at least a threshold cytosine-guanine content.

[0005] In other embodiments, the method includes the step of preparing at least a portion of several samples to be used to generate sequencing reads by combining at least a portion of the said number of nucleic acid fractions with a certain amount of restriction enzyme that cleaves molecules having one or more methylated cytosines, wherein the threshold amount of unmethylated cytosines corresponds to the maximum frequency of uncleaved methylated cytosines within a region having at least a threshold cytosine-guanine content. In other embodiments, the method includes the step of testing for MRD, which includes the steps of sequencing nucleic acid molecules derived from a sample obtained from a subject, analyzing sequence reads derived from the sequencing step to identify one or more driver mutations in the nucleic acid molecules, and using information on the presence, absence, or amount of one or more driver mutations in the nucleic acid molecules to identify tumors in the subject. In other embodiments, the nucleic acid molecules are cell-free DNA. In other embodiments, the sample is at least one of blood, serum, plasma, or tissue. In other embodiments, the method includes determining a treatment for the subject. In other embodiments, the detection limit of the model for determining the tumor fraction of the sample is 0.05% or less. In other embodiments, one or more driver mutations include somatic variants detected at a mutant allele frequency (MAF) of 0.05% or less. In other embodiments, one or more driver mutations include fusions detected at a mutant allele frequency (MAF) of 0.1% or less. In other embodiments, the method includes the step of detecting the mutation distribution of each of the one or more driver mutations, wherein the mutation distribution of each of the one or more driver mutations is detected with a correlation of at least 0.99 to the mutation distribution of driver mutations detected in the subject cohort by histogenetic typing. In other embodiments, the method detects tumors in the subject with at least 85% sensitivity, at least 99% specificity, and at least 99% diagnostic accuracy. In other embodiments, the method includes the step of identifying circulating tumor DNA (ctDNA) and one or more driver mutations in the ctDNA.In other embodiments, the method includes the steps of: obtaining test sequence data from a subject using a computer computing system having one or more hardware processors and memory, wherein the test sequence data includes test sequencing reads derived from a sample of the subject; analyzing the test sequencing reads using the computer computing system to determine a first quantitative measure derived from the test sequencing reads for a genomic region of a reference genome; analyzing the test sequencing reads using the computer computing system to determine a second quantitative measure derived from the test sequencing reads for a genomic region of a reference genome; determining a metric based on the first and second quantitative measures using the computer computing system; generating an input vector including the metric using the computer computing system; and determining an indicator of cancer status in a subject by providing the input vector to a model that implements one or more machine learning techniques to generate an indicator of cancer status in the subject, wherein the model includes weights for individual classification regions of a plurality of classification regions, and at least some of the weights for the individual classification regions are different from each other. In other embodiments, each test sequencing read comprises a nucleotide sequence corresponding to a nucleic acid fragment contained in the sample, and each test sequencing read corresponds to a molecule having a threshold amount of methylated cytosine contained within a region of a nucleotide sequence having at least a threshold cytosine-guanine content; the first quantitative measure is derived from test sequencing reads corresponding to individual taxonomic regions of multiple taxonomic regions, where at least a portion of each taxonomic region of the multiple taxonomic regions corresponds to a genomic region of a reference genome having a threshold amount of methylated cytosine and at least a threshold cytosine-guanine content in subjects in which cancer is detected; the second quantitative measure is derived from test sequencing reads corresponding to individual control regions of multiple control regions, where each control region of the multiple control regions corresponds to an additional genomic region of a reference genome having at least a threshold cytosine-guanine content and at least a threshold amount of methylated cytosine in subjects in which cancer is detected and additional subjects in which cancer is not detected.In other embodiments, the method includes the steps of obtaining training sequence data, comprising training sequence reads derived from a plurality of samples of a plurality of training targets, using a computer computing system having one or more hardware processors and memory, wherein each training sequence read comprises a nucleotide sequence corresponding to a nucleic acid fragment contained in one of the plurality of samples, and each training sequence read corresponds to a molecule having a threshold amount of methylated cytosine contained within a region of a nucleotide sequence having at least a threshold cytosine-guanine content; analyzing the training sequence reads using the computer computing system to determine an additional first quantitative measure derived from the training sequence reads corresponding to individual classification regions of a plurality of classification regions; and analyzing the training sequence reads using the computer computing system to determine a plurality of control The method includes the steps of: determining an additional second quantitative measure derived from training sequence reads corresponding to a region; determining additional metrics for individual classification regions of multiple classification regions based on an additional first quantitative measure for individual classification regions and an additional second quantitative measure for multiple control regions using a computer computing system; generating training data using a computer computing device that includes additional metrics for individual classification regions of multiple classification regions for training sequence reads derived from multiple training target samples; and using the training data to implement one or more machine learning algorithms using a computer computing system to generate a model that determines an index of cancer status in a subject based on the amount of methylated cytosine in at least some of the multiple classification regions. In other embodiments, one or more machine learning algorithms include one or more classification algorithms. In other embodiments, one or more machine learning algorithms include one or more regression algorithms, and the index corresponds to an estimate of the tumor fraction of the sample.In other embodiments, a training sequencing read includes a first portion of the training sequence data, and an additional training sequencing read includes a second portion of the training sequence data, and the additional training sequencing read differs from the training sequencing read in that the method includes the steps of: analyzing at least one of the first portion of the training sequence data or the second portion of the training sequence data by a computer computing system to determine the individual frequencies of a plurality of variants present in individual samples of a plurality of samples; determining, with respect to an individual sample, a variant among the plurality of variants having the highest frequency corresponding to the individual frequency having the highest value among the individual frequencies originating from the individual sample; and determining, with respect to an individual sample, an individual scale of tumor fraction for an individual sample by a computer computing system based on the highest value of the individual frequencies originating from the individual sample. In other embodiments, the training data includes individual scales of tumor fraction for individual samples of a plurality of samples, and the model is generated based on individual scales of tumor fraction for individual samples of a plurality of samples.In other embodiments, the method includes the steps of: generating a data file using a computer system including a processing network and memory, which includes first tokens generated using a first hash function, wherein each first token corresponds to each individual in a group of individuals whose data is stored in a molecular data repository; transmitting the data file to a medical insurance claims data management system using the computer system; obtaining health data corresponding to the group of individuals in response to the data file from the medical insurance claims data management system using the computer system; generating a number of identifiers using a second hash function different from the first hash function, wherein each identifier corresponds to one or more tokens associated with each individual in the group of individuals; obtaining second data from a molecular data repository for the group of individuals using the number of identifiers using the computer system; determining, for the group of individuals, each part of the first data corresponding to each part of the second data; and generating an integrated data repository in which each part of the first data and each part of the second data are stored in association with each identifier of the number of identifiers. In other embodiments, the method includes the steps of: having a computer system determine a first set of data processing instructions that can be executed in relation to first data stored in an integrated data repository; having the computer system execute the first set of data processing instructions to analyze first medical insurance claim codes contained in the first data to determine a first subset of individuals in which a biological condition exists; and having the computer system generate a first dataset representing the subset of individuals in which a biological condition exists.In other embodiments, the method includes the steps of: having a computer system determine a second set of data processing instructions that can be executed in relation to second data stored in an integrated data repository; having the computer system execute the second set of data processing instructions to analyze second medical insurance claim codes contained in the second data to determine one or more treatments to be provided to a second subset of individuals; and having the computer system generate a second dataset showing one or more treatments to be provided to a second subset of individuals. In other embodiments, the method includes the steps of: having a computer system determine a third subset of individuals, which includes a portion of a first subset of individuals that overlaps with a portion of a second subset of individuals; having a computer system receive a request to perform an analysis of the first dataset and the second dataset in relation to the third subset of individuals; and having the computer system, in response to the request, analyze the first dataset and the second dataset for the third subset of individuals to determine a measure of the significance of features of the third subset of individuals with respect to biological status.

[0006] In other embodiments, the method includes the steps of: determining one or more genomic mutations present in a third subset of individuals using a computer computing system; determining a set of treatments to be administered to the third subset of individuals using a computer computing system; and determining the survival rates for each of the third subset of individuals using a computer computing system. In other embodiments, the measure of significance corresponds to the survival rate for one of the treatments and one of the genomic mutations. In other embodiments, the method includes the step of determining the effectiveness of the treatment for the third subset of individuals based on the measure of significance using a computer computing system. In other embodiments, the method includes the step of determining individuals in the third subset of individuals that have not received treatment using a computer computing system. In other embodiments, the method includes administering one or more therapeutically effective doses of treatment to individuals in the third subset that have not received treatment. In other embodiments, the integrated data repository is arranged according to a data repository schema that includes multiple data tables and multiple logical links between the multiple data tables, where each logical link of the multiple logical links points to one or more rows in a data table among the multiple data tables, corresponding to one or more additional rows in an additional data table among the multiple data tables.

[0007] In other embodiments, the data tables include a first data table storing genomics data of a group of individuals; a second data table storing data relating to one or more patient visits by individuals to one or more healthcare providers; a third data table storing information corresponding to each service provided to an individual relating to one or more patient visits to one or more healthcare providers as shown by the second data table; a fourth data table storing personal information of a group of individuals; a fifth data table storing information relating to a health insurance company or government agency that made payments for services provided to the group of individuals; a sixth data table storing information corresponding to health insurance coverage information for a group of individuals; and a seventh data table storing information relating to medical treatments received by a group of individuals. In other embodiments, the number of identifiers generated using the second hash function includes intermediate identifiers, and the method includes the step of applying a salt function to the intermediate identifiers using a computer computing system to generate a final set of identifiers. In other embodiments, the method includes the steps of: using a computer system to obtain information from an additional data repository containing electronic medical records of an additional group of individuals; using a computer system to determine a subset of the additional group of individuals corresponding to a group of individuals whose data is stored in the genomics data repository; and using a computer system to modify the integrated data repository to store at least a portion of the medical record information of the subset of the additional group of individuals in association with the identifier of that number. In other embodiments, the method includes the steps of: using a computer system to perform one or more optical character recognition operations with respect to the additional information; and using a computer system to analyze the additional information obtained from the additional data repository to determine one or more portions of the additional information to be deleted and to create a corpus of information.In other embodiments, the method includes the steps of: using a computer computing system to analyze a corpus of information to determine a portion of an additional group of individuals corresponding to one or more biomarkers; and using a computer computing system to store identifiers for the portion of the additional group of individuals corresponding to one or more biomarkers and to generate one or more data structures that store an index that the portion of the additional group of individuals corresponds to one or more biomarkers.

[0008] In other embodiments, the method includes the steps of: using a computer system to store one or more data structures in an intermediate data repository; and using a computer system to perform one or more de-identification operations with respect to some identifiers of an additional group subset of individuals, and then modifying the integrated data repository to store at least some additional information of medical records of some of the additional group subset of individuals in association with that number of identifiers. In other embodiments, the molecular data repository stores at least one or more of genomic information, genetic information, metabolomics information, transcriptomics information, fragmentomic information, immune receptor information, methylation information, epigenomic information, or proteomics information. In other embodiments, the method includes the step of determining the likelihood of recurrence, including MRD testing, real-world evidence (RWE), or both.

[0009] This specification describes a method comprising the steps of determining the state of biomolecules obtained from a sample derived from a human subject, testing for minimal residual disease (MRD), determining the likelihood of recurrence based on the MRD test, and recommending and / or implementing treatment. In various embodiments, the method described herein determines an assessment including a comprehensive assessment, diagnostic testing, molecular and genetic profiling, and / or risk assessment. In various embodiments, the method described herein determines a treatment plan including patient consultation, treatment strategy, and / or tailored treatment. In various embodiments, the method described herein determines treatment including pre-treatment and / or treatment administration. In various embodiments, the method described herein determines monitoring and / or adjustments including one or more follow-ups and / or response evaluations. In various embodiments, the method described herein determines long-term management and / or survivorship, which may include post-treatment surveillance and / or recurrence management that may support long-term management and / or survivorship.

[0010] A system configured to carry out the method described in any of the prior claims is described herein.

[0011] Computer-readable media comprising the method of any of the prior claims are further described herein.

[0012] Novel features of this disclosure are described in detail in the appended claims. A better understanding of the features and advantages of this disclosure will be obtained by referring to the following detailed description, which includes exemplary embodiments in which the principles of this disclosure are utilized, and to the accompanying drawings (hereinafter also referred to as "fig." and "FIG."). [Brief explanation of the drawing]

[0013] [Figure 1]Figure 1 shows an example architecture for generating an integrated data repository containing multiple types of health management data, following one or more implementations.

[0014] [Figure 2] Figure 2 shows an example of a framework for arranging data tables in an integrated data repository, according to one or more implementations.

[0015] [Figure 3] Figure 3 shows an architecture for generating one or more datasets from information retrieved from a data repository that integrates health-related data from a number of sources, according to one or more implementations.

[0016] [Figure 4] Figure 4 shows an architecture for generating an integrated data repository containing de-identified health insurance claims data and de-identified genomics data, according to one or more implementations.

[0017] [Figure 5] Figure 5 illustrates a framework for generating datasets based on data stored in an integrated data repository using a data pipeline system, according to one or more implementations.

[0018] [Figure 6] Figure 6 is a schematic diagram of the architecture for integrating medical record data into an integrated data repository.

[0019] [Figure 7] Figure 7 is a data flow diagram of an example process for generating an integrated data repository where health insurance claims data and genomics data are stored, according to one or more implementations.

[0020] [Figure 8] Figure 8 is a data flow diagram of an example of a process for generating a number of datasets used to analyze information stored by an integrated data repository in which healthcare claim data and genomics data are stored according to one or more implementations.

[0021] [Figure 9] Figure 9 shows a graphical representation of a machine in the form of a computer system capable of executing a set of instructions for causing a machine to implement any one or more of the methodologies discussed herein according to one or more implementations.

[0022] [Figure 10] Figure 10 shows a graphical representation for adjusting the aggressiveness of patient surveillance based on the likelihood of tumor recurrence obtained from MRD test outcomes using test data and real-world evidence.

[0023] [Figure 11] Figure 11 shows a graphical representation of a treatment plan.

[0024] [Figure 12] Figure 12 shows a graphical representation of treatment implementation.

[0025] [Figure 13] Figure 13 shows a graphical representation of monitoring and adjustment.

[0026] [Figure 14] Figure 14 shows a graphical representation of long-term management and planning.

Best Mode for Carrying Out the Invention

[0027] Detailed Description While various embodiments of the Disclosure are shown and described herein, it will be understood by those skilled in the art that such embodiments are presented merely as examples. Those skilled in the art can conceive of numerous variations, alterations, and substitutions without departing from the Disclosure. It should be understood that various alternatives to the embodiments of the Disclosure described herein are available.

[0028] With respect to a reference number, the term "approximately" and its grammatical equivalents can encompass a range of values ​​up to plus or minus 10% from that value. For example, the quantity "approximately 10" can encompass quantities from 9 to 11. With respect to a reference number, the term "approximately" can encompass a range of values ​​up to plus or minus 10%, 9%, 8%, 7%, 6%, 5%, 4%, 3%, 2%, or 1% from that value.

[0029] With respect to a reference number, the term "at least" and its grammatical equivalents can encompass the reference number and any value greater than that. For example, the quantity "at least 10" can encompass the value of 10 and any number greater than 10, such as 11, 100, and 1,000.

[0030] With respect to a reference number, the term "at most" and its grammatical equivalent can encompass the reference number and any values ​​less than that value. For example, the quantity "at most 10" can encompass the value 10 and any numbers less than 10, such as 9, 8, 5, 1, 0.5, and 0.1.

[0031] As used herein, the singular forms “a,” “an,” and “the” may encompass multiple references unless the context explicitly indicates otherwise. For example, a reference to “a cell” may encompass multiple such cells, and a reference to “the culture” may encompass one or more cultures and their equivalents known to those skilled in the art. All scientific and technical terms used herein may have the same meaning as commonly understood by those skilled in the art in which this disclosure is made, unless explicitly indicated otherwise.

[0032] Current methods either omit testing both genomic and epigenomic attributes of patient samples, or perform multiple tests separately. Omitting genomic or epigenomic information may lead to the prescription of cancer treatments that would have been known to be ineffective, or to the withholding of cancer treatments that would have been known to be effective, if both genomic and epigenomic information were available. Cancer can be indicated by epigenetic variations, such as methylation. An example of methylation changes in cancer is the localized acquisition of DNA methylation in CpG islands at transcription start sites (TSSs) of genes involved in normal growth regulation, DNA repair, cell cycle regulation, and / or cell differentiation. This hypermethylation may be associated with an abnormal loss of transcriptional ability of the genes involved and occurs at least as frequently as point mutations and deletions as a cause of altered gene expression. DNA methylation profiling can be used to detect regions of the genome with different degrees of methylation ("differential methylation regions" or "DMRs") that are altered during development or perturbed by disease, such as cancer or any cancer-related disease. The genomes of cancer cells exhibit imbalances in the DNA methylation patterns described above, and therefore in the functional packaging of DNA. Thus, abnormalities in chromatin organization, when analyzed together with methylation changes, can contribute to the enhancement of cancer profiling. MBD distribution and fragment mix data, such as fragments mapped to start and stop positions (correlated with nucleosome locations), fragment length, and associated nucleosome occupancy, can be used for chromatin structure analysis in hypermethylation studies with the aim of improving biomarker detection rates.

[0033] Methylation profiling can involve determining methylation patterns across different regions of the genome. For example, molecules can be partitioned based on their degree of methylation (e.g., the relative number of methylation sites per molecule), sequenced, and then the sequences of molecules in different partitions can be mapped to a reference genome. This can reveal regions of the genome that are more or less methylated than other regions. Thus, genomic regions can have different degrees of methylation, in contrast to individual molecules.

[0034] A defining characteristic of nucleic acid molecules is modification, which can include various chemical modifications or protein modifications (i.e., epigenetic modifications). Non-exclusive examples of chemical modifications include, but are not limited to, covalent DNA modifications, including DNA methylation. In some embodiments, DNA methylation involves the addition of a methyl group to cytosine at a CpG site (where guanine follows cytosine in the nucleic acid sequence). In some embodiments, DNA methylation involves the addition of a methyl group to adenine, such as N6-methyladenine. In some embodiments, DNA methylation is 5-methylation (modification of the fifth carbon in the six-membered carbon ring of cytosine). In some embodiments, 5-methylation involves the addition of a methyl group to the 5C position of cytosine, thereby creating 5-methylcytosine (m5c). In some embodiments, methylation includes derivatives of m5c. Derivatives of m5c include, but are not limited to, 5-hydroxymethylcytosine (5-hmC), 5-formylcytosine (5-fC), and 5-carboxylcytosine (5-caC). In some embodiments, DNA methylation is 3C methylation (modification of the third carbon of the 6-membered carbon ring of cytosine). In some embodiments, 3C methylation involves the addition of a methyl group to the 3C position of cytosine, thereby producing 3-methylcytosine (3mC). Other examples include N6-methyladenine or glycosylation. DNA methylation involves the addition of a methyl group to DNA (e.g., CpG), which can alter the expression of methylated DNA regions. Methylation can also occur at non-CpG sites; for example, methylation can occur at CpA, CpT, or CpC sites. DNA methylation can alter the activity of methylated DNA regions. For example, methylation of DNA within a promoter region can suppress gene transcription. DNA methylation is crucial for normal development, and abnormalities in methylation can disrupt epigenetic regulation. This disruption, or suppression, of epigenetic regulation may lead to diseases such as cancer. Promoter methylation in DNA can be an indicator of cancer.

[0035] A CpG dyad is a dinucleotide CpG (cytosine-phosphate-guanine, i.e., cytosine followed by guanine in the 5'→3' direction of the nucleic acid sequence) on the sense strand of a double-stranded DNA molecule and its complementary CpG on the antisense strand. CpG dyads can be fully methylated or semi-methylated (methylated on only one strand).

[0036] CpG dinucleotides are relatively rare in the normal human genome, and the vast majority of CpG dinucleotide sequences are transcriptionally inactive (e.g., in DNA heterochromatin regions and repeat elements around the centromere of chromosomes) and are also methylated. However, many CpG islands are protected from such methylation, particularly around transcription start sites (TSSs).

[0037] Protein modifications include binding to chromatin components, particularly histones, including their modified forms, and binding to other proteins, such as proteins involved in replication or transcription. This disclosure provides a method for processing and analyzing nucleic acids with varying degrees of modification, wherein the nature of the original modification of the nucleic acid is correlated with a nucleic acid tag, which can be deciphered by sequencing during nucleic acid analysis. The genetic variation of the sample nucleic acid modification can then be associated with the degree of modification of that nucleic acid in the original sample (epigenetic variation). This includes single-stranded (e.g., ssDNA or RNA) or double-stranded (e.g., dsDNA) molecules.

[0038] DNA loss can reduce the presence of one or more types of DNA, making it difficult to detect the presence of one or more types of DNA, such as cfDNA. In one or more additional scenarios, existing methods for measuring DNA methylation, such as enrichment or depletion methods, may have relatively high resolutions, such as about 100 base pairs (bp) to about 200 bp, but it may be difficult to accurately determine the amount of DNA methylation. The accuracy of determining DNA methylation may affect the accuracy of the tumor fraction estimate of a sample. Since tumor fraction can be used to determine whether a sample originates from an object in which a tumor is present, the accuracy of determining the tumor fraction estimate may affect diagnostic and / or treatment decisions for an individual.

[0039] Beyond these detection schemes, more data is needed to understand tumor behavior and the performance of treatments and guidelines, often outside the highly selective framework of randomized controlled trials, which are frequently designed and implemented by entities with a commercial interest in their success. Real-world evidence (RWE), particularly the use of databases featuring the integration of clinical and molecular data, is playing an increasingly important role in high-precision oncology research. However, the vast majority of these databases feature tumor-derived genomic information limited to a single point in time, generally at the time of diagnosis, partly due to the practical difficulty of genomic profiling of consecutive tumor specimens in real-world clinical practice. Despite evidence that treatments can significantly alter the tumor genomic landscape and lead to drug resistance, tumor genomic data is often limited to those that have not received systemic treatment. Combining data from liquid biopsy assays with rich clinical information can overcome these challenges, aid in improving our understanding of tumor evolution and the emergence of biomarkers conferring resistance, and guide the development of novel therapeutics to address areas of unmet needs.

[0040] Analysis of health management data using existing systems and techniques is typically performed with respect to medical records created by healthcare providers. As used herein, a healthcare provider may refer to an entity, individual, or group of individuals engaged in providing care to an individual in connection with the treatment or prevention of at least one of one or more biological conditions. Furthermore, as used herein, a biological condition may refer to an abnormality in function and / or structure in an individual to the extent that a detectable feature of the abnormality occurs or is likely to occur. A biological condition may be characterized by external and / or internal features, signs, and / or symptoms that indicate a deviation from the biological standard in one or more populations. In various examples, a biological condition may include one or more molecular phenotypes. For example, a biological condition may correspond to a genetic or epigenetic lesion. In one or more additional examples, a biological condition may include at least one of one of the following: one or more diseases, one or more disorders, one or more injuries, one or more syndromes, one or more disability, one or more infections, one or more single symptoms, or other atypical variations of the biological structure and / or function of an individual. Furthermore, a treatment may, as used herein, refer to a substance, procedure, routine, device, and / or other intervention that can be administered or performed with the intention of treating one or more effects of a biological condition in an individual. In one or more examples, a treatment may include a substance metabolized by the individual. A substance may include a composition of a substance, e.g., a pharmaceutical composition. A substance can be delivered to an individual by several means, such as ingestion, injection, absorption, or inhalation. A treatment may also include a physical intervention, such as one or more surgical procedures. In at least some examples, a treatment may include a therapeutically meaningful intervention.

[0041] Healthcare data typically analyzed by existing systems includes unstructured data. Unstructured data may include data that is not organized according to a predefined or standardized format. For example, unstructured data may include notes made by healthcare providers, consisting of free text. That is, the format in which the notes are taken does not include predefined inputs that the healthcare provider can select, such as drop-down menus or lists. Rather, the notes consist of text written by the healthcare provider, which may include sentences, sentence fragments, words, letters, symbols, abbreviations, or one or more combinations thereof. In some cases, unstructured data may be partially structured. For example, a provider may select an insurance billing code from a predefined list of insurance billing codes and add unstructured notes to the data associated with that billing code.

[0042] Existing systems typically expend significant computing resources to analyze unstructured data in order to extract information relevant to the analysis performed by the system. In some cases, existing systems may analyze unstructured data and convert it to a structured format to facilitate analysis of previously unstructured data. Analysis of unstructured data by existing systems can be not only inefficient but also inaccurate. In scenarios where unstructured data is obtained from health management data, the accuracy of the analysis is crucial because the analysis may relate to at least one treatment or diagnosis for a certain number of individuals regarding one or more biological conditions. Therefore, inaccurate analysis of health management data could result in suboptimal treatment for individuals.

[0043] The techniques, architectures, frameworks, systems, processes, and implementations of computer-readable instructions described herein are intended for analyzing health insurance claim data to derive information about at least one aspect of an individual's health or treatment. In contrast to existing systems, health insurance claim data is structured according to one or more formats and stored in a number of data tables. The data tables may contain codes or other alphanumeric information indicating treatments received by an individual, treatment dates, dosage information, diagnoses of the individual for one or more biological conditions, information about visits to healthcare providers, dates of visits to healthcare providers, billing information, etc. Using the implementations described herein, health insurance claim data can be accurately analyzed for hundreds, thousands, tens of thousands, or even more individuals who have one or more biological conditions. In various examples, rows and / or columns of health insurance claim data up to tens of thousands, hundreds of thousands, or millions can be analyzed to determine health-related information about individuals who have one or more biological conditions.

[0044] In various examples, the implementations described herein can integrate molecular data with health insurance claims data. Molecular data may include information derived from tissue samples extracted from a number of individuals. Molecular data may also include information derived from blood samples extracted from a number of individuals. In one or more exemplary examples, molecular data may include genomics data. Furthermore, in one or more examples, health insurance claims data can be integrated with germline genetic information for a number of individuals.

[0045] An integrated data repository can be created that combines medical insurance claim data and molecular data for an individual. In one or more examples, an identifier can be generated for an individual that is associated with both the individual's medical insurance claim data and its molecular data. Both the molecular data and medical insurance claim data stored in the integrated data repository can be accessed using a single identifier for the individual. In one or more exemplary examples, the identifier for an individual may include an encryption security key. In various examples, the integrated data repository may include a number of data tables corresponding to different aspects of the data stored in the data repository. For example, a first data table may be created to include summary data about the individual included in the integrated data repository, such as personal information, and a second data table may be created to include data corresponding to visits to healthcare providers. Furthermore, a third data table may be created to show the medical procedures performed on the individual, and a fourth data table may be created to show information about prescriptions received by the individual. Furthermore, a fifth data table may be created to include multi-omics profiling of the individual. A multi-omics profile may include at least one of the following: a genomic profile, a transcriptomics profile, an epigenetic profile, or a proteomics profile.

[0046] Data tables contained within an integrated data repository can be linked by logical links. Thus, a query that retrieves information from one data table can retrieve information from one or more additional data tables. By accessing the information stored in the linked data tables, several different datasets can be generated that can be used to analyze the information stored in the integrated data repository. For example, the information stored in the integrated data repository can be analyzed by one or more algorithms to generate datasets organized according to one or more schemas. These datasets may show the treatments an individual received for a biological condition over a period of time. The datasets may also show a cohort of individuals with several common characteristics contained within the integrated data repository. In various examples, datasets can organize, integrate, and arrange information from several different data sources, including the integrated data repository. Analyzing datasets in relation to several queries can reveal information that may be of interest to at least one of the following: healthcare providers, patients, or providers of treatment for biological conditions. For example, analyzing one or more datasets can more accurately determine the survival rate of individuals who have a biological condition and possess a specific genomic profile in response to receiving a particular treatment.

[0047] The implementation described herein can provide a platform for integrating individual health insurance claims data and molecular data, which is not found in existing systems that typically rely on electronic health records containing a certain amount of unstructured data. By generating and analyzing structured health insurance claims data integrated with molecular data, the implementation described herein can provide more accurate characterization of the integrated data compared to existing systems that rely on relatively inaccurate unstructured electronic health record data. Furthermore, the implementation described herein generates analysis-ready datasets that enable the analysis of individual health information in a confidential and anonymized manner. sample

[0048] The sample may be any biological sample isolated from the subject. The sample may be a body sample. The sample may include body tissues, such as known or suspected solid tumors, whole blood, platelets, serum, plasma, stool, red blood cells, white blood cells or leucocytes, endothelial cells, tissue biopsy material, cerebrospinal fluid, synovial fluid, lymph, ascites, interstitial fluid or extracellular fluid, intercellular fluid including gingival crevicular exudate, bone marrow, pleural fluid, cerebrospinal fluid, saliva, mucus, sputum, semen, sweat, and urine. Preferably, the sample is a body fluid, particularly blood and its fractions, as well as urine. The sample may be in the form originally isolated from the subject, or may have been subjected to further processing to remove or add components such as cells, or to enrich one component with another. Therefore, preferred body fluids for analysis are plasma or serum containing cell-free nucleic acids. The sample can be isolated or obtained from the subject and transported to the sample analysis site. Samples can be stored and transported at a desired temperature, e.g., room temperature, 4°C, -20°C, and / or -80°C. Samples can be isolated or obtained from subjects at the sample analysis site. Subjects may be humans, mammals, animals, companion animals, service animals, or pets. Subjects may have cancer. Subjects may not have cancer or detectable cancer symptoms. Subjects may have been treated with one or more cancer treatments, e.g., chemotherapy, antibodies, vaccines, or biological agents. Subjects may be in remission. Subjects may or may not have been diagnosed as susceptible to cancer or any cancer-related genetic mutation / disorder.

[0049] The volume of plasma may depend on the desired read depth of the region to be sequenced. Exemplary volumes are 0.4–40 ml, 5–20 ml, and 10–20 ml. For example, the volume could be 0.5 mL, 1 mL, 5 mL, 10 mL, 20 mL, 30 mL, or 40 mL. The volume of sampled plasma may be 5–20 mL.

[0050] The sample may contain varying amounts of nucleic acids, including genome equivalents. For example, a sample of approximately 30 ng of DNA may contain approximately 10,000 (10⁴) haploid human genome equivalents, or approximately 200 billion (2 × 10¹¹) individual polynucleotide molecules in the case of cfDNA. Similarly, a sample of approximately 100 ng of DNA may contain approximately 30,000 haploid human genome equivalents, or approximately 600 billion individual molecules in the case of cfDNA.

[0051] The sample may include nucleic acids from different sources, e.g., nucleic acids and cell-free nucleic acids from the same target cells, or nucleic acids and cell-free nucleic acids from different target cells. The sample may include nucleic acids with mutations. For example, the sample may include DNA with germline mutations and / or somatic mutations. Germline mutations refer to mutations present in the germline DNA of the target. Somatic mutations refer to mutations originating from somatic cells of the target, e.g., cancer cells. The sample may include DNA with cancer-associated mutations (e.g., cancer-associated somatic mutations). The sample may include epigenetic variants (i.e., chemical or protein modifications), where the epigenetic variants are associated with the presence of genetic variants such as cancer-associated mutations. In some embodiments, the sample includes epigenetic variants associated with the presence of genetic variants, where the sample does not include genetic variants.

[0052] Exemplary amounts of cell-free nucleic acids in the sample before amplification range from approximately 1 fg to approximately 1 μg, for example, 1 pg to 200 ng, 1 ng to 100 ng, and 10 ng to 1000 ng. For example, the amount may be up to approximately 600 ng, up to approximately 500 ng, up to approximately 400 ng, up to approximately 300 ng, up to approximately 200 ng, up to approximately 100 ng, up to approximately 50 ng, or up to approximately 20 ng of cell-free nucleic acid molecules. The amount may be at least 1 fg, at least 10 fg, at least 100 fg, at least 1 pg, at least 10 pg, at least 100 pg, at least 1 ng, at least 10 ng, at least 100 ng, at least 150 ng, or at least 200 ng of cell-free nucleic acid molecules. The quantity may be up to 1 femtogram (fg), 10 fg, 100 fg, 1 picogram (pg), 10 pg, 100 pg, 1 ng, 10 ng, 100 ng, 150 ng, or 200 ng of cell-free nucleic acid molecules. The method may include obtaining 1 femtogram (fg) to 200 ng.

[0053] Cell-free nucleic acids are nucleic acids that are not contained within cells and are not bound to cells in any other form, or in other words, nucleic acids that remain in a sample after intact cells have been removed. Cell-free nucleic acids include DNA, RNA, and hybrids thereof, including genomic DNA, mitochondrial DNA, siRNA, miRNA, circulating RNA (cRNA), tRNA, rRNA, small nucleolar RNA (snoRNA), Piwi-interacting RNA (piRNA), long non-coding RNA chains (long ncRNA), or fragments of any of these. Cell-free nucleic acids may be double-stranded, single-stranded, or hybrids thereof. Cell-free nucleic acids may be released into body fluids by secretion or cell death processes, such as cell necrosis and apoptosis. Some cell-free nucleic acids are released into body fluids from cancer cells, such as circulating tumor DNA (ctDNA). Other cell-free nucleic acids are released from healthy cells. In some embodiments, cfDNA is cell-free embryonic DNA (cffDNA). In some embodiments, cell-free nucleic acids are produced by tumor cells. In some embodiments, cell-free nucleic acids are produced by a mixture of tumor cells and non-tumor cells.

[0054] Cell-free nucleic acids have an exemplary size distribution of approximately 100–500 nucleotides, with molecules of 110–approximately 230 nucleotides accounting for about 90% of the molecules, the mode being approximately 168 nucleotides, and a second minor peak in the range of 240–440 nucleotides. Cell-free nucleic acids can be isolated from body fluids by a fractionation or partitioning step that separates the cell-free nucleic acids found in solution from intact cells and other insoluble components of the body fluid. Partitioning may involve techniques such as centrifugation or filtration. Alternatively, cells in the body fluid can be lysed and the cell-free and cellular nucleic acids processed together. Generally, after buffer addition and washing steps, nucleic acids can be precipitated using alcohol. Furthermore, contaminants or salts can be removed using a cleanup step, such as a silica-based column. In certain aspects of the procedure, for example to optimize yield, nonspecific bulk carrier nucleic acids, such as Cot-1 DNA, DNA, or proteins for bisulfite sequencing, hybridization, and / or ligation, can be added throughout the reaction.

[0055] After such processing, the sample may contain various forms of nucleic acids, including double-stranded DNA, single-stranded DNA, and single-stranded RNA. In some embodiments, single-stranded DNA and RNA can be converted to double-stranded form for subsequent processing and analysis steps.

[0056] specimen The specimen may include nucleic acid specimens and non-nucleic acid specimens. This disclosure provides for the detection of genetic variations in biological specimens derived from the subject. The biological specimen may include polynucleotides derived from cancer cells. Polynucleotides may be DNA (e.g., genomic DNA, cDNA), RNA (e.g., mRNA, small RNA), or any combination thereof. The biological specimen may include tumor tissue derived from a biopsy, for example. In some cases, the biological specimen may include blood or saliva. In certain cases, the biological specimen may include cell-free DNA ("cfDNA") or circulating tumor DNA ("ctDNA"). Cell-free DNA may be present in blood, for example.

[0057] Examples of non-nucleic acid samples include, but are not limited to, lipids, carbohydrates, peptides, proteins, glycoproteins (N-linked or O-linked), lipoproteins, phosphorylated proteins, specific phosphorylated or acetylated variants of proteins, amidated variants of proteins, hydroxylated variants of proteins, methylated variants of proteins, ubiquitinated variants of proteins, sulfated variants of proteins, viral proteins (e.g., viral capsids, viral envelopes, viral coats, viral accessories, viral glycoproteins, viral spikes, etc.), extracellular and intracellular proteins, antibodies, and antigen-binding fragments. Non-nucleic acid samples include receptors, antigens, surface proteins, transmembrane proteins, surface antigen classification proteins, protein channels, protein pumps, carrier proteins, phospholipids, glycoproteins, glycolipids, intercellular interaction protein complexes, antigen presentation complexes, major histocompatibility complexes, engineered T cell receptors, T cell receptors, B cell receptors, chimeric antigen receptors, extracellular matrix proteins, post-translational modification states of cell surface proteins (e.g., phosphorylation, glycosylation, ubiquitination, nitrosylation, methylation, acetylation, or lipid addition), gap junctions, and adhesion junctions.

[0058] In general, any number of samples, including both nucleic acid and non-nucleic acid samples, can be analyzed using the systems, apparatus, methods, and compositions. For example, the number of samples analyzed may be at least about 2, at least about 3, at least about 4, at least about 5, at least about 6, at least about 7, at least about 8, at least about 9, at least about 10, at least about 11, at least about 12, at least about 13, at least about 14, at least about 15, at least about 20, at least about 25, at least about 30, at least about 40, at least about 50, at least about 100, at least about 1,000, at least about 10,000, at least about 100,000 or more different samples present within a region of the sample or within individual features of the substrate. Methods for performing multiplexed assays to analyze two or more different samples will be discussed in later sections of this disclosure.

[0059] One or more nucleic acid and / or non-nucleic acid samples constitute a set of intermolecular interactions in the biological system under test (e.g., cells), which can be thought of as an "interactome"—intermolecular interactions that occur between molecules belonging to different biochemical families (proteins, nucleic acids, lipids, carbohydrates, etc.) and also between molecules within a given family. In various embodiments, the interactome is a protein-DNA interactome (a network formed by transcription factors (and DNA or chromatin regulatory proteins) and their target genes). In other embodiments, the interactome refers to a protein-protein interaction network (PPI) or protein-protein interaction network (PIN). The methods described herein enable the testing and analysis of interactomes. Techniques such as proteogenomics (e.g., whole-genome sequencing, whole-exome sequencing and RNA-seq, and mass spectrometry) can support testing of interactomes.

[0060] analysis This method can be used to diagnose a condition in a subject, particularly the presence of cancer; to characterize a condition (e.g., to stage cancer or determine cancer heterogeneity); to monitor the response to treatment of a condition; and to determine the risk of developing a condition or the prognosis of the subsequent course of the condition. This disclosure may also be useful in determining the efficacy of a particular treatment option. A successful treatment option may increase the amount of copy number variation or rare mutations detected in the subject's blood, as more cancer cells may be killed and DNA may be shed if the treatment is successful. In other cases, this may not occur. In another case, a particular treatment option may correlate over time with the genetic profile of the cancer. This correlation may be useful in selecting a treatment. Furthermore, if the cancer is observed to be in remission after treatment, this method can be used to monitor residual disease or disease recurrence.

[0061] The types and number of cancers that can be detected include blood cancers, brain cancers, lung cancers, skin cancers, nasal cancers, throat cancers, liver cancers, bone cancers, lymphomas, pancreatic cancers, intestinal cancers, rectal cancers, thyroid cancers, bladder cancers, kidney cancers, oral cancers, stomach cancers, solid tumors, heterogeneous tumors, and homogeneous tumors. The type and / or stage of cancer can be detected from genetic variations, including mutations, rare mutations, indels, copy number variations, base transpositions, translocations, inversions, deletions, aneuploidy, partial aneuploidy, polyploidy, chromosomal instability, chromosomal structural changes, gene fusions, chromosome fusions, gene shortening, gene amplification, gene duplication, chromosomal damage, DNA damage, abnormal changes in nucleic acid chemical modifications, abnormal changes in epigenetic patterns, and abnormal changes in nucleic acid 5-methylcytosine.

[0062] Genetic and other specimen data can also be used to characterize specific forms of cancer. Cancers are often heterogeneous in both composition and staging. Genetic profiling data may enable the characterization of specific subtypes of cancer, which may be important in the diagnosis or treatment of that particular subtype. This information may also provide subjects or practitioners with clues regarding prognosis for specific types of cancer, enabling either the subject or practitioner to adapt treatment options as the disease progresses. Some cancers may progress and become more invasive and genetically unstable. Other cancers may remain benign, inactive, or quiescent. The systems and methods of this disclosure may be useful in determining disease progression.

[0063] This analysis is also useful in determining the efficacy of specific treatment options. A successful treatment option may increase the amount of copy number variations or rare mutations detected in the subject's blood, as more cancers may be killed and DNA may be released if the treatment is successful. In other cases, this may not occur. In other cases, a particular treatment option may correlate with the cancer's genetic profile over time. This correlation may be useful in selecting a treatment. Furthermore, if the cancer is observed to be in remission after treatment, this method can be used to monitor residual disease or disease recurrence.

[0064] This method can also be used to detect genetic variations in non-cancerous conditions. Immune cells, such as B cells, can undergo rapid clonal expansion in the presence of certain diseases. Clonal expansion can be monitored using copy number variation detection, and a particular immune state can be monitored. In this example, copy number variation analysis can be performed over time to generate a profile of how a particular disease may progress. Using the detection of copy number variations, or even rare mutations, it is possible to determine how a population of pathogens changes during the course of infection. This can be particularly important during chronic infections, such as HIV / AIDS or hepatitis infections, where the virus may change its life cycle and / or mutate into a more virulent form during the course of infection. This method can be used to determine or profile the rejection activity of the host body as immune cells attempt to destroy transplanted tissue, thereby monitoring the status of the transplanted tissue and modifying strategies for treating or preventing rejection.

[0065] Furthermore, the methods of this disclosure can be used to characterize heterogeneity of an abnormal condition in a subject. Such a method may include, for example, a step of generating a gene profile of extracellular polynucleotides derived from the subject, wherein the gene profile includes multiple data obtained from the analysis of copy number variations and rare mutations. In some embodiments, the abnormal condition is cancer. In some embodiments, the abnormal condition may result in a heterogeneous genomic population. In the example of cancer, it is known that several tumors may contain tumor cells at different stages of cancer. In other examples, heterogeneity may include multiple lesions of the disease. Again, in the example of cancer, multiple tumor lesions may be present, and perhaps one or more lesions are the result of metastases that have spread from the primary site.

[0066] This method can be used to generate or profile fingerprints or sets of data, which are summaries of genetic information derived from different cells in heterogeneous diseases. These data sets may include, alone or in combination with, analysis of copy number variations and mutations.

[0067] This method can be used to diagnose, prognose, monitor, or observe cancer or other diseases. In some embodiments, the methods described herein do not involve diagnosing, prognosing, or monitoring a fetus, and therefore do not apply to non-invasive prenatal testing. In other embodiments, these methodologies can be used to diagnose, prognose, monitor, or observe cancer or other diseases in unborn subjects, in pregnant subjects, whose DNA and other polynucleotides may co-circulate with maternal molecules.

[0068] Determination of the 5-methylcytosine pattern of nucleic acids Bisulfite-based sequencing and its variations provide means for determining the methylation patterns of nucleic acids. In some embodiments, determining the methylation pattern includes distinguishing 5-methylcytosine (5mC) from unmethylated cytosine. In some embodiments, determining the methylation pattern includes distinguishing N6-methyladenine from unmethylated adenine. In some embodiments, determining the methylation pattern includes distinguishing 5-hydroxymethylcytosine (5hmC), 5-formylcytosine (5fC), and 5-carboxylcytosine (5caC) from unmethylated cytosine. Examples of bisulfite sequencing, but not limited to these, include oxidative bisulfite sequencing (OX-BS-seq), Tet-assisted bisulfite sequencing (TAB-seq), and reductive bisulfite sequencing (redBS-seq).

[0069] Oxidative bisulfite sequencing (OX-BS-seq) is used to distinguish between 5mC and 5hmC by first converting 5hmC to 5fC and then proceeding with bisulfite sequencing as previously described. Tet-assisted bisulfite sequencing (TAB-seq) can also be used to distinguish between 5mC and 5hmC. In TAB-seq, 5hmC is protected by glucosylation. Then, 5mC is converted to 5caC using the Tet enzyme, and then bisulfite sequencing is proceeded as previously described. Reductive bisulfite sequencing is used to distinguish between 5fC and modified cytosine.

[0070] Generally, bisulfite sequencing involves dividing a nucleic acid sample into two aliquots and treating one aliquot with bisulfite. Bisulfite converts native cytosines and certain modified cytosine nucleotides (e.g., 5-formylcytosine or 5-carboxylcytosine) to uracil, while other modified cytosines (e.g., 5-methylcytosine, 5-hydroxymethylcytosine) are not. Comparison of the nucleic acid sequences of molecules from the two aliquots reveals which cytosines were converted to uracil and which were not. Consequently, modified and unmodified cytosines can be determined. Initially dividing the sample into two aliquots is inconvenient for samples containing only small amounts of nucleic acids and / or composed of heterogeneous cell / tissue origins, such as body fluids containing cell-free DNA.

[0071] This disclosure provides bisulfite sequencing and methods therefor, in which variations thereof are permitted. These methods function by linking nucleic acids in a population to a capture moiety, i.e., a label that can be captured or immobilized. Examples of capture moieties include, but are not limited to, biotin, avidin, streptavidin, nucleic acids containing specific nucleotide sequences, haptens recognized by antibodies, and magnetically attractable particles. The extraction moiety may be a member of a binding pair, such as biotin / streptavidin or hapten / antibody. In some embodiments, the capture moiety attached to the sample is captured by its binding pair attached to an isolateable portion, such as a magnetically attractable particle or a large particle that can be settled by centrifugation. The capture moiety may be any type of molecule that enables affinity separation of nucleic acids having the capture moiety from nucleic acids lacking the capture moiety. Exemplary capture moieties are biotin that enables affinity separation by binding to streptavidin linked to or linkable to a solid phase, or oligonucleotides that enable affinity separation by binding to complementary oligonucleotides linked to or linkable to a solid phase. After the capture portion is attached to the sample nucleic acid, the sample nucleic acid functions as an amplification template. After amplification, the original template remains attached to the capture portion, but the amplicon is not attached to the capture portion.

[0072] The capture portion can be ligated to the sample nucleic acid as a component of the adapter, thereby also providing amplification and / or sequencing primer binding sites. In some methods, both ends of the sample nucleic acid are ligated to the adapter, and both adapters have capture portions. Preferably, any cytosine residues in the adapter are modified with, for example, 5-methylcytosine to protect them from the action of bisulfites. In some cases, the capture portion is ligated to the original template by a cleavable linkage (e.g., a photocleavable desthiobiotin-TEG or a uracil residue cleavable with USER® enzyme, Chem. Commun. (Camb). 2015 Feb 21; 51(15): 3266-3269), in which case the capture portion can be removed as desired.

[0073] The amplicon is denatured and brought into contact with an affinity reagent for the capture tag. The original template binds to the affinity reagent, but the nucleic acid molecules produced by amplification do not. Therefore, the original template can be separated from the nucleic acid molecules produced by amplification.

[0074] After separation or segregation, each population of nucleic acids (i.e., the original template and the amplified product) can be subjected to bisulfite treatment, with the original template population being treated and the amplified product not. Alternatively, the amplified product can be subjected to bisulfite treatment, while the original template population is not. After such treatment, each population can be amplified (in the case of the original template population, uracil is converted to thymine). The populations can also be subjected to biotin probe hybridization for enrichment. Then, each population is analyzed and its sequences are compared to determine which cytosines were 5-methylated (or 5-hydroxymethylated) in the original population. Unmodified C is indicated by the detection of T nucleotides (corresponding to unmethylated cytosines converted to uracil) in the template population and C nucleotides at the corresponding positions in the amplified population. Modified C is indicated in the original sample by the presence of C at the corresponding positions in the original template and the amplified population.

[0075] In some embodiments, the method utilizes sequential DNA-seq and bisulfite-seq (BIS-seq) NGS library preparation of molecularly tagged DNA libraries. This process is carried out by labeling an adapter (e.g., biotin), DNA-seq amplification of the entire library, parent molecule recovery (e.g., streptavidin bead pulldown), bisulfite conversion, and BIS-seq. In some embodiments, the method identifies 5-methylcytosine at single-nucleotide resolution by sequential NGS preparative amplification of parent library molecules with and without bisulfite treatment. This can be achieved by modifying one of the two adapter strands of the 5-methylated NGS adapter (directional adapter; Y-shaped / fork-shaped due to the substitution of 5-methylcytosine) used in BIS-seq with a label (e.g., biotin). The adapter is then ligated to the sample DNA molecule and amplified (e.g., by PCR). Since only the parent molecule has a labeled adapter end, the parent molecule can be selectively recovered from its amplified offspring by a label-specific capture method (e.g., streptavidin-magnetic beads). Because the parent molecule retains a 5-methylation mark, bisulfite conversion of the captured library provides 5-methylation status at single-nucleotide resolution during BIS-seq, and the molecular information is preserved in the corresponding DNA-seq. In some embodiments, enrichment / NGS can be performed in a standard multiplexed NGS workflow by combining the bisulfite-treated library with an untreated library and then adding a sample-tagged DNA sequence. Bioinformatics analysis can be performed for genomic alignment and 5-methylated base identification, as in the BIS-seq workflow. In short, this method provides the ability to selectively recover ligated molecules with the parent 5-methylcytosine mark after library amplification, thereby enabling parallel processing of bisulfite-converted DNA. This overcomes the destructive nature of bisulfite treatment on the quality / sensitivity of the DNA-seq information extracted by the workflow.This method allows for the parallel application of complete DNA library amplification and processes to extract epigenetic DNA modifications using recovered, ligated parental DNA molecules (via labeled adapters). While this disclosure discusses, but is not limited to, the use of the BIS-seq method for identifying cytosine 5-methylation (5-methylcytosine), variations of BIS-seq have been developed to identify hydroxymethylated cytosine (5hmC; OX-BS-seq, TAB-seq), formylcytosine (5fC; redBS-seq), and carboxylcytosine. These methodologies can be implemented in conjunction with the sequential / parallel library preparation described herein.

[0076] Alternative methods for modified nucleic acid analysis This disclosure provides alternative methods for analyzing modified nucleic acids (e.g., methylation, histone linkage, and other modifications described above). Some such methods involve contacting a population of nucleic acids having different degrees of modification (e.g., 0, 1, 2, 3, 4, 5, or more methyl groups per nucleic acid molecule) with an adapter, and then fractionating the population according to the degree of modification. The adapter is attached to either one or both ends of the nucleic acid molecules in the population. Preferably, the adapter contains a sufficient number of different tags such that the probability of two nucleic acids having the same start and end points receiving the same tag combination is low (e.g., 95%, 99%, or 99.9%). After the adapter is attached, the nucleic acid is amplified from a primer that binds to a primer-binding site in the adapter. The adapter may contain the same or different primer-binding sites, whether they have the same or different tags, but preferably the adapter contains the same primer-binding site. After amplification, the nucleic acids are brought into contact with an activator that preferably binds to modified nucleic acids (e.g., such activators previously described). The nucleic acids are separated from binding to the activator into at least two sets of nucleic acids with different degrees of modification. For example, if the activator has affinity for modified nucleic acids, nucleic acids with a high degree of modification (compared to the median degree of modification in the population) will preferentially bind to the activator, while nucleic acids with a low degree of modification will not bind to the activator or will elute more easily from the activator. After separation, the different sets can then be subjected to further processing steps, which typically include further amplification and sequence analysis, performed in parallel but separately. The sequence data from the different sets can then be compared.

[0077] Both ends of a nucleic acid can be ligated to a Y-shaped adapter containing a primer-binding site and a tag. The molecule is amplified. The amplified molecule is then fractionated by contacting it with an antibody that preferentially binds to 5-methylcytosine to produce two fractions. One fraction contains the original molecule lacking methylation and the amplified copy with lost methylation. The other fraction contains the original DNA molecule with methylation. The two fractions are then processed and sequenced separately, with further amplification of the methylated fraction. The sequence data of the two fractions can then be compared. In this example, the tag is used not to distinguish between methylated and unmethylated DNA, but to distinguish between different molecules within these fractions, and thus it can be determined whether reads with the same start and end points are based on the same molecule or different molecules.

[0078] This disclosure provides further methods for analyzing a population of nucleic acids in which at least a portion of the nucleic acids contain one or more modified cytosine residues, e.g., 5-methylcytosine and any of the other modifications described above. These methods involve contacting the population of nucleic acids with an adapter containing one or more cytosine residues modified at the 5C position, e.g., 5-methylcytosine. Preferably, all cytosine residues of such an adapter are also modified, or all such cytosines within the primer-binding region of the adapter are modified. The adapter is attached to both ends of the nucleic acid molecules in the population. Preferably, the adapter contains a sufficient number of different tags such that the probability of two nucleic acids having the same start and end points receiving the same tag combination is low (e.g., 95%, 99%, or 99.9%), depending on the number of tag combinations. The primer-binding sites of such an adapter may be the same or different, but are preferably the same. After the adapter is attached, the nucleic acids are amplified from a primer that binds to the primer-binding site of the adapter. The amplified nucleic acids are split into a first aliquot and a second aliquot. The first aliquot is assayed for sequence data with or without further processing. Therefore, the sequence data for the molecule in the first aliquot is determined independently of the initial methylation state of the nucleic acid molecule. The nucleic acid molecule in the second aliquot is treated with bisulfite. This treatment converts unmodified cytosine to uracil. The bisulfite-treated nucleic acid is then subjected to amplification, primed with a primer targeting the original primer-binding site of the adapter ligated to the nucleic acid. At this point, only the nucleic acid molecule originally ligated to the adapter (separate from its amplified product) is amplified because these nucleic acids retain cytosine at the primer-binding site of the adapter, while the amplified product loses methylation of these cytosine residues that were converted to uracil by bisulfite treatment. Therefore, only the original molecule in the population, which is at least partially methylated, undergoes amplification. After amplification, these nucleic acids are subjected to sequence analysis.By comparing the sequences determined from the first and second aliquots, it may be possible to indicate, in particular, which cytosines within the nucleic acid population were subjected to methylation.

[0079] Distributing the sample into multiple subsamples; sample characteristics; analysis of epigenetic features. In certain embodiments described herein, a population of different forms of nucleic acids (e.g., highly methylated and hypomethylated DNA in a sample, e.g., a set of captured cfDNA described herein) can be physically distributed based on one or more characteristics of the nucleic acids before further analysis, e.g., differential modification or isolation, tagging, and / or sequencing of nucleic acid bases. This technique can be used, for example, to determine whether a particular sequence is highly methylated or hypomethylated. In some embodiments, highly methylated variable epigenetic target regions are analyzed to determine whether they exhibit the highly methylated characteristics of tumor cells, and / or hypomethylated variable epigenetic target regions are analyzed to determine whether they exhibit the hypomethylated characteristics of tumor cells. Furthermore, by distributing heterogeneous nucleic acid populations, rare signals can be increased, for example, by enriching rare nucleic acid molecules that are more dominant in one fraction (or section) of the population. For example, genetic variations that are present in highly methylated DNA but less so (or absent) in less methylated DNA can be more easily detected by distributing the sample into highly methylated and less methylated nucleic acid molecules. By analyzing multiple fractions of the sample, multidimensional analysis of a single locus or nucleic acid species in the genome can be performed, thus achieving higher sensitivity.

[0080] In some cases, heterogeneous nucleic acid samples are distributed into two or more sections (e.g., at least three, four, five, six, or seven sections). In some embodiments, each section is differentially tagged. The tagged sections can then be pooled together for collective sample preparation and / or sequencing. The distribution-tagging-pooling step may be performed more than once, in which case each round of distribution is based on different characteristics (examples are presented herein) and tagged using differential tags that distinguish them from other sections and distribution means.

[0081] Examples of features that can be used for partitioning include sequence length, methylation level, nucleosome binding, sequence mismatch, immunoprecipitation, and / or proteins that bind to DNA. The resulting partitions may include one or more of the following nucleic acid forms: single-stranded DNA (ssDNA), double-stranded DNA (dsDNA), shorter DNA fragments, and longer DNA fragments. In some embodiments, partitioning based on cytosine modification (e.g., cytosine methylation) or methylation is commonly performed and, if necessary, combined with at least one additional partitioning step which may be based on any of the aforementioned features or DNA forms. In some embodiments, a heterogeneous population of nucleic acids is partitioned into nucleic acids having one or more epigenetic modifications and nucleic acids without one or more epigenetic modifications. Examples of epigenetic modifications include the presence or absence of methylation; the level of methylation; the type of methylation (e.g., 5-methylcytosine or other types of methylation, e.g., adenine methylation and / or cytosine hydroxymethylation); and association with one or more proteins, such as histones, and the level of association. Alternatively, or in addition to the above, a heterogeneous population of nucleic acids can be distributed between nucleic acid molecules associated with nucleosomes and nucleic acid molecules lacking nucleosomes. Alternatively, or in addition to the above, a heterogeneous population of nucleic acids can be distributed between single-stranded DNA (ssDNA) and double-stranded DNA (dsDNA). Alternatively, or in addition to the above, a heterogeneous population of nucleic acids can be distributed based on the length of the nucleic acid (e.g., molecules up to 160 bp and molecules with a length greater than 160 bp).

[0082] In some cases, each segment (representing a different nucleic acid morphology) is differentially labeled before sequencing, and the segments are pooled together. In other cases, the different morphologies are sequenced separately. In some embodiments, a population of different nucleic acids is divided into two or more distinct segments. Each segment represents a different nucleic acid morphology, and the first segment (also called a subsample) contains a higher proportion of cytosine-modified DNA than the second subsample. Each segment is tagged separately. The first subsample is subjected to a procedure that affects the first nucleic acid base of the DNA in the first subsample differently from the second nucleic acid base of the DNA, where the first nucleic acid base is modified or unmodified, and the second nucleic acid base is a different modified or unmodified nucleic acid base from the first nucleic acid base, and the first and second nucleic acid bases have the same base-pairing specificity. The tagged nucleic acids are pooled together before sequencing. Sequence reads are obtained and analyzed in silico, including distinguishing between the first and second nucleic acid bases of the DNA in the first partial sample. Tags are used to sort reads from different segmentes. Analysis to detect genetic variants can be performed at the segmental level and at the whole nucleic acid population level. For example, the analysis may include in silico analysis to determine genetic variants in the nucleic acids within each segment, such as CNVs, SNVs, indels, and fusions. In some cases, in silico analysis may include determining chromatin structure. For example, the coverage of sequence reads can be used to determine nucleosome positioning in chromatin. Higher coverage may correlate with higher nucleosome occupancy in genomic regions, while lower coverage may correlate with lower nucleosome occupancy or nucleosome-depleted regions (NDRs).

[0083] The samples may contain nucleic acids with various modifications, including post-replication modifications to nucleotides, and typically non-covalent bonding to one or more proteins.

[0084] In one embodiment, the nucleic acid population is obtained from serum, plasma, or blood samples from subjects suspected of having neoplasia, tumor, or cancer, or who have been previously diagnosed with neoplasia, tumor, or cancer. The nucleic acid population includes nucleic acids with varying levels of methylation. Methylation can result from any one or more post-replication or post-transcriptional modifications. Post-replication modifications include modifications of nucleotide cytosines, particularly those at the 5-position of the nucleic acid base, e.g., 5-methylcytosine, 5-hydroxymethylcytosine, 5-formylcytosine, and 5-carboxylcytosine. The affinity agonist may be an antibody with desired specificity, its natural binding partner or variant (Bock et al., Nat Biotech 28: 1106-1114 (2010); Song et al., Nat Biotech 29: 68-72 (2011)), or an artificial peptide selected, for example, by phage display to have specificity for a given target.

[0085] Examples of capture portions intended herein include methyl-binding domains (MBDs) and methyl-binding proteins (MBPs) described herein, including proteins such as antibodies that preferentially bind to MeCP2 and 5-methylcytosine. Similarly, the distribution of different forms of nucleic acids can be carried out using histone-binding proteins that can separate histone-bound nucleic acids from free or unbound nucleic acids. Examples of histone-binding proteins that can be used in the methods disclosed herein include RBBP4, RbAp48, and SANT domain peptides. With respect to some affinity agonists and modifications, depending on whether the nucleic acid is modified, binding to the agonist may occur in an all-or-nothing manner, while separation may be degree-dependent. In such cases, nucleic acids with a high degree of modification will bind to the agonist to a greater degree than nucleic acids with a low degree of modification. Alternatively, modified nucleic acids may bind in an all-or-nothing manner. Nevertheless, various levels of modification can be sequentially eluted from the binding agonist.

[0086] For example, in some embodiments, partitioning may be binary or based on the degree / level of modification. For instance, all methylated fragments can be partitioned from unmethylated fragments using a methyl-binding domain protein (e.g., MethylMiner Methylated DNA Enrichment Kit (ThermoFisher Scientific)). Subsequent partitioning may involve eluting fragments with different levels of methylation by adjusting the salt concentration in the solution containing the methyl-binding domain and the bound fragments. As the salt concentration increases, fragments with higher levels of methylation are eluted. In some cases, the final partitioning may represent nucleic acids with different degrees of modification (high or low abundance of modification). High and low abundance can be defined by the number of modifications a nucleic acid has compared to the median number of modifications per strand in the population. For example, if the median number of 5-methylcytosine residues in nucleic acids in a sample is 2, then nucleic acids containing more than 2 5-methylcytosine residues have a high abundance of this modification, while nucleic acids with 1 or zero 5-methylcytosine residues have a low abundance. The effect of affinity separation is to enrich the bound phase with nucleic acids that have a high degree of modification, and enrich the unbound phase (i.e., in solution) with nucleic acids that have a low degree of modification. After eluting the nucleic acids in the bound phase, further processing can be performed.

[0087] When using the MethylMiner Methylated DNA Enrichment Kit (ThermoFisher Scientific), various levels of methylation can be partitioned using sequential elution. For example, a low-methylation fraction (e.g., no methylation) can be separated from a methylated fraction by contacting the nucleic acid population with MBD from the kit, which has been attached to magnetic beads. The beads are used to separate methylated nucleic acids from unmethylated nucleic acids. Then, one or more elution steps are performed sequentially to elute nucleic acids with different levels of methylation. For example, a first set of methylated nucleic acids can be eluted at a salt concentration of 160 mM or higher, e.g., at least 150 mM, at least 200 mM, at least 300 mM, at least 400 mM, at least 500 mM, at least 600 mM, at least 700 mM, at least 800 mM, at least 900 mM, at least 1000 mM, or at least 2000 mM. After eluting such methylated nucleic acids, magnetic separation is used again to separate the nucleic acids with higher levels of methylation from those with lower levels of methylation. The elution and magnetic separation steps themselves can be repeated to create various segments, such as a low-methylation segment (representing no methylation), a methylated segment (representing low levels of methylation), and a high-methylation segment (representing high levels of methylation).

[0088] In some methods, nucleic acids bound to the activator used for affinity separation are subjected to a washing step. The washing step washes away nucleic acids that are weakly bound to the affinity activator. Such nucleic acids may be enriched with nucleic acids having a degree of modification close to the mean or median (i.e., an intermediate value between nucleic acids that remained bound to the solid phase and nucleic acids that did not bind to the solid phase when the sample was first brought into contact with the activator). Affinity separation yields at least two, and sometimes three or more, partitions of nucleic acids with different degrees of modification. Although the partitions remain separated, nucleic acids from at least one, and usually two or three (or more), partitions are ligated to nucleic acid tags, usually provided as components of an adapter, so that nucleic acids in different partitions receive different tags that distinguish members of one partition from members of another. Tags ligated to nucleic acid molecules of the same partition may be the same or different. However, if they are different, the tags may share a common part of their code to identify the molecule to which they are attached as belonging to a particular partition. For further details regarding the portioning of nucleic acid samples based on characteristics such as methylation, see WO2018 / 119452, which is incorporated herein by reference. In some embodiments, nucleic acid molecules can be fractionated into different parts based on nucleic acid molecules bound to a particular protein or fragment thereof and nucleic acid molecules not bound to that particular protein or fragment thereof.

[0089] Nucleic acid molecules can be fractionated based on DNA-protein binding. Protein-DNA complexes can be fractionated based on specific properties of the protein. Examples of such properties include various epitopes, modifications (e.g., histone methylation or acetylation), or enzymatic activity. Examples of proteins that can bind to DNA and serve as a basis for fractionation include, but are not limited to, protein A and protein G. Nucleic acid molecules can be fractionated based on the region bound to the protein using any suitable method. Examples of methods used to fractionate nucleic acid molecules based on the region bound to the protein include, but are not limited to, SDS-PAGE, chromatin immunoprecipitation (ChIP), heparin chromatography, and asymmetric field flow fractionation (AF4).

[0090] In some embodiments, nucleic acid distribution is carried out by contacting the nucleic acid with the methylation-binding domain ("MBD") of a methylation-binding protein ("MBP"). The MBD binds to 5-methylcytosine (5mC). The MBD is coupled to paramagnetic beads such as Dynabeads® M-280 streptavidin via a biotin linker. Distribution to fractions with different degrees of methylation can be carried out by eluting the fractions by increasing the NaCl concentration.

[0091] An example method for identifying molecular tags in a library distributed with MBD beads using NGS is as follows:

[0092] Physical distribution of extracted DNA samples (e.g., plasma DNA extracted from human samples) using a methyl-binding domain protein-bead purification kit. All eluates from the process are stored for downstream processing.

[0093] Parallel application of differential molecular tags and adapter sequences enabling NGS for each segment. For example, ligating a highly methylated segment, a residually methylated ("washed") segment, and a lowly methylated segment with an NGS adapter having a molecular tag.

[0094] All molecularly tagged segments are recombined and then amplified using adapter-specific DNA primer sequences.

[0095] Enrichment / hybridization of the combined and amplified total library. Targeting the desired genomic region (e.g., cancer-specific genetic variants and differential methylation regions).

[0096] Re-amplification of the enriched total DNA library, and addition of sample tags. Different samples are pooled and assayed in multiplex in an NGS instrument.

[0097] Bioinformatics analysis of NGS data. Molecular tags are used to identify unique molecules and to deconvolve samples into differentially MBD-distributed molecules. This analysis allows for relative information about 5-methylcytosine in genomic regions, in parallel with standard gene sequencing / variant detection.

[0098] Examples of MBPs intended in this specification include, but are not limited to, the following:

[0099] (a) MeCP2 is a protein that preferentially binds to 5-methylcytosine rather than unmodified cytosine.

[0100] (b) RPL26, PRP8, and the DNA mismatch repair protein MHS6 preferentially bind to 5-hydroxymethylcytosine rather than unmodified cytosine.

[0101] (c)FOXK1, FOXK2, FOXP1, FOXP4, and FOXI3 preferentially bind to 5-formyl-cytosine rather than unmodified cytosine (Iurlaro et al., Genome Biol. 14: R119 (2013)).

[0102] (d) An antibody specific to one or more methylated nucleotide bases.

[0103] Generally, elution is dependent on the number of methylation sites per molecule, with higher salt concentrations eluting molecules that have more methylation. A series of elution buffers with increasing NaCl concentrations can be used to elute DNA into separate populations based on the degree of methylation. Salt concentrations can range from approximately 100 nM to approximately 2500 mM NaCl. In one embodiment, the process yields three divisions. Molecules are brought into contact with a solution of a first salt concentration containing molecules that have methyl-binding domains and can be attached to a capture moiety such as streptavidin. At the first salt concentration, some populations of molecules bind to the MBD, while others remain unbound. The unbound populations can be separated as a "low-methylated" population. For example, the first division, representing the low-methylated form of DNA, remains unbound at low salt concentrations, e.g., 100 mM or 160 mM. The second segment, representing intermediate methylated DNA, is eluted using an intermediate salt concentration, e.g., between 100 mM and 2000 mM. This segment is also separated from the sample. The third segment, representing the highly methylated form of DNA, is eluted using a high salt concentration, e.g., at least about 2000 mM.

[0104] This disclosure provides further methods for analyzing a population of nucleic acids in which at least a portion of the nucleic acids contains one or more modified cytosine residues, e.g., 5-methylcytosine and any of the other modifications described above. In these methods, after distribution, a portion of the nucleic acid sample is brought into contact with an adapter containing one or more cytosine residues modified at the 5C position, e.g., 5-methylcytosine. Preferably, all cytosine residues of such an adapter are also modified, or all such cytosines within the primer-binding region of the adapter are modified. The adapter is attached to both ends of the nucleic acid molecule in the population. Preferably, the adapter contains a sufficient number of different tags such that the probability of two nucleic acids having the same start and end points receiving the same tag combination is low (e.g., 95%, 99%, or 99.9%), depending on the number of tag combinations. The primer-binding sites of such an adapter may be the same or different, but are preferably the same. After the adapter is attached, the nucleic acid is amplified from a primer that binds to the primer-binding site of the adapter. The amplified nucleic acid is split into a first aliquot and a second aliquot. The first aliquot is assayed for sequence data with or without further processing. Thus, the sequence data for the molecule in the first aliquot is determined independently of the initial methylation state of the nucleic acid molecule. The nucleic acid molecule in the second aliquot is subjected to a procedure that affects the first nucleic acid base of the DNA differently than the second nucleic acid base of the DNA, where the first nucleic acid base contains cytosine modified at position 5 and the second nucleic acid base contains unmodified cytosine. This procedure may be bisulfite treatment or another procedure that converts unmodified cytosine to uracil. The nucleic acid subjected to the procedure is then amplified using a primer to the original primer-binding site of the adapter ligated to the nucleic acid. At this point, only the nucleic acid molecule originally ligated to the adapter (which is separate from its amplified product) is amplified because these nucleic acids retain cytosine at the primer-binding site of the adapter, while in the amplified product, the methylation of these cytosine residues that were converted to uracil by bisulfite treatment is lost.Therefore, only the original molecules in the population that are at least partially methylated undergo amplification. After amplification, these nucleic acids are subjected to sequence analysis. By comparing the sequences determined from the first and second aliquots, it may be possible to indicate, in particular, which cytosines in the nucleic acid population were subjected to methylation.

[0105] Such analysis can be performed using the following exemplary procedure: After distribution, both ends of the methylated DNA are ligated to a Y-shaped adapter containing a primer-binding site and a tag. The cytosine on the adapter is modified at position 5 (e.g., 5-methylation). The modification of the adapter serves to protect the primer-binding site during subsequent conversion steps (e.g., bisulfite treatment, TAP conversion, or any other conversion that does not affect the modified cytosine but does affect the unmodified cytosine). After the adapter is attached, the DNA molecule is amplified. The amplified product is split into two aliquots for sequencing with and without conversion. The aliquot not subjected to conversion can be subjected to sequencing analysis with or without further processing. The other aliquot is subjected to a procedure that affects the first nucleic acid base of the DNA differently than the second nucleic acid base of the DNA, where the first nucleic acid base contains cytosine modified at position 5 and the second nucleic acid base contains unmodified cytosine. This procedure may be bisulfite treatment or another procedure to convert unmodified cytosine to uracil. When contacted with a primer specific to the original primer-binding site, only the primer-binding site protected by cytosine modification can support amplification. Therefore, only the original molecule is subjected to further amplification, and the copy from the first amplification is not subjected to further amplification. The further amplified molecule is then subjected to sequence analysis. The sequences from the two aliquots can then be compared. Similar to the separation scheme described above, the nucleic acid tag on the adapter is used to distinguish nucleic acid molecules within the same segment, rather than to distinguish between methylated and unmethylated DNA.

[0106] A step of subjecting a first partial sample to a procedure that affects the first nucleic acid base of the DNA of the first partial sample in a way that is different from that of the second nucleic acid base of the DNA. A method disclosed herein is a step of subjecting a first partial sample to a procedure that affects a first nucleic acid base of the DNA of the first partial sample differently from a second nucleic acid base of the DNA, wherein the first nucleic acid base is a modified or unmodified nucleic acid base, the second nucleic acid base is a modified or unmodified nucleic acid base different from the first nucleic acid base, and the first and second nucleic acid bases have the same base-pairing specificity. In some embodiments, if the first nucleic acid base is a modified or unmodified adenine, then the second nucleic acid base is a modified or unmodified adenine; if the first nucleic acid base is a modified or unmodified cytosine, then the second nucleic acid base is a modified or unmodified cytosine; if the first nucleic acid base is a modified or unmodified guanine, then the second nucleic acid base is a modified or unmodified guanine; and if the first nucleic acid base is a modified or unmodified thymine, then the second nucleic acid base is a modified or unmodified thymine (for the purposes of this step, modified and unmodified uracil are included in modified thymine).

[0107] In some embodiments, the first nucleic acid base is modified or unmodified cytosine, and the second nucleic acid base is modified or unmodified cytosine. For example, the first nucleic acid base may contain unmodified cytosine (C), and the second nucleic acid base may contain one or more of 5-methylcytosine (mC) and 5-hydroxymethylcytosine (hmC). Alternatively, the second nucleic acid base may contain C, and the first nucleic acid base may contain one or more of mC and hmC. Other combinations are also possible, for example, as shown in the above summary and the following discussion, such as when one of the first and second nucleic acid bases contains mC and the other contains hmC.

[0108] In some embodiments, a procedure that affects the first nucleic acid base of the DNA in a first partial sample differently from that affecting the second nucleic acid base of the DNA involves bisulfite conversion. Treatment with bisulfite converts unmodified cytosine and certain modified cytosine nucleotides (e.g., 5-formylcytosine (fC) or 5-carboxylcytosine (caC)) to uracil, while other modified cytosines (e.g., 5-methylcytosine, 5-hydroxymethylcytosine) are not converted. Therefore, when using bisulfite conversion, the first nucleic acid base includes one or more of unmodified cytosine, 5-formylcytosine, 5-carboxylcytosine, or other cytosine forms affected by bisulfite, and the second nucleic acid base may include one or more of mC and hmC, e.g., mC and optionally hmC. Sequencing of the bisulfite-treated DNA identifies the positions read as cytosine as mC or hmC positions. On the other hand, positions read as T are identified as T or bisulfite-sensitive forms of C, such as unmodified cytosine, 5-formylcytosine, or 5-carboxylcytosine. Therefore, by performing bisulfite conversion on the first partial sample described herein, it becomes easier to identify mC or hmC-containing positions using sequence reads obtained from the first partial sample. For an illustrative description of bisulfite conversion, see, for example, Moss et al., Nat Commun. 2018; 9: 5068.

[0109] In some embodiments, a procedure that affects the first nucleic acid base of the DNA in the first sample differently from the second nucleic acid base of the DNA includes oxidative bisulfite (Ox-BS) conversion. In some embodiments, a procedure that affects the first nucleic acid base of the DNA in the first sample differently from the second nucleic acid base of the DNA includes Tet-assisted bisulfite (TAB) conversion. In some embodiments, a procedure that affects the first nucleic acid base of the DNA in the first sample differently from the second nucleic acid base of the DNA includes Tet-assisted conversion using a substituted borane reducing agent, which may optionally be 2-picoline borane, borampyridine, tert-butylamine borane, or ammonia borane. In some embodiments, a procedure that affects the first nucleic acid base of the DNA in the first sample differently from the second nucleic acid base of the DNA includes chemical-assisted conversion using a substituted borane reducing agent, which may optionally be 2-picoline borane, borampyridine, tert-butylamine borane, or ammonia borane. In some embodiments, the procedure that affects the first nucleic acid base of the DNA in the first partial sample differently from the second nucleic acid base of the DNA includes APOBEC coupling epigenetic (ACE) conversion.

[0110] In some embodiments, the procedure for affecting the first nucleic acid base of the DNA in a first partial sample differently from the second nucleic acid base of the DNA includes, for example, enzymatic conversion of the first nucleic acid base, similar to EM-Seq. See, for example, Vaisvila R, et al. (2019) EM-seq: Detection of DNA methylation at single base resolution from picograms of DNA. bioRxiv; DOI: 10.1101 / 2019.12.20.884692v1, available at www.biorxiv.org / content / 10.1101 / 2019.12.20.884692v1. For example, 5mC and 5hmC can be converted to substrates that cannot be deaminated by deaminase (e.g., APOBEC3A) using TET2 and T4-βGT, and then the unmodified cytosines can be deaminated using deaminase (e.g., APOBEC3A) to convert them to uracil.

[0111] In some embodiments, the procedure for affecting the first nucleic acid base of the DNA of a first partial sample differently from that affecting the second nucleic acid base of the DNA includes separating the DNA that originally contains the first nucleic acid base from the DNA that originally does not contain the first nucleic acid base.

[0112] In some embodiments, the first nucleic acid base is a modified or unmodified adenine, and the second nucleic acid base is a modified or unmodified adenine. In some embodiments, the modified adenine is N6-methyladenine (mA). In some embodiments, the modified adenine is one or more of N6-methyladenine (mA), N6-hydroxymethyladenine (hmA), or N6-formyladenine (fA).

[0113] Techniques including methylated DNA immunoprecipitation (MeDIP) can be used to isolate DNA containing modified bases such as mA from other DNA. See, for example, Kumar et al., Frontiers Genet. 2018; 9: 640; Greer et al., Cell 2015; 161: 868-878. Antibodies specific to mA are described in Sun et al., Bioessays 2015; 37: 1155-62. Antibodies against various modified nucleic acid bases, such as thymine / uracil forms including halogenated forms like 5-bromouracil, are commercially available. Various modified bases can also be detected based on changes in their base pairing specificity. For example, hypoxanthine is a modified form of adenine that can result from deamination and is read as G in sequencing. For example, see U.S. Patent 8,486,630; Brown, Genomes, 2nd Ed., John Wiley & Sons, Inc., New York, NY, 2002, chapter 14, "Mutation, Repair, and Recombination."

[0114] Enrichment / capture steps, amplification, adapter, barcode In some embodiments, the methods disclosed herein include the step of capturing one or more sets of target regions of DNA, such as cfDNA. Capture can be carried out using any suitable method known in the art. In some embodiments, the capture step includes contacting the DNA to be captured with a set of target-specific probes. The set of target-specific probes may have any of the features described herein with respect to a set of target-specific probes, including, but not limited to, those in the embodiments described above and in the section on probes below. The capture step can be carried out on one or more partial samples prepared during the methods disclosed herein. In some embodiments, DNA is captured from at least a first or second partial sample, for example, at least a first and a second partial sample. If the first partial sample undergoes a separation step (for example, separating DNA that originally contains a first nucleic acid base (e.g., hmC) from DNA that originally does not contain a first nucleic acid base (e.g., hmC-seal)), the capture step can be performed on any, any two, or all of the DNA that originally contains a first nucleic acid base (e.g., hmC), the DNA that originally does not contain a first nucleic acid base, and the second partial sample. In some embodiments, the partial samples are differentially tagged (e.g., as described herein), then pooled, and then captured.

[0115] The capture step can be carried out using conditions suitable for a particular nucleic acid hybridization, and these conditions generally depend to some extent on the characteristics of the probe, such as its length and base composition. Those skilled in the art will be able to determine the appropriate conditions based on general knowledge in the art regarding nucleic acid hybridization. In some embodiments, a complex is formed between the target-specific probe and DNA.

[0116] In some embodiments, the methods described herein include a step of capturing cfDNA obtained from a test subject for multiple sets of target regions. The target regions include epigenetic target regions, which may exhibit differences in methylation levels and / or fragmentation patterns depending on whether they originate from tumors or healthy cells. The target regions also include sequence-variable target regions, which may exhibit differences in sequence depending on whether they originate from tumors or healthy cells. The capture step generates a set of captured cfDNA molecules, where cfDNA molecules corresponding to the sequence-variable target region set are captured in a higher capture yield than cfDNA molecules corresponding to the epigenetic target region set in the set of captured cfDNA molecules. For further consideration of the capture step, capture yields, and related embodiments, see WO2020 / 160414, incorporated herein by reference for all purposes.

[0117] In some embodiments, the method described herein includes the step of contacting cfDNA obtained from a test subject with a set of target-specific probes, wherein the set of target-specific probes is configured to capture cfDNA corresponding to a sequence-variable target region set with a higher capture yield than cfDNA corresponding to an epigenetic target region set.

[0118] To analyze sequence-variable target regions with sufficient confidence or accuracy, higher sequencing depths may be required than those necessary for analyzing epigenetic target regions. Therefore, capturing cfDNA corresponding to a set of sequence-variable target regions at a higher capture yield than cfDNA corresponding to a set of epigenetic target regions may be beneficial. The amount of data required to determine fragmentation patterns (e.g., for examining perturbations at transcription start sites or CTCF-binding sites) or fragment abundances (e.g., in highly methylated and hypomethylated segments) is generally less than the amount of data required to determine the presence or absence of cancer-associated sequence mutations. By capturing target region sets at different yields, it may be easier to sequence target regions to different sequencing depths in the same sequencing run (e.g., using a pooled mixture and / or in the same sequencing cell).

[0119] In various embodiments, the method further includes the step of sequencing the captured cfDNA to varying degrees of sequencing depth with respect to, for example, an epigenetic target region set and a sequence variable target region set, consistent with the discussion herein. In some embodiments, the target-specific probe-DNA complex is separated from DNA not bound to the target-specific probe. For example, if the target-specific probe is covalently or noncovalently bound to a solid support, washing or aspiration steps can be used to separate the unbound material. Alternatively, if the complex has chromatographic properties distinct from the unbound material (for example, if the probe contains a ligand that binds to a chromatography resin), chromatography can be used.

[0120] As discussed in detail elsewhere in this specification, a set of target-specific probes may include multiple sets, such as probes for sequence-variable target regions and probes for epigenetic target regions. In some such embodiments, the capture step is performed simultaneously in the same container using probes for sequence-variable target regions and probes for epigenetic target regions, for example, the probes for sequence-variable target regions and probes for epigenetic target regions are in the same composition. This method provides a relatively streamlined workflow. In some embodiments, the concentration of probes for sequence-variable target regions is higher than the concentration of probes for epigenetic target regions.

[0121] Alternatively, the capture step may be performed using the sequence-variable target region probe set in a first container and the epigenetic target region probe set in a second container, or the contact step may be performed using the sequence-variable target region probe set at a first time point and in the first container, and the epigenetic target region probe set at a second time point before or after the first time point. This method allows for the preparation of separate first and second compositions, each containing captured DNA corresponding to the sequence-variable target region set and captured DNA corresponding to the epigenetic target region set. The compositions can be processed separately as desired (e.g., to fractionate based on methylation as described elsewhere in this specification) and then recombined in proportions suitable for further processing and analysis, such as sequencing.

[0122] In some embodiments, DNA is amplified. In some embodiments, amplification is performed before the capture step. In some embodiments, amplification is performed after the capture step.

[0123] In some embodiments, the adapter is included in the DNA. This can be done in parallel with the amplification procedure, for example, by attaching the adapter to the 5' portion of the primer as described above. Alternatively, the adapter can be added by other methods such as ligation.

[0124] In some embodiments, the DNA is accompanied by a tag, which may include a barcode. The tag can facilitate the identification of the nucleic acid's origin. For example, a barcode can be used to identify the origin (e.g., target) from which the DNA originates after pooling multiple samples for parallel sequencing. This can be done in parallel with the amplification procedure, for example, by affixing a barcode to the 5' portion of the primer, as described above. In some embodiments, the adapter and tag / barcode are provided by the same primer or primer set. For example, the barcode may be located on the 3' side of the adapter and on the 5' side of the portion that hybridizes with the primer's target. Alternatively, the barcode may be added together with the adapter in the same ligation substrate as needed, by other methods such as ligation.

[0125] Further details regarding amplification, tagging, and barcodes are discussed in the following section, “General Features of the Method,” and these can be combined to an operational degree with any of the embodiments described above, as well as those described in the Introduction and Summary sections.

[0126] Captured set In some embodiments, a set of captured DNA (e.g., cfDNA) is provided. With respect to the disclosed method, the set of captured DNA can be obtained, for example, by performing a capture step after a distribution step, as described herein. The captured set may include DNA corresponding to a set of sequence variable target regions, a set of epigenetic target regions, or a combination thereof. In some embodiments, when normalized for differences in the size (footprint size) of the targeting regions, the quantity of captured sequence variable target region DNA is greater than the quantity of captured epigenetic target region DNA.

[0127] Alternatively, this can result in a first captured set and a second captured set, each containing DNA corresponding to a sequence-variable target region set and DNA corresponding to an epigenetic target region set, respectively. The first and second captured sets can be combined to result in a combined captured set.

[0128] In some embodiments, including the combined captured sets discussed above, the captured sets containing DNA corresponding to the sequence variable target region set and the epigenetic target region set are present at higher concentrations than the DNA corresponding to the epigenetic target region set, e.g., 1.1 to 1.2 times, 1.2 to 1.4 times, or 1.4 to 1.6 times. Concentration of 0, 1.6 to 1.8 times, 1.8 to 2.0 times, 2.0 to 2.2 times, 2.2 to 2.4 times, 2.4 to 2.6 times, 2.6 to 2.8 times, 2.8 to 3.0 times, 3.0 to 3.5 times, 3.5 to 4.0 times, 4.0 to 4.5 times, 4.5 to 5.0 times, 5.0 to 5.5 times, 5.5 to 6.0 times, 6.0 to 6.5 2x concentration, 6.5x to 7.0x concentration, 7.0x to 7.5x concentration, 7.5x to 8.0x concentration, 8.0x to 8.5x concentration, 8.5x to 9.0x concentration, 9.0x to 9.5x concentration, 9.5x to 10.0x concentration, 10x to 11x concentration, 11x to 12x concentration, 12x to 13x concentration, 13x to 14x concentration, 14x to 15x concentration, 15x to 16x concentration, 16x to 17x concentration, 17x to 18x The concentrations can be twice as high, 18 to 19 times higher, 19 to 20 times higher, 20 to 30 times higher, 30 to 40 times higher, 40 to 50 times higher, 50 to 60 times higher, 60 to 70 times higher, 70 to 80 times higher, 80 to 90 times higher, 90 to 100 times higher, 10 to 20 times higher, 10 to 40 times higher, 10 to 50 times higher, 10 to 70 times higher, or 10 to 100 times higher. The degree of concentration difference is the main cause of normalization with respect to the footprint size of the target region, as discussed in the definition section.

[0129] Epigenetic target region set An epigenetic target region set may include one or more types of target regions that can discriminate DNA from neoplastic (e.g., tumor or cancer) cells from DNA from healthy cells, e.g., non-neoplastic circulating cells. Exemplary types of such regions are discussed in detail herein. An epigenetic target region set may also include, for example, one or more control regions described herein. In some embodiments, an epigenetic target region set has a footprint of at least 100 kb, e.g., at least 200 kb, at least 300 kb, or at least 400 kb. In some embodiments, the epigenetic target region set has footprints ranging from 100 to 1000 kb, for example, 100 to 200 kb, 200 to 300 kb, 300 to 400 kb, 400 to 500 kb, 500 to 600 kb, 600 to 700 kb, 700 to 800 kb, 800 to 900 kb, and 900 to 1,000 kb.

[0130] Highly methylated variable target region In some embodiments, the epigenetic target region set includes one or more hypermethylated variable target regions. Generally, a hypermethylated variable target region refers to a region, for example in a cfDNA sample, where the observed elevated level of methylation indicates an increased likelihood that the sample (e.g., a cfDNA sample) contains DNA produced by neoplastic cells such as tumor or cancer cells. For example, hypermethylation of tumor suppressor gene promoters has been repeatedly observed. See, for example, Kang et al., Genome Biol. 18:53 (2017) and the references cited therein. In one example, a hypermethylated variable target region may include a region in cancerous tissue that does not necessarily have different methylation compared to DNA from the same type of healthy tissue, but has different methylation (e.g., more methylation) compared to cfDNA that is typical in healthy subjects. For example, if the presence of cancer leads to increased cell death, e.g., apoptosis, of cells of the tissue type corresponding to cancer, such cancer can be detected, at least in part, using such hypermethylated variable target regions. In some embodiments, the hypermethylated variable target regions include one or more genomic regions in which cfDNA molecules in those regions have no different methylation status in cancer subjects compared to cfDNA from healthy subjects, but the increased presence / quantity of hypermethylated cfDNA in those regions serves as an indicator of a specific tissue type (e.g., cancer origin) and is represented as cfDNA with increased apoptosis (e.g., tumor efflux) into circulation.

[0131] Highly methylated target regions can be obtained, for example, from the Cancer Genome Atlas. Kang et al., Genome Biology 18:53 (2017) describe the construction of a stochastic method called CancerLocator using highly methylated target regions derived from breast, colon, kidney, liver, and lung. In some embodiments, highly methylated target regions may be specific to one or more types of cancer. Thus, in some embodiments, the highly methylated target regions include one, two, three, four, or five subsets of highly methylated target regions that collectively represent highly methylated in one, two, three, four, or five of the following: breast cancer, colon cancer, kidney cancer, liver cancer, and lung cancer.

[0132] In some embodiments, the probes for the epigenetic target region set include probes specific to one or more highly methylated variable target regions. The highly methylated variable target regions may be any of the above. For example, in some embodiments, the probes specific to highly methylated variable target regions include probes specific to a plurality of loci listed in Table 1, e.g., probes specific to at least 10%, 20%, 30%, 40%, 50%, 60%, 70%, 80%, 90%, or 100% of the loci listed in Table 1. In some embodiments, the probes specific to highly methylated variable target regions include probes specific to a plurality of loci listed in Table 2, e.g., probes specific to at least 10%, 20%, 30%, 40%, 50%, 60%, 70%, 80%, 90%, or 100% of the loci listed in Table 2. In some embodiments, probes specific to highly methylated variable target regions include probes specific to a plurality of loci listed in Table 1 or Table 2, for example, probes specific to at least 10%, 20%, 30%, 40%, 50%, 60%, 70%, 80%, 90%, or 100% of the loci listed in Table 1 or Table 2. In some embodiments, for each locus included as a target region, there may be one or more probes having a hybridization site that binds between the transcription start site and the stop codon (the last stop codon of a gene that undergoes alternative splicing). In some embodiments, one or more probes bind within 300 bp, for example, within 200 bp or 100 bp, of the listed locations. In some embodiments, the probes have hybridization sites that overlap with the locations listed above. In some embodiments, the hypermethylation target region-specific probes include probes specific to one, two, three, four, or five subsets of hypermethylation target regions that collectively represent hypermethylation in one, two, three, four, or five of the following cancers: breast cancer, colon cancer, kidney cancer, liver cancer, and lung cancer.

[0133] Low-methylation variable target region Overall hypomethylation is a phenomenon commonly observed in various cancers. See, for example, Hon et al., Genome Res. 22:246-258 (2012) (breast cancer); Ehrlich, Epigenomics 1:239-259 (2009) (a review article mentioning the observation of hypomethylation in colon cancer, ovarian cancer, prostate cancer, leukemia, hepatocellular carcinoma, and cervical cancer). For example, regions such as repeat elements, e.g., LINE1 elements, Alu elements, centromere tandem repeats, periconomere tandem repeats, and satellite DNA, as well as intergenetic regions that are normally methylated in healthy cells, may show reduced methylation in tumor cells. Therefore, in some embodiments, a set of epigenetic target regions includes hypomethylated variable target regions in which the observed decrease in methylation levels indicates an increased likelihood that the sample (e.g., a cfDNA sample) contains DNA produced by neoplastic cells such as tumor or cancer cells. For example, a hypomethylated variable target region may include regions in cancerous tissue that are not necessarily methylated differently from DNA from healthy tissue of the same type, but are methylated differently (e.g., less methylated) than cfDNA that is typical in healthy subjects. For example, if the presence of cancer leads to increased cell death, e.g., apoptosis, of cells of the tissue type corresponding to cancer, such cancer can be detected, at least in part, using such a hypomethylated variable target region. In some embodiments, a hypomethylated variable target region includes one or more genomic regions in which cfDNA molecules in those regions are not methylated differently from cfDNA from healthy subjects in cancer subjects, but the increased presence / quantity of hypomethylated cfDNA in those regions serves as an indicator of a particular tissue type (e.g., cancer origin) and is indicated as cfDNA with increased apoptosis (e.g., tumor efflux) into circulation.

[0134] In some embodiments, the low-methylation variable target region includes repeat elements and / or intergenetic regions. In some embodiments, the repeat elements include one, two, three, four, or five of the following: LINE1 elements, Alu elements, centromere tandem repeats, peristromere tandem repeats, and / or satellite DNA.

[0135] Exemplary specific genomic regions exhibiting cancer-associated hypomethylation include nucleotides 8403565–8953708 and 151104701–151106035 on human chromosome 1. In some embodiments, the hypomethylation variable target region overlaps with or includes one or both of these regions.

[0136] In some embodiments, a probe for a set of epigenetic target regions includes a probe specific to one or more hypomethylated variable target regions. The hypomethylated variable target regions may be any of the above. For example, a probe specific to one or more hypomethylated variable target regions may include probes for repeating elements, such as LINE1 elements, Alu elements, centromere tandem repeats, peristromereal tandem repeats, and satellite DNA, where intergenetic regions that are normally methylated in healthy cells may show reduced methylation in tumor cells.

[0137] In some embodiments, probes specific to low-methylation variable target regions include probes specific to repeat elements and / or intergenetic regions. In some embodiments, probes specific to repeat elements include probes specific to one, two, three, four, or five of the following: LINE1 elements, Alu elements, centromere tandem repeats, pericentromere tandem repeats, and / or satellite DNA.

[0138] Exemplary probes specific to genomic regions exhibiting cancer-related hypomethylation include probes specific to human chromosome 1 nucleotides 8403565-8953708 and / or 151104701-151106035. In some embodiments, probes specific to hypomethylation variable target regions include probes specific to regions that overlap with or contain human chromosome nucleotides 8403565-8953708 and / or 151104701-151106035.

[0139] Probes for detecting a panel of regions include probes for detecting target genomic regions (hotspot regions), as well as nucleosome recognition probes (e.g., KRAS codons 12 and 13), which can be designed to optimize capture based on analysis of cfDNA coverage and fragment size variations influenced by nucleosome binding patterns and GC sequence composition. Regions used herein may also include non-hotspot regions optimized based on nucleosome location and GC model.

[0140] subject In some embodiments, DNA (e.g., cfDNA) is obtained from subjects having cancer. In some embodiments, DNA (e.g., cfDNA) is obtained from subjects suspected of having cancer. In some embodiments, DNA (e.g., cfDNA) is obtained from subjects having a tumor. In some embodiments, DNA (e.g., cfDNA) is obtained from subjects suspected of having a tumor. In some embodiments, DNA (e.g., cfDNA) is obtained from subjects having a neoplasia. In some embodiments, DNA (e.g., cfDNA) is obtained from subjects in remission of tumor, cancer, or neoplasia (e.g., after chemotherapy, surgical resection, radiation therapy, or a combination thereof). In any of the embodiments described above, the cancer, tumor, or neoplasia or suspected cancer, tumor, or neoplasia may be of the lung, colon, rectum, kidney, breast, prostate, or liver. In some embodiments, the cancer, tumor, or neoplasia or suspected cancer, tumor, or neoplasia may be of the lung. In some embodiments, the cancer, tumor, neoplasia, or suspected cancer, tumor, or neoplasia is of the colon or rectum. In some embodiments, the cancer, tumor, neoplasia, or suspected cancer, tumor, or neoplasia is of the breast. In some embodiments, the cancer, tumor, neoplasia, or suspected cancer, tumor, or neoplasia is of the prostate. In any of the embodiments described above, the subject may be a human subject.

[0141] In some embodiments, the sequence-variable target region probe set has a footprint of at least 0.5kb, for example, at least 1kb, at least 2kb, at least 5kb, at least 10kb, at least 20kb, at least 30kb, or at least 40kb. In some embodiments, the epigenetic target region probe set has a footprint ranging from 0.5 to 100kb, for example, in the range of 0.5 to 2kb, 2 to 10kb, 10 to 20kb, 20 to 30kb, 30 to 40kb, 40 to 50kb, 50 to 60kb, 60 to 70kb, 70 to 80kb, 80 to 90kb, and 90 to 100kb.

[0142] In some embodiments, probes specific to a sequence-variable target region set include probes specific to target regions derived from at least 10, 20, 30, or 35 cancer-related genes, such as AKT1, ALK, BRAF, CCND1, CDK2A, CTNNB1, EGFR, ERBB2, ESR1, FGFR1, FGFR2, FGFR3, FOXL2, GATA3, GNA11, GNAQ, GNAS, HRAS, IDH1, IDH2, KIT, KRAS, MED12, MET, MYC, NFE2L2, NRAS, PDGFRA, PIK3CA, PPP2R1A, PTEN, RET, STK11, TP53, and U2AF1.

[0143] composition containing captured DNA A combination comprising a first population and a second population of captured DNA is provided herein. The first population may contain, or may be derived from, DNA having a higher proportion of cytosine modifications than the second population. The first population may include a first nucleic acid base form in which the base-pairing specificity originally present in the DNA has been altered, and a second nucleic acid base in which the base-pairing specificity has not been altered, where the first nucleic acid base form before alteration of base-pairing specificity originally present in the DNA is a modified or unmodified nucleic acid base, and the second nucleic acid base is a modified or unmodified nucleic acid base different from the first nucleic acid base, and the first nucleic acid base form before alteration of base-pairing specificity originally present in the DNA and the second nucleic acid base have the same base-pairing specificity. The second population does not include a first nucleic acid base form in which the base-pairing specificity originally present in the DNA has been altered. In some embodiments, the cytosine modification is cytosine methylation. In some embodiments, the first nucleic acid base is a modified or unmodified cytosine, and the second nucleic acid base is a modified or unmodified cytosine. The first and second nucleic acid bases may be any of those discussed herein, in summary, or in relation to the step of subjecting a first partial sample to a procedure that affects the first nucleic acid base of the DNA of the first partial sample differently from the second nucleic acid base of the DNA.

[0144] In some embodiments, the first group includes array tags selected from a first set of one or more array tags, and the second group includes array tags selected from a second set of one or more array tags, wherein the second set of array tags is different from the first set of array tags. The array tags may include barcodes.

[0145] In some embodiments, the first population includes protected hmC, e.g., glucosylated hmC. In some embodiments, the first population has been subjected to one of the conversion procedures discussed herein, e.g., bisulfite conversion, Ox-BS conversion, TAB conversion, ACE conversion, TAP conversion, TAPSβ conversion, or CAP conversion. In some embodiments, the first population has been subjected to protection of hmC, followed by deamination of mC and / or C. In some embodiments of the combination, the first population includes or is derived from DNA having a higher proportion of cytosine modifications than the second population, and the first population includes a first subpopulation and a second subpopulation, where the first nucleic acid bases are modified or unmodified nucleic acid bases, the second nucleic acid bases are modified or unmodified nucleic acid bases different from the first nucleic acid bases, and the first and second nucleic acid bases have the same base pairing specificity. In some embodiments, the second population does not include the first nucleic acid bases. In some embodiments, the first nucleic acid base is modified or unmodified cytosine, and the second nucleic acid base is modified or unmodified cytosine, and the modified cytosine is optionally mC or hmC. In some embodiments, the first nucleic acid base is modified or unmodified adenine, and the second nucleic acid base is modified or unmodified adenine, and the modified adenine is optionally mA.

[0146] In some embodiments, the first nucleic acid base (e.g., modified cytosine) is biotinylated. In some embodiments, the first nucleic acid base (e.g., modified cytosine) is the product of hysgen cycloaddition to β-6-azido-glucosyl-5-hydroxymethylcytosine containing an affinity label (e.g., biotin).

[0147] In any of the combinations described herein, the captured DNA may include cfDNA. The captured DNA may have any of the features described herein with respect to the captured set, including, for example, a higher concentration of DNA corresponding to a sequence variable target region set (normalized with respect to footprint size as discussed above) than the concentration of DNA corresponding to an epigenetic target region set. In some embodiments, the DNA of the captured set includes a sequence tag, which can be attached to the DNA as described herein. Generally, including a sequence tag results in a DNA molecule that differs from the naturally occurring, untagged form.

[0148] The combination may further include the probe sets or sequencing primers described herein, each of which may differ from naturally occurring nucleic acid molecules. For example, the probe sets described herein may include a capture portion, and the sequencing primers may include labels that do not exist in nature.

[0149] Computer systems, processing of real-world evidence (RWE) The methods of the present disclosure can be implemented using a computer system or with the assistance of a computer system. For example, such a method may include the steps of: distributing a sample into a plurality of subsamples, including a first subsample and a second subsample, wherein the first subsample contains a higher proportion of cytosine-modified DNA than the second subsample; subjecting the first subsample to a procedure that affects the first nucleic acid base of the DNA of the first subsample differently from the second nucleic acid base of the DNA, wherein the first nucleic acid base is a modified or unmodified nucleic acid base, the second nucleic acid base is a modified or unmodified nucleic acid base different from the first nucleic acid base, and the first and second nucleic acid bases have the same base-pairing specificity; and sequencing the DNA in the first subsample and the DNA in the second subsample in such a manner that the first and second nucleic acid bases of the DNA of the first subsample are distinguishable.

[0150] In one embodiment, the Disclosure provides a non-temporary computer-readable medium containing computer-executable instructions that, when executed by at least one electronic processor, carries out at least part of a method comprising: collecting cfDNA from a test subject; capturing a plurality of sets of target regions from the cfDNA, wherein the plurality of target region sets include a sequence-variable target region set and an epigenetic target region set, thereby creating a set of captured cfDNA molecules; sequencing the captured cfDNA molecules, wherein the captured cfDNA molecules of the sequence-variable target region set are sequenced to a higher sequencing depth than the captured cfDNA molecules of the epigenetic target region set; obtaining a plurality of sequence reads generated by sequencing the captured cfDNA molecules using a nucleic acid sequencer; mapping the plurality of sequence reads to one or more reference sequences to generate mapped sequence reads; and processing the mapped sequence reads corresponding to the sequence-variable target region set and the mapped sequence reads corresponding to the epigenetic target region set to determine the likelihood that the subject has cancer.

[0151] The code may be pre-compiled and configured for use in a machine having a processor adapted to run the code, or it may be compiled during execution time. The code may be supplied in a programming language that can be selected to allow the code to be executed in a pre-compiled manner or in an as-compiled manner.

[0152] Further details regarding computer systems and networks, databases, and computer program products are also provided, for example, in Peterson, Computer Networks: A Systems Approach, Morgan Kaufmann, 5th Ed. (2011), Kurose, Computer Networking: A Top-Down Approach, Pearson, 7th Ed. (2016), each of which is thus incorporated herein by reference in its entirety, in Elmasri, Fundamentals of Database Systems, Addison Wesley, 6th Ed. (2010), in Coronel, Database Systems: Design, Implementation, & Management, Cengage Learning, 11th Ed. (2014), in Tucker, Programming Languages, McGraw-Hill Science / Engineering / Math, 2nd Ed. (2006), and in Rhoton, Cloud Computing Architected: Solution Design Handbook, Recursive Press (2011).

[0153] Figure 1 shows an example of architecture 100 for generating an integrated data repository containing multiple types of health management data, according to one or more implementations. Architecture 100 may include a data integration and analysis system 102. The data integration and analysis system 102 can obtain data from several data sources and integrate the data from the data sources into the integrated data repository 104. For example, the data integration and analysis system 102 can obtain data from a health insurance claims data repository 106. In various examples, the data integration and analysis system 102 and the health insurance claims data repository 106 may be created and maintained by different entities. In one or more additional examples, the data integration and analysis system 102 and the health insurance claims data repository 106 may be created and maintained by the same entity.

[0154] The data integration and analysis system 102 can be implemented by one or more computer computing devices. These may include one or more server computer computing devices, one or more desktop computer computing devices, one or more laptop computer computing devices, one or more tablet computer computing devices, one or more mobile computer computing devices, or a combination thereof. In certain implementations, at least some of the one or more computer computing devices can be implemented in a distributed computing environment. For example, at least some of the one or more computer computing devices can be implemented in a cloud computing architecture. In scenarios where the computer computing system used to implement the data integration and analysis system 102 is configured as a distributed computing architecture, processing operations can be performed in parallel by a large number of virtual machines. In various examples, the data integration and analysis system 102 may implement multithreading techniques. The implementation of a distributed computing architecture and multithreading techniques allows the data integration and analysis system 102 to utilize fewer computing resources compared to a computer computing architecture that does not implement these techniques.

[0155] The medical insurance claims data repository 106 can store information obtained from one or more medical insurance companies corresponding to insurance claims made by policyholders of one or more medical insurance companies. The medical insurance claims data repository 106 can be sorted (e.g., filtered) by patient identifiers. Patient identifiers may be based on the patient's first name, last name, date of birth, social security number, address, employer, etc. The data stored by the medical insurance claims data repository 106 may include structured data placed in one or more data tables. One or more data tables storing structured data may include a number of rows and a number of columns that show information about medical insurance claims made by policyholders of one or more medical insurance companies in relation to procedures and / or treatments received by the policyholders from healthcare providers. At least some of the rows and columns of the data tables stored by the medical insurance claims data repository 106 may include medical insurance codes that may indicate diagnoses and treatments and / or procedures for biological conditions obtained by policyholders of one or more medical insurance companies. In various examples, medical insurance codes may also indicate diagnostic procedures for one or more biological conditions obtained or that may be present in an individual. In one or more examples, a diagnostic procedure may provide information used to detect the presence of a biological condition. The diagnostic procedure may also provide information used to determine the progression of the biological condition. In one or more exemplary examples, a diagnostic procedure may include one or more imaging procedures, one or more assays, one or more test procedures, or one or more combinations thereof.

[0156] The data integration and analysis system 102 can also obtain information from the molecular data repository 108. The molecular data repository 108 can store data on genomic information, genetic information, metabolomics information, transcriptomics information, fragmentomics information, immune receptor information, methylation information, epigenomic information, and / or proteomics information for a certain number of individuals. In one or more examples, the data integration and analysis system 102 and the molecular data repository 108 may be created and maintained by different entities. In one or more additional examples, the data integration and analysis system 102 and the molecular data repository 108 may be created and maintained by the same entity.

[0157] Genomic information may indicate one or more mutations corresponding to an individual's genes. Mutations in an individual's genes may correspond to differences between the individual's nucleic acid sequence and one or more reference genomes. The reference genome may include known reference genomes such as hg19. In various examples, mutations in an individual's genes may correspond to differences in the individual's germline genes compared to a reference genome. In one or more additional examples, the reference genome may include the individual's germline genome. In one or more further examples, mutations in an individual's genes may include somatic mutations. Mutations in an individual's genes may be associated with insertions, deletions, single nucleotide variants, loss of heterozygosity, duplications, amplifications, translocations, fusion genes, or one or more combinations thereof.

[0158] In one or more exemplary cases, the genomic information stored in the molecular data repository 108 may include the genomic profile of tumor cells present in an individual. In these situations, the genomic information may be obtained, but is not limited to, from genetic material derived from circulating nucleic acids (e.g., cell-free DNA) present as a result of the degradation of tumor cells present in the individual, such as from tissue samples or tumor biopsy materials, circulating tumor cells (CTCs), exosomes or efferosomes, or found in blood samples of the individual, e.g., by analysis of deoxyribonucleic acid (DNA) and / or ribonucleic acid (RNA). In one or more examples, the genomic information of tumor cells in an individual may correspond to one or more target regions. One or more mutations present in one or more target regions may indicate the presence of tumor cells in the individual. The genomic information stored in the molecular data repository 108 may be generated in association with assays or other diagnostic tests that can determine one or more mutations in one or more target regions of a reference genome.

[0159] In one or more additional examples, the data integration and analysis system 102 may obtain information from one or more additional data repositories 110. One or more additional data repositories 110 may store data relating to electronic medical records of individuals whose data exists in at least one of the health insurance claims data repository 106 or the molecular data repository 108. Furthermore, one or more additional data repositories 110 may store data relating to pathology reports of individuals whose data exists in at least one of the health insurance claims data repository 106 or the molecular data repository 108. In various examples, one or more additional data repositories 110 may store data relating to biological conditions and / or treatments for biological conditions. In one or more examples, the data integration and analysis system 102 and at least a portion of one or more additional data repositories 110 may be created and maintained by different entities. In one or more further examples, the data integration and analysis system 102 and at least a portion of one or more additional data repositories 110 may be created and maintained by the same entity.

[0160] In one or more further implementations, the data integration and analysis system 102 may obtain information from one or more reference information data repositories 112. One or more reference information data repositories 112 may store information including definitions, standards, protocols, terminology, and one or more combinations thereof. In various examples, the information stored by one or more reference information data repositories may correspond to biological conditions and / or treatments for biological conditions. In one or more exemplary examples, one or more reference information data repositories 112 may include RxNorm (RxNorm provides standardized names for clinical drugs, associating these names with many of the drug terms used in pharmacy management and drug interaction software). In one or more examples, the data integration and analysis system 102 and at least a portion of one or more reference information data repositories 112 may be created and maintained by different entities. In one or more further examples, the data integration and analysis system 102 and at least a portion of one or more reference information data repositories 112 may be created and maintained by the same entity.

[0161] The data integration and analysis system 102 is accessible from at least one of the following: the medical insurance claims data repository 106, the molecular data repository 108, one or more additional data repositories 110, or the reference information data repository 112. It can also obtain data via one or more communication networks that provide access to at least one of these. Furthermore, the data integration and analysis system 102 can obtain data from at least one of the following: the medical insurance claims data repository 106, the molecular data repository 108, one or more additional data repositories 110, or the reference information data repository 112, via one or more secure communication channels. Additionally, the data integration and analysis system 102 can obtain data from at least one of the following: the medical insurance claims data repository 106, the molecular data repository 108, one or more additional data repositories 110, or the reference information data repository 112, via one or more calls to an application programming interface (API).

[0162] The data integration and analysis system 102 may include a data integration system 114. The data integration system 114 can obtain data from the health insurance claims data repository 106 and the molecular data repository 108 to generate an integrated data repository 104. The data integration system 114 can also obtain data from one or more additional data repositories 110 to generate the integrated data repository 104. In various examples, the data integration system 114 can implement one or more natural language processing techniques to integrate data from one or more additional data repositories 110 into the integrated data repository 104.

[0163] In one or more examples, the data integration system 114 can generate one or more tokens to identify individuals whose data is stored in the medical insurance claims data repository 106 and whose data is stored in the molecular data repository 108. In various examples, the data integration system 114 can generate one or more tokens by implementing one or more hash functions. The data integration system 114 can implement one or more hash functions to generate one or more tokens based on information stored in at least one of the medical insurance claims data repository 106 or the molecular data repository 108. For example, the information that the data integration system 114 uses to generate individual tokens by implementing a hash function may include at least one of the following: the identifier of each individual, the date of birth of each individual, the postal code of each individual, the date of birth of each individual, or the gender of each individual. In one or more exemplary examples, the identifier of each individual may include a combination of at least part of the first name and at least part of the last name of each individual. Tokens generated using data from different data repositories may correspond to the same or similar information or the same or similar type of information stored in the different data repositories. For example, a token can be generated using part of an individual's full name, date of birth, at least part of their postal code, and gender, obtained from the medical insurance claims data repository 106 and the molecular data repository 108.

[0164] The data integration system 114 can integrate data from a number of different data sources by parsing tokens generated by implementing one or more hash functions using the data obtained from those different data sources. For example, the data integration system 114 can obtain one or more first tokens generated from data stored in the medical insurance claims data repository 106 and one or more second tokens generated from data stored in the molecular data repository 108. The data integration system 114 can parse one or more first tokens in relation to one or more second tokens to determine individual first tokens corresponding to individual second tokens. In one or more exemplary examples, the data integration system 114 can identify individual first tokens that match individual second tokens. A first token may match a second token if its data has at least a threshold amount of similarity to the data of the second token. In one or more examples, a first token may match a second token if its data is the same as the data of the second token. For example, if the alphanumeric string of the first token is the same as the alphanumeric string of the second token, then the first token may match the second token.

[0165] By determining the first token generated using data stored in the medical insurance claims data repository 106, which corresponds to a second token generated using data stored in the molecular data repository 108, the data integration system 114 can identify individuals whose data is stored in both the medical insurance claims data repository 106 and the molecular data repository 108. In this way, the data integration system 114 can obtain data from the medical insurance claims data repository 106 and data from the molecular data repository 108 from a certain number of individuals, and store the medical insurance claims data and molecular data for that number of individuals in the integrated data repository 104.

[0166] The data integration system 114 can also generate an integrated data repository 104 by integrating data stored in one or more additional data repositories 110 with data from the medical insurance claims data repository 106 and the molecular data repository 108. For example, the data integration system 114 can obtain one or more third tokens generated from data stored in an additional data repository 110, for example, a data repository where data corresponding to pathology reports is stored. The data integration system 114 can parse one or more third tokens in relation to a first token generated using information stored in the medical insurance claims data repository 106 and a second token generated using information stored in the molecular data repository 108 to determine each third token corresponding to each individual first token and each individual second token. In one or more exemplary examples, the data integration system 114 can identify a third token generated using one or more hash functions and a common set of information obtained from the medical insurance claims data repository 106, the molecular data repository 108, and the additional data repositories 110.

[0167] By determining a third token generated using data stored in an additional data repository 110, corresponding to a first token generated using data stored in the medical insurance claims data repository 106 and a second token generated using data stored in the molecular data repository 108, the data integration system 114 can identify individuals whose data is stored in the medical insurance claims data repository 106, the molecular data repository 108, and the additional data repository 110. In this way, the data integration system 114 can obtain data from the medical insurance claims data repository 106 from a certain number of individuals, as well as data from the molecular data repository 108 and the additional data repository 110 from the same number of individuals, and store the medical insurance claims data, molecular data, and additional data for that number of individuals in the integrated data repository 104.

[0168] The data stored in the integrated data repository 104 for that number of individuals may be accessible using the identifier of each individual. The data integration system 114 may implement several techniques as part of the de-identification process for storing and retrieving information about individuals in the integrated data repository 104. An individual identifier may correspond to a key generated using at least one hash function. An individual identifier may also be generated by implementing one or more salting processes with respect to a key generated using at least one hash function. A token is generated using one or more hash functions and a common set of information obtained from the health insurance claims data repository 106, the molecular data repository 108, and / or additional data repositories 110. In one or more exemplary examples, the identifier generated by the data integration system 114 for accessing information about each individual stored in the integrated data repository 104 may be unique to each individual. In one or more examples, an individual identifier may be generated using at least some of the information used to generate the token associated with the individual. In one or more additional examples, an individual identifier may be generated using information different from the information used to generate the token associated with the individual.

[0169] The data integration system 114 can also generate the integrated data repository 104 from several different combinations of data repositories. For example, the data integration system 114 can obtain tokens generated from information stored in the health insurance claims data repository 106 and additional tokens generated from information stored in one or more additional data stores 110. The data integration system 114 can determine individual tokens generated from information stored in the health insurance claims data repository 106 that correspond to individual additional tokens generated from information stored in one or more additional data repositories 110. By determining tokens generated using data stored in the health insurance claims data repository 106 that correspond to additional tokens generated using data stored in the additional data repository 110, the data integration system 114 can identify individuals whose data is stored in both the health insurance claims data repository 106 and the additional data repositories 110. Thus, the data integration system 114 can obtain data from the medical insurance claims data repository 106 originating from a certain number of individuals and data from an additional data repository 110 originating from the same number of individuals, and store the medical insurance claims data and additional data for that number of individuals in the integrated data repository 104. The medical insurance claims data and the additional data stored in the integrated data repository 104 for that number of individuals may be accessible using the identifier of each individual.

[0170] In one or more further examples, the data integration system 114 can obtain tokens generated from information stored in the molecular data repository 108 and tokens generated from information stored in one or more additional data stores 110. The data integration system 114 can determine individual tokens generated from information stored in the molecular data repository 108 that correspond to individual additional tokens generated from information stored in one or more additional data repositories 110. By determining tokens generated using data stored in the molecular data repository 108 that correspond to additional tokens generated using data stored in the additional data repositories 110, the data integration system 114 can identify individuals for whom data is stored in both the molecular data repository 108 and the additional data repositories 110. In this way, the data integration system 114 can obtain data from the molecular data repository 108 and data from the additional data repositories 110 from a certain number of individuals, and store the molecular data and additional data for that number of individuals in the integrated data repository 104. The molecular data and the additional data stored in the integrated data repository 104 for that number of individuals may be accessible using the respective identifiers of the individuals.

[0171] The data stored in the integrated data repository 104 may be stored in accordance with one or more regulatory frameworks that protect privacy and ensure the security of individuals' medical records, health information, and insurance information. For example, data may be stored in the integrated data repository 104 in accordance with one or more government regulatory frameworks that target the protection of personal information, such as the Health Insurance Portability and Accountability Act (HIPAA) and / or the General Data Protection Regulation (GDPR). The integrated data repository 104 also stores data in an anonymized and non-identifiable form to ensure the protection of privacy of individuals whose data is stored in the integrated data repository 104. To further ensure the privacy of individuals whose data is stored in the integrated data repository 104, the data integration system 114 may periodically regenerate the integrated data repository 104. For example, the data integration system 114 may create the integrated data repository 104 once quarterly. In one or more additional examples, the data integration system 114 may generate the integrated data repository 104 once monthly, weekly, or every two weeks. By periodically regenerating the integrated data repository 104, rather than simply refreshing it when new data becomes available, the integrated data repository 104 enhances privacy protections regarding the data stored by the integrated data repository 104. In other words, in a situation where the data repository is simply refreshed with new data, the number of new individuals added at a given time is typically less than the number of existing individuals whose data is already stored in the data repository, making it potentially easier to track individuals associated with newly added data in the data repository.

[0172] In various examples, the data stored in the integrated data repository 104 can be accessed via a database management system. Furthermore, the integrated data repository 104 can store data according to one or more database models. In one or more examples, the integrated data repository 104 can store data according to one or more relational database technologies. For example, the integrated data repository 104 can store data according to a relational database model. In one or more additional examples, the integrated data repository 104 can store data according to an object-oriented database model. In one or more further examples, the integrated data repository 104 can store data according to an extensible markup language (XML) database model. In yet another example, the integrated data repository 104 can store data according to a structured query language (SQL) database model. In yet another example, the integrated data repository can store data according to an image database model.

[0173] The data integration system 114 can generate an integrated data repository 104 by generating a certain number of data tables and creating links between the data tables. The links may represent logical couplings between the data tables. The data integration system 114 can generate data tables by extracting specific sets of data from information obtained from data repositories 106, 108, 110, and 112 and storing the data in the rows and columns of the respective data tables. In various examples, logical couplings between data tables may include at least one of the following: a one-to-one link where a row of information in one data table corresponds to a row of information in another data table; a one-to-many link where a row of information in one data table corresponds to multiple rows of information in another data table; or a many-to-many link where multiple rows of information in one data table correspond to multiple rows of information in another data table.

[0174] The number of data tables can be arranged according to the data repository schema 116. In the exemplary example in Figure 1, the data repository schema 114 includes the first data table 118, the second data table 120, the third data table 122, the fourth data table 124, and the fifth data table 124. While the exemplary example in Figure 1 includes five data tables, additional implementations may include more or fewer data tables in the data repository schema 116. The data repository schema 116 may also include links between data tables 118, 120, 122, 124, and 126. Links between data tables 118, 120, 122, 124, and 126 may indicate that information retrieved from one of the data tables 118, 120, 122, 124, and 126 leads to the retrieval of additional information stored in one or more additional data tables 118, 120, 122, 124, and 126. Furthermore, not all data tables 118, 120, 122, 124, and 126 may be linked to each other. In the example in Figure 1, the first data table 118 is logically joined to the second data table 118 by the first link 128, and the first data table 118 is logically joined to the fourth data table 124 by the second link 130. Furthermore, the second data table 120 is logically joined to the third data table 122 via the third link 132, and the fourth data table 124 is logically joined to the fifth data table 126 via the fourth link 134. Furthermore, the third data table 122 is logically joined to the fifth data table 126 via the fifth link 136.

[0175] In various examples, when a data table is added to and / or removed from the data repository schema 116, additional links between data tables can be added to or removed from the data repository schema 116. In one or more exemplary examples, the integrated data repository 104 can store data tables according to the data repository schema 116 for at least a portion of individuals from which the data integration system 114 has obtained information from at least two combinations of the health insurance claims data repository 106, the molecular data repository 108, one or more additional data repositories 110, and one or more reference information data repositories 112. As a result, the integrated data repository 104 can store each of the data table 118, 120, 122, 124, and 126 according to the data repository schema 116 for individuals ranging from thousands, tens of thousands, hundreds of thousands, or more.

[0176] The data integration and analysis system 102 may also include a data pipeline system 138. The data pipeline system 138 may include several algorithms, software code, scripts, macros, or other computer-executable instructions that process information stored in the integrated data repository 104 to generate additional datasets. The additional datasets may include information obtained from one or more of the data tables 118, 120, 122, 124, and 126. The additional datasets may also include information derived from data obtained from one or more of the data tables 118, 120, 122, 124, and 126. The components of the data pipeline system 138 implemented to generate a first additional dataset may differ from the components of the data pipeline system 138 used to generate a second additional dataset.

[0177] In one or more examples, the data pipeline system 138 can generate a dataset showing the pharmaceutical procedures received by a certain number of individuals. In one or more exemplary examples, the data pipeline system 138 can analyze information stored in at least one of the data tables 118, 120, 122, 124, and 126 to determine the health insurance codes corresponding to the pharmaceutical procedures received by a certain number of individuals. The data pipeline system 138 can analyze the health insurance codes corresponding to the pharmaceutical procedures in relation to a library of data showing specific pharmaceutical procedures corresponding to one or more health insurance codes to determine the names of the pharmaceutical procedures received by individuals. In one or more additional examples, the data pipeline system 138 can analyze information stored in the integrated data repository 104 to determine the medical procedures received by a certain number of individuals. For example, the data pipeline system 138 can analyze information stored in one of the data tables 118, 120, 122, 124, and 126 to determine at least one injection or intravenous procedure received by an individual. In one or more further examples, the data pipeline system 138 can analyze information stored in the integrated data repository 104 to determine episodes of care for an individual, lines of treatment received, progression of biological condition, or time to the next treatment. In various examples, the datasets generated by the data pipeline system 138 may differ for different biological conditions. For example, the data pipeline system 138 can generate a first number of datasets for a first type of cancer, e.g., lung cancer, and a second number of datasets for a second type of cancer, e.g., colorectal cancer.

[0178] The data pipeline system 138 can also determine one or more confidence levels to be assigned to information relating to individuals whose data is stored in the integrated data repository 104. Each confidence level may correspond to a different measure of accuracy for information relating to individuals whose data is stored in the integrated data repository 104. The information associated with each confidence level may correspond to one or more characteristics of individuals derived from the data stored in the integrated data repository 104. The confidence level values ​​for one or more characteristics can be generated by the data pipeline system 138 in conjunction with the generation of one or more datasets from the integrated data repository 104. In one or more examples, the first confidence level may correspond to a first range of accuracy measures, the second confidence level may correspond to a second range of accuracy measures, and the third confidence level may correspond to a third range of accuracy measures. In one or more additional examples, the second range of accuracy measures may include values ​​smaller than those of the first range of accuracy measures, and the third range of accuracy measures may include values ​​smaller than those of the second range of accuracy measures. In one or more exemplary examples, information corresponding to a first confidence level may be referred to as gold standard information, information corresponding to a second confidence level may be referred to as silver standard information, and information corresponding to a third confidence level may be referred to as bronze standard information.

[0179] The data pipeline system 138 can determine the confidence level value for an individual's features based on several factors. For example, an individual's features can be determined using each set of information. The data pipeline system 138 can determine the confidence level for an individual's features based on the degree of completeness of each set of information used to determine the features of the individual. In a situation where one or more parts of information are missing from the set of information relating to a first number of individuals, the confidence level for the features may be lower than the confidence level for the features of a second number of individuals where no information is missing from the set of information. In one or more examples, the data pipeline system 138 can use the degree of information loss to determine the confidence level for an individual's features. For example, a higher degree of information loss used to determine an individual's features may result in a lower confidence level for the features than in a situation where the information loss is lower. Furthermore, different types of information can correspond to different confidence levels for features. In one or more examples, the presence of a first part of information used to determine an individual's features may result in a higher confidence level for the features than the presence of a second part of information used to determine the features.

[0180] In one or more exemplary examples, the data pipeline system 138 can determine the number of individuals that belong to a cohort with a primary diagnosis of lung cancer (or other biological condition). For each individual, the data pipeline system 138 can determine a confidence level regarding that the individual is classified as having a primary diagnosis of lung cancer. The data pipeline system 138 can use information from a certain number of columns contained in data tables 118, 120, 122, 124, and 126 to determine a confidence level regarding that an individual belongs to the lung cancer cohort. The number of columns may include health insurance codes relating to the diagnosis and / or treatment of the biological condition. Furthermore, the number of columns may correspond to the date of diagnosis and / or treatment for the biological condition. The data pipeline system 138 can determine that the confidence level for an individual characterized as part of the lung cancer cohort is higher in scenarios where information is available for each of the number of columns or at least a threshold number of columns than when information is available for fewer than a threshold number of columns. Furthermore, the data pipeline system 138 can determine the confidence level for individuals belonging to the lung cancer cohort based on the type of information associated with one or more columns and the availability of that information. For example, in a situation where one or more diagnostic codes exist for a group of individuals in association with one or more time periods, and one or more treatment codes do not exist, the data pipeline system 138 can determine that the confidence level for the inclusion of the group of individuals in the lung cancer cohort is higher than in a situation where at least one diagnostic code is absent and there are treatment codes used to determine whether an individual is included in the lung cancer cohort.

[0181] The data integration and analysis system 102 may include a data analysis system 140. The data analysis system 148 can receive integrated data repository requests 142 from one or more computer computing devices, for example, an exemplary computer computing device 144. One or more integrated data repository requests 142 may cause data to be retrieved from the integrated data repository 104. In various examples, one or more integrated data repository requests 142 may cause data to be retrieved from one or more datasets generated by the data pipeline system 138. The integrated data repository request 142 may specify the data to be retrieved from one or more datasets generated by the integrated data repository 104 and / or the data pipeline system 138. In one or more additional examples, the integrated data repository request 142 may include one or more pre-built queries corresponding to computer-executable instructions to retrieve a specified set of data from one or more datasets generated by the integrated data repository 104 and / or the data pipeline system 138.

[0182] In response to one or more integrated data repository requests 142, the data analysis system 140 may analyze data retrieved from the integrated data repository 104 or at least one of the one or more datasets generated by the data pipeline system 138 to generate data analysis results 146. The data analysis results 146 may be sent to one or more computer computing devices, for example, an exemplary computer computing device 148. In the exemplary example of Figure 1, one or more integrated data repository requests 142 and the data analysis results 146 are shown to be sent from one computer computing device 144 to another computer computing device 148, but in one or more additional implementations, the data analysis results 146 may be received by the same computer computing device that sent the one or more integrated data repository requests 142. The data analysis results 146 can be displayed by one or more user interfaces rendered by computer computing device 144 or computer computing device 148.

[0183] In one or more examples, the data analysis system 140 can implement at least one of one or more machine learning techniques or one or more statistical techniques to analyze data retrieved in response to one or more integrated data repository requests 142. In one or more examples, the data analysis system 140 can implement one or more artificial neural networks to analyze data retrieved in response to one or more integrated data repository requests 142. For example, the data analysis system 140 can implement at least one of one or more convolutional neural networks or one or more residual neural networks to analyze data retrieved from the integrated data repository 104 in response to one or more integrated data repository requests 142. In at least some examples, the data analysis system 140 can implement one or more random forest techniques, one or more support vector machines, or one or more hidden Markov models to analyze data retrieved in response to one or more integrated data repository requests 142. To identify at least one measure of correlation or significance between individual features, one or more statistical models can be implemented to analyze the data retrieved in response to one or more integrated data repository requests 142. For example, a log-rank test can be applied to the data retrieved in response to one or more integrated data repository requests 142. Furthermore, a Cox proportional hazards model can be implemented on the data retrieved in response to one or more integrated data repository requests 142. Furthermore, a Wilcoxon signed-rank test can be applied to the data retrieved in response to one or more integrated data repository requests 142. In yet another example, z-score analysis can be performed on the data retrieved in response to one or more integrated data repository requests 142.In further examples, Kaplan-Meier analysis can be performed on data retrieved in response to one or more integrated data repository requests 142. In at least some examples, one or more machine learning techniques can be implemented in combination with one or more statistical techniques to analyze data retrieved in response to one or more integrated data repository requests 142.

[0184] In one or more exemplary examples, the data analysis system 140 can determine the survival rate of an individual with lung cancer in response to one or more treatments. In one or more additional exemplary examples, the data analysis system 140 can determine the survival rate of an individual with lung cancer and one or more genomic region mutations in response to one or more treatments. In various examples, the data analysis system 140 can generate data analysis results 146 when data retrieved from at least one of one or more datasets generated by the integrated data repository 104 or the data pipeline system 138 meets one or more criteria. For example, the data analysis system 140 can determine whether at least a portion of the data retrieved in response to one or more integrated data repository requests 142 meets a threshold confidence level. If the confidence level for at least a portion of the data retrieved in response to one or more integrated data repository requests 142 is lower than the threshold confidence level, the data analysis system 140 may not generate at least a portion of the data analysis results 146. In scenarios where the confidence level for at least a portion of the data retrieved in response to one or more integrated data repository requests 142 is at least the threshold confidence level, the data analysis system 140 can generate at least a portion of the data analysis results 146. In various examples, the threshold confidence level may be related to the type of data analysis results 146 generated by the data analysis system 140.

[0185] In one or more exemplary examples, the data analysis system 140 may receive an integrated data repository request 142 and generate a data analysis result 146 showing the survival rate of one or more individuals. In these cases, the data analysis system 140 can determine whether the data stored in one or more datasets generated by the integrated data repository 104 and / or the data pipeline system 138 meets a threshold confidence level, such as the gold standard confidence level. In one or more additional examples, the data analysis system 140 may receive an integrated data repository request 142 and generate a data analysis result 146 showing the treatment received by one or more individuals. In these implementations, the data analysis system 140 can determine whether the data stored in one or more datasets generated by the integrated data repository 104 and / or the data pipeline system 138 meets a lower threshold confidence level, such as the bronze standard confidence level.

[0186] In one or more additional exemplary examples, the data analysis system 140 can receive an integrated data repository request 142 and identify individuals that have one or more genomic mutations and have received one or more treatments for a biological condition. Continuing this example, the data analysis system 140 can determine the survival rate of individuals with one or more genomic mutations in relation to the one or more treatments they have received. The data analysis system 140 can then identify the effectiveness of the treatments on the individuals, based on their survival rates, in relation to the genomic mutations that may be present in them. In this way, the health outcomes of individuals can be improved by identifying treatments that may be more effective than those currently offered to individuals in a population of individuals with one or more genomic mutations.

[0187] Figure 2 shows an example of a framework 200 corresponding to the arrangement of data tables in an integrated data repository, according to one or more implementations. In the exemplary example in Figure 2, the framework 200 includes a data repository schema 202, which includes the first data table 204, the second data table 206, the third data table 208, the fourth data table 210, the fifth data table 212, the sixth data table 214, and the seventh data table 216. While the exemplary example in Figure 2 includes seven data tables, additional implementations may include more or fewer data tables in the data repository schema 202. The data repository schema 202 may also include links between data tables 204, 206, 208, 210, 212, 214, and 216. Links between data tables 204, 206, 208, 210, 212, 214, and 216 may indicate that information retrieved from one of data tables 204, 206, 208, 210, 212, 214, and 216 may lead to the retrieval of additional information stored in one or more additional data tables 204, 206, 208, 210, 212, 214, and 216. Furthermore, not all data tables 204, 206, 208, 210, 212, 214, and 216 may be linked to each of the other data tables 204, 206, 208, 210, 212, 214, and 216. In the example shown in Figure 2, the first data table 204 is logically joined to the second data table 206 by the first link 218, the third data table 208 is logically joined to the second data table 206 by the second link 220, the second data table 206 is also logically joined to the fourth data table 210 by the third link 222, the second data table 206 is logically joined to the fifth data table 212 by the fourth link 224, and the second data table 206 is logically joined to the sixth data table 214 by the fifth link 226.Furthermore, the fifth data table 212 is logically joined to the sixth data table 214 by the sixth link 228, and the sixth data table 214 is logically joined to the seventh data table 216 by the seventh link 230. Furthermore, the seventh data table 216 is logically joined to the fourth data table 210 by the eighth link 232. In various examples, when a data table is added to and / or removed from the data repository schema 202, additional links between data tables can be added to or removed from the data repository schema 202. In one or more exemplary examples, the integrated data repository 104 may store data tables according to the data repository schema 202 for at least some of the individuals from which the data integration system 114 has obtained information from at least two combinations of the medical insurance claims data repository 106, the molecular data repository 108, and one or more additional data repositories 110. As a result, the integrated data repository 104 can store thousands, tens of thousands, hundreds of thousands, or even more individuals, with each case in data tables 204, 206, 208, 210, 212, 214, and 216 according to the data repository schema 204.

[0188] In one or more examples, the first data table 204 may store data corresponding to genomics and genomic testing for an individual. For example, the first data table 204 may include columns containing genomics data, genomic region mutations, mutation types, copy numbers of genomic regions, coverage data indicating the number of nucleic acid molecules identified in a sample with one or more mutations, the test date, and information corresponding to the panel used to generate patient information. The first data table 204 may also include one or more columns containing health insurance data codes that may correspond to one or more diagnostic codes. Furthermore, the information in the first data table 204 may include at least one identifier for the individual related to the case in the first data table 204.

[0189] The second data table 206 can store data related to one or more patient visits by an individual to one or more healthcare providers. The third data table 208 can store information corresponding to each service provided to the individual in relation to one or more patient visits to one or more healthcare providers indicated by the second data table 206. For example, an individual may visit a healthcare provider, and multiple services may be performed on the individual during the visit. The second data table 206 may include columns showing information about each of the multiple services performed during the patient visit. Multiple third data tables 208 can be generated for each patient visit, and each third data table 208 may include columns showing a more granular level of information for each service provided during the patient visit than the information stored in the second data table 206 related to the patient visit. For example, the second data table 206 may include a number of columns showing the health insurance codes for different services provided to the individual during the patient visit, and the third data table 208 for one of the services may include a number of columns for additional health insurance codes corresponding to additional information about that service. The second data table 206 and the third data table (or more) 208 for patient visits may show the dates of one or more services corresponding to the patient visit.

[0190] The fourth data table 210 may include columns that show information about individuals whose information is stored in the integrated data repository 104. For example, the fourth data table 210 may include columns that show information about the individual's location, sex, date of birth, date of death (if applicable), or at least one of the one or more keys associated with the individual. In one or more examples, the fourth data table 210 may include one or more columns that indicate whether incorrect data has been identified about the individual. In various examples, a separate fourth data table 210 can be generated for each individual. Thus, the data repository schema 202 may include multiple instances of the fourth data table 210, for example, thousands, tens of thousands, hundreds of thousands, or even more.

[0191] The fifth data table 212 may include columns showing information relating to a health insurance company or government agency that makes payments for one or more services provided to each individual. For example, the fifth data table 212 may include one or more payer identifiers. The sixth data table 214 may include columns showing information corresponding to health insurance coverage information for each individual. In one or more examples, the sixth data table 214 may include columns showing the existence of health insurance for the individual, the existence of prescription drug coverage for the individual, and the type of health insurance plan for the individual, such as a Health Maintenance Organization (HMO), a Priority Medical Benefit Organization (PPO), etc.

[0192] The seventh data table 216 may include columns showing information related to the medical treatment received by each individual. In one or more examples, the seventh data table 216 may include one or more columns showing the health insurance codes corresponding to the medical treatment, which are available through the pharmacy. The health insurance codes may correspond to individual medical treatments. Furthermore, the health insurance codes may indicate a diagnosis of the biological condition of the individual. The seventh data table 216 may also include additional information such as dosage, supply days, dispensing amount, number of permitted refills, date of service, or at least one of the information about the individual receiving the medical treatment.

[0193] In various examples, data repository schema 202 can provide the results of parsing the information stored by data tables 204, 206, 208, 210, 212, 214, and 216 more efficiently than a typical data repository schema. For example, the logical connections between data tables 204, 206, 208, 210, 212, 214, and 216 are arranged so that related data can be efficiently retrieved across different data tables 204, 206, 208, 210, 212, 214, and 216. In situations where data tables 204, 206, 208, 210, 212, 214, and 216 are arranged contiguously and / or where a larger number of data tables 204, 206, 208, 210, 212, 214, and 216 are logically joined, responding to a request for information from the integrated data repository 104 by retrieving data from one or more of the data tables 204, 206, 208, 210, 212, 214, and 216 from the integrated data repository 104 is less efficient than when the data repository schema 202 is implemented.

[0194] Figure 3 shows an architecture 300 for generating one or more datasets from information retrieved from a data repository that integrates health-related data from a number of sources, according to one or more implementations. Architecture 300 may include a data integration and analysis system 102 and an integrated data repository 104. Furthermore, the data integration and analysis system 102 may include at least a data pipeline system 138 and a data analysis system 140. The data pipeline system 138 may include a set of several data processing instructions that can be executed to generate each dataset, which can be analyzed by the data analysis system 140 in response to an integrated data repository request 142 to generate a data analysis result 146.

[0195] The data pipeline system 138 may include a first data processing instruction 302, a second data processing instruction 304, and up to a nth data processing instruction 306. The data processing instructions 302, 304, and 306 may be executable by one or more processing units to perform several operations for generating their respective datasets using information obtained from the integrated data repository 104. In one or more exemplary examples, the data processing instructions 302, 304, and 306 may include at least one of the following: software code, scripts, API calls, macros, etc. The first data processing instruction 302 may be executable to generate a first dataset 308. Furthermore, the second data processing instruction 304 may be executable to generate a second dataset 310. Furthermore, the nth data processing instruction 306 may be executable to generate a nth dataset 312. In various examples, after the integrated data repository 104 is generated by the data integration and analysis system 102, the data pipeline system 138 can execute data processing commands 302, 304, and 306 to generate datasets 308, 310, and 312. In one or more examples, datasets 308, 310, and 312 may be stored in the integrated data repository 104 or in an additional data repository accessible to the data integration and analysis system 102. At least some of the data processing commands 302, 304, and 306 can parse health insurance codes to generate at least some of datasets 308, 310, and 312. Furthermore, at least some of the data processing commands 302, 304, and 306 can parse genomics data to generate at least some of datasets 308, 310, and 312.

[0196] In one or more examples, the first data processing instruction 302 may be executable to retrieve data from one or more first data tables stored by the integrated data repository 104. The first data processing instruction 302 may also be executable to retrieve data from one or more specified columns of one or more first data tables. In various examples, the first data processing instruction 302 may be executable to identify individuals having health insurance codes stored in one or more combinations of columns and rows corresponding to one or more diagnostic codes. The first data processing instruction 302 may then be executable to parse one or more diagnostic codes to determine the biological condition to which the individual was diagnosed. In one or more exemplary examples, the first data processing instruction 302 may be executable to parse one or more diagnostic codes in relation to a library of diagnostic codes indicating one or more biological conditions corresponding to each diagnostic code. The library of diagnostic codes may include hundreds to thousands of diagnostic codes. The first data processing instruction 302 may also be executable to determine the biological condition and the individual diagnosed by parse time information about the individual, such as the date of treatment, the date of diagnosis, the date of death, or one or more combinations thereof.

[0197] The second data processing instruction 304 may be executable to retrieve data from one or more second data tables stored in the integrated data repository 104. The second data processing instruction 304 may also be executable to retrieve data from one or more specified columns in one or more second data tables. In various examples, the second data processing instruction 304 may be executable to identify individuals having health insurance codes stored in one or more combinations of columns and rows corresponding to one or more treatment codes. One or more treatment codes may correspond to treatments obtained from a pharmacy. In one or more additional examples, one or more treatment codes may correspond to medical procedures, such as injections or intravenous treatments. The second data processing instruction 304 may be executable to determine one or more treatments corresponding to each health insurance code contained in one or more second data tables by parsing the health insurance codes in relation to a predetermined set of information. The predetermined set of information may include a data library showing one or more treatments corresponding to one of several hundred to several thousand health insurance codes. The second data processing instruction 304 can generate a second dataset 310 showing each treatment received by a group of individuals. In one or more exemplary examples, a group of individuals may correspond to individuals contained in a first dataset 308. The second dataset 310 may be arranged in rows and columns, where one or more rows correspond to a single individual and one or more columns indicate the treatment received by each individual.

[0198] A nth processing instruction 306 (where N is any positive integer) may be executed to generate a nth dataset 312 by combining information from a number of previously generated datasets, for example, a first dataset 308 and a second dataset 310. Furthermore, a nth processing instruction 306 may be executed to retrieve additional information from one or more additional columns of the integrated data repository 104 and to generate a nth dataset 312 that incorporates the additional information from the integrated data repository 104 together with the information obtained from the first dataset 308 and the second dataset 310. For example, a nth processing instruction 306 may be executed to identify individuals in the first dataset 308 diagnosed with a biological condition, and to parse a specified column of one or more additional data tables in the integrated data repository 104 to determine the treatment date shown in the second dataset 210 corresponding to the individuals in the first dataset 308. In one or more further examples, the nth processing instruction 306 may be executable to analyze columns of one or more additional data tables in the integrated data repository 104 to determine the dosage of the treatments shown in the second dataset 310 received by individuals included in the first dataset 308. Thus, the nth processing instruction 306 may be executable to generate an episode dataset of care based on the information included in the cohort dataset and the treatment dataset.

[0199] In one or more exemplary examples, in response to receiving an integrated data repository request 142, the data analysis system 140 may determine one or more datasets corresponding to the characteristics of the query associated with the integrated data repository request 142. For example, the data analysis system 140 may determine that the information contained in a first dataset 308 and a second dataset 310 is applicable to the response to the integrated data repository request 142. In these scenarios, the data analysis system 140 may analyze at least a portion of the data contained in the first dataset 308 and the second dataset 310 to generate a data analysis result 146. In one or more additional examples, the data analysis system 140 may determine different datasets to respond to different queries contained in the integrated data repository request 142 in order to generate a data analysis result 146.

[0200] By generating each dataset using a specific set of data processing instructions, the amount of user input to the data integration and analysis system 102 can be reduced, as can the computational load, such as the amount of processing resources and memory used to process the integrated data repository requests 142. For example, without a specific architecture for the data pipeline system 138, each time an integrated data repository request 142 is received, the data used to respond to the integrated data repository request 142 is assembled from the data repository 104. In contrast, by implementing the data pipeline system 138 to generate datasets 308, 310, and 312 by executing data processing instructions 302, 304, and 306, the data necessary to respond to various integrated data repository requests 142 is already assembled and accessible to the data analysis system 140 to respond to the integrated data repository requests 142. Therefore, the computing resources used to respond to the integrated data repository requests 142 by implementing the data pipeline system 138 to generate datasets 308, 310, and 312 are less than those used in a typical system that performs the information parsing and collection process for each integrated data repository request 142. Furthermore, in situations where the data pipeline system 138 is not implemented, users of the data integration and analysis system 102 may need to present multiple integrated data repository requests 142 to analyze the information they intend to analyze, either due to inaccuracies in ad-hoc data collection in response to integrated data repository requests 142 in a typical system, or due to the data analysis system 140 being called multiple times to perform information analysis in a typical system that could be done using a single integrated data repository request 142 if the data pipeline system 138 is implemented.

[0201] Figure 4 shows an architecture 400 for generating an integrated data repository containing de-identified health insurance claims data and de-identified genomics data, according to one or more implementations. Architecture 400 may include a data integration and analysis system 102, a health insurance claims data repository 106, and a molecular data repository 108. The data integration and analysis system 102 can obtain patient information 402 from the molecular data repository 108. Patient information 402 may include genomics data 404 for individuals whose data is stored in the molecular data repository 108. Genomics data 404 may represent the results of one or more nucleic acid sequencing operations that analyze the sequences of nucleic acid molecules contained in a sample obtained from an individual in relation to one or more target genomic regions. In one or more examples, the sample may be obtained from the tissue of one or more individuals. In one or more additional examples, the sample may be obtained from the bodily fluids of one or more individuals, e.g., blood or plasma. One or more target genomic regions may correspond to genomic regions corresponding to the presence of one or more biological conditions. For example, the target region may correspond to a genomic region of a reference genome that has a mutation present in an individual with a biological condition. In one or more exemplary cases, the target region may correspond to a genomic region of a reference human genome that has one or more mutations in an individual with one or more forms of cancer. Patient information 402 may also include information indicating personal information about an individual whose data is stored in the molecular data repository 108, as well as information corresponding to tests and analyses performed on samples provided by the individual.

[0202] The data integration and analysis system 102 can perform an anonymization process 406 to anonymize personal information obtained from the molecular data repository 108. The data integration and analysis system 102 can implement one or more computer techniques as part of the anonymization process to anonymize data about individuals stored in the molecular data repository 108, so that the anonymized data is personal, privacy protected, and compliant with one or more privacy regulatory frameworks. The anonymization process 406 may include accessing tokens in 408. In various examples, tokens may include alphanumeric strings. In one or more examples, tokens may be generated by the data integration and analysis system 102. In one or more additional examples, tokens may be generated by a third party and obtained by the data integration and analysis system 102.

[0203] Tokens can be generated in association with a subset 410 of patient information 402 using one or more hash functions. For example, for individuals whose information is stored in the molecular data repository 108, tokens can be generated using a combination of at least part of each individual's name, at least part of each individual's surname, at least part of each individual's date of birth, the individual's sex, and at least part of each individual's location identifier. The deidentification process 406 may also include generating identifiers for individuals whose data is stored in the molecular data repository 108 in 412. Identifiers can be generated by the data integration and analysis system 102 using one or more hash functions different from the one or more hash functions used to generate tokens. In one or more exemplary examples, the data integration and analysis system 102 can generate intermediate versions of each identifier using one or more hash functions, and then apply one or more salting techniques to the intermediate versions of the identifiers to generate the final versions of the identifiers. The salting functions include functions configured to add at least one random bit to each intermediate identifier to generate each final identifier. In various examples, the data integration and analysis system 102 may generate identifiers in 412 using at least some of the information about each individual stored in the molecular data repository 108. In one or more exemplary examples, identifiers may be generated based on patient identifiers contained in patient information 402. Identifiers generated by the data integration and analysis system 102 may be unique to each individual whose data is stored in the molecular data repository 108.

[0204] In operation 414, the data integration and analysis system 102 can generate modified patient information 416 based on identifiers. The modified patient information 416 may include genomics data 404 related to individuals associated with the molecular data repository 108 and identifiers for each individual. The modified patient information 416 may have a data structure 418. The data structure 418 may include columns containing identifiers for each individual associated with the molecular data repository 108 and several columns containing genomics data 404 related to the individuals, such as identifiers for one or more genes, one or more gene modifications, and types of gene modifications.

[0205] The data integration and analysis system 102 can generate a token file 420. The token file 420 may contain a first token 422 accessed in operation 408 for each individual whose data is stored in the molecular data repository 108. The token file 420 may have a data structure 424 which includes a number of columns containing information about each individual. The data structure 424 may include columns showing each identifier generated by the data integration and analysis system 102 and columns showing one or more first tokens 422 associated with each identifier. The data integration and analysis system 102 can transmit the token file 420 to the medical insurance claims data management system 426 coupled with the medical insurance claims data repository 106. The medical insurance claims data management system 426 can analyze the first token 422 in relation to the corresponding second token 428. The second token 428 can be accessed or generated by the medical insurance claims data management system 426. The second token 428 can be generated for individuals whose data is stored in the medical insurance claims data repository 106 using the same or similar subset of information as the subset 410 of patient information 402. For example, the second token 428 can be generated using a combination of at least part of each individual's first name, at least part of each individual's last name, at least part of each individual's date of birth, each individual's sex, and at least part of each individual's location identifier.

[0206] In various examples, the medical insurance claims data management system 426 can retrieve medical insurance claims data from the medical insurance claims data repository 106 for each individual associated with a second token 428 that matches a corresponding first token 422. The first token 422 may match the second token 428 if the data of the first token 422 has at least a threshold amount of similarity to the data of the second token 428. In one or more examples, the first token 422 may match the second token 428 if the data of the first token 422 is the same as the data of the second token 428.

[0207] In response to the identification of medical insurance claim data for individuals having each second token 428 corresponding to each first token 422, the medical insurance claim data management system 426 can generate corrected medical insurance claim data 430. The medical insurance claim data management system 426 can transmit the corrected medical insurance claim data 430 to the data integration and analysis system 102. In one or more examples, the corrected medical insurance claim data 430 can be formatted according to a data structure 432. The data structure 432 may include columns containing a subset of the second tokens 428 corresponding to the first tokens 422 and a number of columns containing the medical insurance claim data.

[0208] In operation 434, the data integration and analysis system 102 can integrate genomics data and health insurance claims data of individuals common to both the molecular data repository 108 and the health insurance claims data repository 106. The data integration and analysis system 102 can determine individuals common to both the molecular data repository 108 and the health insurance claims data repository 106 by determining the genomics data and health insurance claims data corresponding to common tokens. The data integration and analysis system 102 can determine that a first token 422 related to a portion of the genomics data 404 corresponds to a second token 428 related to a portion of the health insurance claims data by determining a measure of similarity between the first token 422 and the second token 428. In scenarios where the first token 422 has at least a threshold amount of similarity to the second token 428, the data integration and analysis system 102 can store the corresponding portions of the genomics data 404 and the corresponding portions of the health insurance claims data in association with individual identifiers in the integrated data repository, for example, the integrated data repository 104 in Figures 1, 2, and 3.

[0209] The implementation of Architecture 400 can implement cryptographic protocols that enable the integration of de-identified information from entirely different data repositories into a single data repository. This increases the security of the data stored in the integrated data repository 104. Furthermore, the cryptographic protocols implemented by Architecture 400 may enable more efficient retrieval and accurate analysis of the information stored in the integrated data repository 104 than would be possible without the use of Architecture 400's cryptographic protocols. For example, by using cryptographic techniques to generate a token file 420 containing a first token 422 based on a specified set of information stored in the molecular data repository 104, and utilizing a second token 428 generated using the same or similar cryptographic techniques for a similar or identical set of information stored in the health insurance claims data repository 106, the data integration and analysis system 102 can match information corresponding to the same individual stored in entirely different data repositories. If the cryptographic protocol of architecture 400 is not implemented, the probability of information from a single data repository being incorrectly attributed to one or more individuals increases, thereby reducing the accuracy of the results provided by the data integration and analysis system 102 in response to the integrated data repository request 142 sent to the data integration and analysis system 102.

[0210] Figure 5 shows a framework 500 for generating datasets based on data stored in the integrated data repository 104 by a data pipeline system 138 according to one or more implementations. The integrated data repository 104 can store health insurance claim data and genomics data for a group of individuals 502. For example, the integrated data repository 104 can store information obtained from health insurance claim records 504 for a group of individuals 502. For each individual included in the group of individuals 502, the integrated data repository 104 can store information obtained from numerous health insurance claim records 504. In various examples, the information stored in the integrated data repository 104 may include and / or be derived from thousands, tens of thousands, hundreds of thousands, or even millions of health insurance claim records 504 for a given number of individuals. Furthermore, each health insurance claim record may contain numerous columns. As a result, the integrated data repository 104 may be generated by analyzing millions of columns of health insurance claim data.

[0211] Furthermore, while health insurance claims data can be organized according to a structured data format, it is typically arranged to be viewed by health insurers, patients, and healthcare providers to show monetary and insurance code information about services provided to individuals by healthcare providers. Therefore, health insurance claims data may be available in relation to the characteristics of individuals in which a biological condition exists, and it is not readily analyzed to obtain insights that may be useful in treating individuals for a biological condition. An integrated data repository 104 can be generated and the raw health insurance claims data can be analyzed and modified in a manner that allows for the determination of trends, characteristics, features, and / or insights about individuals in which one or more biological conditions may exist, by further analyzing the data stored in the integrated data repository 104. For example, health insurance codes can be stored in the integrated data repository 104 so that, for a given individual, at least one of the following can be determined based on the health insurance claims data for that individual: medical procedure, biological condition, treatment, dosage, drug manufacturer, drug distributor, or diagnosis. In various examples, the data integration and analysis system 102 can generate and implement one or more tables showing correlations between health insurance claim data and various treatments, symptoms, or biological conditions corresponding to the health insurance claim data. Furthermore, the integrated data repository 104 can be generated using genomics data records 506 of a population of individuals 502. In various examples, a large amount of health insurance claim data can be matched with genomics data for a population of individuals 502 to generate the integrated data repository 104.

[0212] The data integration and analysis system 102 can determine correlations between the presence of one or more biomarkers present in the genomics data record 506 and other characteristics of the individuals indicated by the health insurance claim record 506, by integrating genomics data records 506 and health insurance claim records 504 for a group of individuals 502, which is typically not possible with existing systems. For example, the data integration and analysis system 102 can determine one or more genomic characteristics of an individual that correspond to treatments received by the individual, the timing of treatments, the dosage of treatments, diagnoses for the individual, smoking status, the presence of one or more biological conditions, the presence of one or more symptoms of biological conditions, or one or more combinations thereof. Based on the correlations determined by the data integration and analysis system 102 using the integrated data repository 104, it is possible to identify cohorts of individuals for whom one or more treatments may be beneficial, which could not be identified with existing systems. In one or more examples, the processes and techniques implemented to integrate the medical insurance claims records 504 and the genomics claims records 506 to generate the integrated data repository 104 may be complex and efficient techniques, systems, and processes implemented to minimize the amount of computing resources used to generate the integrated data repository 104.

[0213] In one or more exemplary examples, the data pipeline system 138 can access information stored by the integrated data repository 104 to generate a dataset containing a number of additional data records 508 containing information about at least a subset of individuals in a group 502. In the exemplary example of Figure 5, the additional data records 508 include information indicating whether an individual belongs to a cohort of individuals with lung cancer. The data pipeline system 138 can determine the cohort of individuals with lung cancer by executing a set of several different data processing instructions. In various examples, the additional data records 508 may include information used to determine the status of individual 502 with respect to lung cancer, such as one or more transaction insurance identifiers, one or more International Classification of Diseases (ICD) codes, and one or more health insurance transaction dates. In addition to including a column indicating whether individual 502 belongs to a lung cancer cohort, the additional data records 508 may include a column indicating the confidence level of individual 502's status regarding the presence of lung cancer.

[0214] Figure 6 is a schematic diagram of a computer computing architecture 600 for integrating medical record data into an integrated data repository 104. In various examples, at least part of the operation of the computer computing architecture 600 can be performed by the data integration and analysis system 102 shown in Figures 1, 3, and 4. In one or more examples, at least part of the operation of the computer computing architecture 600 can be performed by one or more additional computer computing systems that are controlled, maintained, or implemented by a service provider that also controls, maintains, or implements at least one of the data integration and analysis system 102. In one or more additional examples, at least part of the operation of the computer computing architecture 600 can be performed by several servers in a distributed computing environment.

[0215] The computer computing architecture 600 may include a medical record data repository 602. The medical record data repository 602 can store medical record data from a certain number of individuals. The medical record data may include imaging information, clinical test results, diagnostic test information, clinical findings, dental health information, notes from healthcare providers, medical history forms, diagnostic request forms, medical procedure order forms, medical information charts, and one or more combinations thereof. In various examples, for a given individual, the medical record data repository 602 may store individual-related information obtained from one or more healthcare providers.

[0216] The computer computing architecture 600 can perform operations 604, which include obtaining data packages from the medical record data repository 602. In one or more examples, data packages can be obtained in response to one or more requests sent to the medical record data repository 602 for medical records corresponding to one or more individuals. In one or more additional examples, data packages can be obtained by the computer computing architecture 600 using one or more application programming interface (API) calls. In one or more exemplary examples, a first data package 606, a second data package 608, and up to a Nth data package 610 can be obtained using the computer computing architecture 600. Each of the individual data packages 606, 608, and 610 may correspond to the medical records of their respective individuals. For example, the first data package 606 may contain the medical records of a first individual, the second data package 608 may contain the medical records of a second individual, and the Nth data package 610 may contain the medical records of a third individual.

[0217] Individual data packages 606, 608, and 610 may contain several components. In one or more examples, individual data packages 606, 608, and 610 may contain individual components corresponding to medical records from different healthcare providers. In one or more additional examples, individual data packages 606, 608, and 610 may contain individual components corresponding to different parts of medical records corresponding to one or more healthcare providers. In the exemplary example of Figure 6, the second data package 608 may contain a first component 612, a second component 614, and up to an Nth component 616. In one or more exemplary examples, the first component 612 may contain a first part of an individual's medical record, the second component 614 may contain a second part of an individual's medical record, and the Nth component 616 may contain a third part of an individual's medical record. In various examples, the first component 612 may correspond to a medical record of an individual by a first healthcare provider, the second component 614 may correspond to a medical record of an individual by a second healthcare provider, and the third component may correspond to a medical record of an individual by a third healthcare provider. In one or more additional exemplary examples, the first component 612 may include a first section of the individual's medical record, e.g., one or more forms relating to diagnostic tests or procedures, and the second component 614 may include a second section of the individual's medical record, e.g., a pathology report relating to the individual.

[0218] In operation 618, the computer architecture 600 can preprocess individual data packages to identify a corpus 620 of information to be analyzed. In one or more examples, preprocessing of data packages obtained from the medical record data repository 602 may include transforming the data contained in the data packages. For example, preprocessing a data package may include converting at least a portion of the data obtained from the medical record data repository 602 into machine-coded information. For example, preprocessing a data package may include performing one or more optical character recognition (OCR) operations on at least a portion of the data packages obtained from the medical record data repository 602. By converting at least a portion of the data packages obtained from the medical record data repository 602 into machine-coded information, the data packages can be subjected to several operations, such as one or more parsing operations to identify one or more characters or strings, or one or more editing operations that cannot be performed on at least a portion of the data packages obtained from the medical record data repository 602.

[0219] In one or more examples, the preprocessing of individual data packages may include determining information contained in each data package that should be excluded from further analysis by the computer architecture 600. In various examples, one or more components of individual data packages can be excluded from the corpus 620 of information to be analyzed. For example, with respect to a second data package 608, the computer architecture 600 may determine that a first component 612 should be excluded from further analysis by the computer architecture 600. In one or more examples, the computer architecture 600 may analyze components 612, 614, and / or 616 in relation to one or more keywords to identify at least one of components 612, 614, and / or 616 to be excluded from further analysis by the computer architecture 600. In one or more exemplary examples, the computer architecture 600 can parse components 612, 614, and / or 616 to identify one or more keywords, and in response to identifying one or more keywords in components 612, 614, and / or 616, the computer architecture 600 can decide to exclude each component 612, 614, and / or 616 from further analysis by the computer architecture 600. For example, the computer architecture 600 can determine that the first component 612 of the second data package 608 is a test request form for one or more diagnostic procedures or tests. In these scenarios, the computer architecture 600 can determine that the first component 612 should be excluded from further analysis by the computer architecture 600. Furthermore, the computer architecture 600 can determine, based on one or more keywords contained in at least one of the second component 614 or the Nth component 616, that at least one of the second component 614 and / or 616 corresponds to one or more pathology reports for an individual.In these cases, the computer architecture 600 may decide that at least a portion of the second component 614 and / or at least a portion of the Nth component 616 should be included in the corpus 620 of information to be further analyzed by the computer architecture 600.

[0220] Furthermore, a subset of the components of individual data packages obtained from the medical record data repository 602 can be included in the corpus of information 620. In various examples, one or more additional actions can be performed to narrow the scope of the corpus of information 620. For example, one or more queries can be applied to a subset of the information obtained from the medical record data repository 602. One or more queries can extract information from one or more data packages that satisfy one or more queries. In at least some examples, one or more queries may be a group of queries applied to individual components of a data package. In one or more exemplary examples, the group of queries can determine which information should be included in the corpus of information 620 and which additional information should be excluded from the corpus of information 620. In one or more additional examples, one or more terms of at least one component of a data package can be excluded from the corpus of information 620.

[0221] In one or more additional exemplary examples, after it is determined that a first component 612 should be excluded from further analysis by the computing architecture 600, the computing architecture 600 can then implement one or more queries against at least one second component 614 or an Nth component 616. In these scenarios, one or more queries may determine that terms of the second component 614, for example, terms indicating the family history of one or more biological conditions, should be excluded from the corpus of information 620. In various examples, one or more queries may be directed to identify several keywords and / or combinations of keywords contained in at least one of the second component 614 or the Nth component 616. In these cases, the computing architecture 600 can exclude one or more parts of individual components of a data package containing one or more keywords or combinations of keywords from the corpus of information 620. In one or more additional examples, the computer architecture 600 may exclude from the corpus of information 620 a certain number of words, a certain number of characters, and / or a certain number of symbols that follow one or more keywords contained in one or more parts of the individual components of the data package.

[0222] Furthermore, in operation 622, the computer architecture 600 can analyze a corpus of information to determine individual characteristics. In one or more examples, the computer architecture 600 can analyze the corpus of information 620 to determine individuals having one or more phenotypes. In various examples, the computer architecture 600 can analyze the corpus of information 620 to determine one or more biomarkers that indicate a biological state. For example, the computer architecture 600 can analyze the corpus of information 620 to determine individuals having one or more genetic characteristics. One or more genetic characteristics may include at least one of one or more variants of a genomic region corresponding to a biological state. In one or more exemplary examples, one or more genetic characteristics may correspond to one or more variants of a genomic region corresponding to a certain type of cancer. In one or more additional exemplary examples, one or more biomarkers may correspond to levels of a sample outside a specified range. For example, the computer architecture 600 can analyze the information corpus 620 to determine individuals who have levels of one or more proteins and / or one or more small molecules that are indicators of a biological state. In these scenarios, the computer architecture 600 can analyze the results of clinical tests to determine the levels of an individual's samples. In one or more additional examples, the computer architecture 600 can analyze the information corpus 620 to determine individuals who have one or more symptoms that are indicators of a biological state. In one or more further examples, the computer architecture 600 can analyze imaging information contained in the information corpus 620 to determine individuals who have one or more biomarkers.

[0223] In one or more examples, the computer architecture 600 can implement one or more machine learning techniques to analyze the corpus of information 620. For example, the computer architecture 600 can implement one or more artificial neural networks, such as at least one of one or more convolutional neural networks or one or more residual neural networks, to analyze the corpus of information 620. The computer architecture 600 can also implement at least one of one or more random forest techniques, one or more hidden Markov models, or one or more support vector machines to analyze the corpus of information 620.

[0224] In at least some implementations, the computer architecture 600 can parse the corpus of information 620 by performing one or more queries against it. One or more queries may correspond to one or more keywords and / or combinations of keywords. One or more keywords and / or combinations of keywords may correspond to at least one of a set of letters or symbols corresponding to one or more biological conditions. For example, a keyword may correspond to a letter relating to a genomic region mutation, such as HER2. In one or more additional exemplary examples, one or more criteria may be associated with a combination of keywords. For example, a criterion corresponding to a combination of keywords may include a certain number of words that are within a specified distance from each other in a portion of the corpus of information 620 about an individual, such as the words fatigue, blood pressure, and swelling, which are within 100 characters of each other. In these cases, the computer architecture 600 can parse the corpus of information 620 for one or more keywords and / or combinations of keywords. In various examples, in response to a determination that one or more keywords and / or combinations of keywords exist according to one or more criteria, the computer architecture 600 can determine that a biological state exists for a given individual.

[0225] In one or more additional examples, one or more queries may be image-based, and the computer architecture 600 can analyze images contained in the corpus of information 620 against a template image. The template image can be generated by analyzing a certain number of images in which a biological state exists and aggregating that number of images into a template image. In these scenarios, the computer architecture 600 can analyze images contained in the corpus of information 620 against one or more template images to determine a measure of similarity between the images contained in the corpus of information 620 and the template image. If the measure of similarity for an individual is at least a threshold, the computer architecture 600 can determine that the features of the biological state are present in the individual.

[0226] After individuals possessing one or more characteristics have been determined, the computer architecture 600 may, in operation 624, generate a data structure to store data about individuals possessing one or more characteristics. In one or more examples, the computer architecture 600 may generate data tables showing individuals possessing individual characteristics and / or groups of characteristics. For example, the computer architecture 600 may generate a first data table 626 and a second data table 628. The first data table 626 may show individuals possessing one or more first characteristics, and the second data table 628 may show individuals possessing one or more second characteristics. In one or more exemplary examples, the first data table 626 may show individuals possessing one or more first biomarkers for a biological state, and the second data table 628 may show individuals possessing one or more second biomarkers for a biological state. One or more first biomarkers may correspond to one or more first genomic variants associated with a biological state, and one or more second biomarkers may correspond to one or more second genomic variants associated with a biological state. In various examples, data tables 626 and 628 may indicate whether one or more features associated with each data table 626 and 628 are present for each individual. For example, the first data table 626 may include a first indicator for individuals in which one or more first genomic variants are present and a second indicator for individuals in which one or more first genomic variants are not present. In one or more additional examples, the first data table 626 may indicate an individual's smoking status, and the second data table 628 may indicate whether an individual has received one or more treatments for a biological condition.

[0227] In one or more exemplary examples, the first data table 626 and the second data table 628 may have rows corresponding to individual individuals. In at least some examples, individual identifiers may reside in individual rows. Each identifier may contain at least one alphanumeric character or symbol corresponding to an individual. In various examples, individual identifiers may reside in data packages corresponding to individuals. Columns in the first data table 626 and the second data table 628 may indicate the status of individual individuals with respect to one or more features. For example, columns in data tables 626 and 628 may contain identifiers containing at least one alphanumeric character or symbol indicating the presence or absence of one or more features for a given individual. Furthermore, while the exemplary example in Figure 6 includes the first data table 626 and the second data table 628, the computer computing architecture 600 may generate more or fewer data tables.

[0228] In operation 630, the computer computing architecture 600 can store data structures in an additional data repository. For example, the computer computing architecture 600 can store at least a first data table 626 and / or a second data table 628 in the intermediate data repository 632. In various examples, the first data table 626 and the second data table 628 can be temporarily stored in the intermediate data repository 632. In one or more exemplary examples, the first data table 626 and the second data table 628 can be stored in the intermediate data repository 632 before being added to the integrated data repository 104. In one or more examples, the integrated data repository 104 can be periodically generated and / or updated. In these scenarios, the data structures generated by the computer computing architecture 600 based on the analysis of the corpus of information 620 can be stored in the intermediate data repository 632 until at least one of the generation or update of the integrated data repository 104 occurs.

[0229] Before adding the data structures stored in the intermediate data repository 632 to the integrated data repository 104, the computer architecture 600 may perform one or more deidentification processes in operation 634. The data structures stored in the intermediate data repository 632 may be deidentified to protect the privacy of individuals. One or more deidentification processes may include applying one or more electronically implemented cryptographic techniques to the information of individuals contained in the data structures stored in the intermediate data repository 632. In one or more examples, the computer architecture 600 may generate tokens corresponding to individual individuals whose information is stored in the data structures of the intermediate data repository 632. Tokens may be generated by applying one or more hash functions to the information of individual individuals. In one or more examples, one or more deidentification processes may include applying a salt function to the information corresponding to individual individuals to generate tokens for individual individuals. In various examples, one or more cryptographic techniques applied to deidentify the data structures stored in the intermediate data repository 632 may be the same as or similar to those applied to the information obtained from the medical insurance claims data repository 106 in Figures 1 and 4.

[0230] In operation 636, the computer architecture 600 can store an unidentified data structure together with the integrated data repository 104. For example, information stored in the intermediate data repository 632 about a given individual can be stored together with additional information about that individual in the integrated data repository 104. For example, the integrated data repository 104 can store information about a given individual obtained from at least two of the molecular data repository 108, the health insurance claims data repository 106, and the intermediate data repository 632. In this way, information about a given individual obtained from a number of completely different data repositories can be stored in the integrated data repository 104. As a result, information about an individual obtained from different data repositories can be analyzed together, rather than separately, as is done with many existing systems.

[0231] In various examples, the information stored in the intermediate data repository 632 can be used to verify one or more decisions made by the data integration and analysis system 102. For example, the data integration and analysis system 102 can analyze information obtained from the health insurance claims data repository 106 and the molecular data repository 108 to determine the characteristics of an individual. Then, the data integration and analysis system 102 can analyze information obtained from the intermediate data repository 632 to determine whether the predicted characteristics identified from the information obtained from the health insurance claims data repository 106 and the molecular data repository 108 correspond to the characteristics of the same individual with respect to the information stored in the intermediate data repository 632.

[0232] One or more cryptographic techniques applied to deidentify the data structures stored in the intermediate data repository 632 can utilize the same or similar information used to generate at least one of the first token 422 or the second token 428 in Figure 4. For example, operation 634 can deidentify the data structures in the intermediate data repository by implementing one or more cryptographic techniques using a combination of at least part of each individual's name, at least part of each individual's surname, at least part of each individual's date of birth, the individual's sex, and at least part of each individual's location identifier. By deidentifying the data structures stored in the intermediate data repository 632 using the same or similar cryptographic techniques and a subset of the same or similar information used to generate at least one of the first token 422 or the second token 428, the information stored in the intermediate data repository 632 can be synchronized with the information about the same individuals for which information is stored in the integrated data repository 104. Both the integrated data repository 104 and the intermediate data repository 632 can store information about thousands, tens of thousands, or even millions of individuals. Therefore, if the data of individuals whose records are stored in the integrated data repository 104 and the intermediate data repository 632 cannot be synchronized by using the specified cryptographic protocol described herein, the data structures of the integrated data repository 104 and the intermediate data repository 632 associated with the same individual cannot be stored in a manner that allows for the combined retrieval of information stored in the integrated data repository 104 and the intermediate data repository 632 for a given individual, which may lead to the data integration and analysis system 102 providing inaccurate information. The absence of the specified cryptographic protocol described herein may also lead to the use of more computing resources to determine the information stored in the integrated data repository 104 from information stored in other data sources and the intermediate data repository 632 corresponding to a given individual.Figures 7 and 8 illustrate an example process for generating an integrated data repository and a dataset used to analyze the information stored in the integrated data repository. The example process is presented as a set of blocks in a logical flow graph, representing a series of actions that can be implemented in hardware, software, or a combination thereof. Blocks are referenced by numbers. In a software context, a block represents a computer-executable instruction stored in one or more computer-readable media that, when executed by one or more processing units (e.g., hardware microprocessors), performs the described action. Generally, computer-executable instructions include routines, programs, objects, components, data structures, etc., that perform a specific function or implement a specific data type. The order in which the actions are described is not intended to be interpreted as limiting, and any number of described blocks can be combined in any order and / or in parallel to implement the process.

[0233] Figure 7 is a data flow diagram of an example process 700 for generating an integrated data repository in which health insurance claims data and genomics data are stored, according to one or more implementations. In operation 702, process 700 may include generating a data file containing tokens generated using a first hash function. Each token may correspond to each individual in a group of individuals whose data is stored in the molecular data repository. In one or more examples, individuals whose data is stored in the molecular data repository can be associated with one or more tokens. Tokens can be generated by applying one or more first hash functions to a subset of information corresponding to a group of individuals stored in the genomics data repository. In various examples, each token can be generated by applying one or more first hash functions to one or more combinations of at least part of the name of each individual in the group of individuals, at least part of the surname of each individual in the group of individuals, the location identifier of each individual in the group of individuals, the sex of each individual in the group of individuals, and the date of birth of each individual in the group of individuals. In one or more exemplary examples, tokens may be generated by a data integration and analysis system coupled with a genomics data repository. In one or more additional exemplary examples, tokens may be generated by a third-party system and accessed by a data integration and analysis system coupled with a molecular data repository. Process 700 may also include, in operation 704, sending a data file to a health insurance claims data management system. The health insurance claims data management system can access the tokens contained in the data file and match them with a second token generated based on information stored in the health insurance claims data repository.

[0234] Furthermore, in operation 706, process 700 may include obtaining first data corresponding to a group of individuals from a health insurance claims data management system in response to a data file, where the first data includes health insurance claims data. In some implementations, positive consent is obtained from members of the group of individuals to obtain data from the health insurance claims data management system. In one or more examples, the data is transferred in an anonymized format, and therefore the data cannot be traced back to individual members. The health insurance claims data management system can be coupled with a health insurance claims data repository that stores health insurance claims information for a certain number of individuals. In one or more examples, the health insurance claims data management system can parse tokens in a data file in relation to additional tokens generated by the health insurance claims data management system. The additional tokens can be generated based on the same set of information used to generate the tokens contained in the data file. However, the identity of individuals cannot be determined based on the tokens. In various examples, a health insurance claims data management system can determine which individuals have information stored in both the health insurance claims data repository and the genomics data repository by matching tokens contained in data files with additional tokens generated based on information stored in the health insurance claims data repository. The technologies disclosed herein comply with legal and best practice privacy standards, such as HIPAA and GDPR.

[0235] In operation 708, process 700 may include generating a number of identifiers using a second hash function different from a first hash function. In one or more examples, each identifier may correspond to one or more tokens relating to each individual in a group of individuals. Identifiers may be unique to a given individual in a group of individuals and may also be deidentifiable. Furthermore, identifiers may be generated using information stored in a genomics data repository different from the information stored in the genomics data repository used to generate the tokens for the group of individuals. In various examples, intermediate identifiers may be generated by applying the second hash function to the information for each individual in the group, and the final version of the identifier may be generated by applying one or more salting techniques to the intermediate identifier. The information stored in the genomics data repository for each individual may be stored in association with the identifier, and thus at least some of the information stored in the genomics data repository for a given individual may be accessed using the respective identifier of the given individual.

[0236] Furthermore, in operation 710, process 700 may include obtaining second data from a molecular data repository for a group of individuals using the identifier of the number, and in operation 712, process 700 may include determining, for a group of individuals, each part of the first data that corresponds to each part of the second data. For example, for a given individual, first data corresponding to health insurance claim data for the given individual can be identified in addition to second data corresponding to molecular data for the given individual, such as genomics data. In this way, both health insurance claim data and molecular data can be identified for a given individual.

[0237] Process 700 may include, in operation 714, generating an integrated data repository in association with each identifier of the number of identifiers, storing each part of the first data and each part of the second data. For example, the integrated data repository may store health insurance claim data and genomics claim data for a given individual, associated with identifiers that can be used to access the health insurance claim data and genomics claim data for a given individual. The information stored in the integrated data repository may be organized according to a data repository schema. For example, the integrated data repository may store health insurance claim data and genomics data for a group of individuals in a number of data tables. In one or more examples, the information stored in the number of data tables can be linked. For example, information about a given individual stored in a first data table of the data repository schema can be linked with additional information about a given individual stored in a second data table of the data repository schema. In this way, it is possible to access additional information stored in another data table of the data repository schema from information accessed in one data table of the data repository schema.

[0238] In one or more exemplary examples, the data repository schema may include a first data table that stores genomic data for a group of individuals. For example, the first data table may store genomic data, genomic region mutations, mutation types, copy numbers of genomic regions, coverage data indicating the number of nucleic acid molecules identified in a sample with one or more mutations, examination dates, and information corresponding to the panel used to generate patient information. The data repository schema may also include a second data table that stores data relating to one or more patient visits by an individual to one or more healthcare providers, and a third data table that stores information relating to each service provided to the individual for one or more patient visits to one or more healthcare providers as shown by the second data table. Furthermore, the data repository schema may include a fourth data table that stores personal information of a group of individuals, and a fifth data table that stores information relating to a health insurance company or government agency that made payments for services provided to the group of individuals. Furthermore, the data repository schema may include a sixth data table that stores health insurance coverage information for a group of individuals, for example, information corresponding to the type of health insurance plan for the group of individuals. The data repository schema may also include a seventh data table that stores information related to the medical treatments received by a group of individuals.

[0239] In one or more examples, the integrated data repository may also store medical records corresponding to at least a portion of a population of individuals. In these examples, the medical records may be obtained from one or more data repositories that store the medical records. One or more optical character recognition (OCR) operations can be performed on the medical records. Furthermore, the medical records can be analyzed to determine one or more portions of additional information to remove and to create a corpus of information. In various examples, the corpus of information can be analyzed to determine a portion of an additional population of individuals corresponding to one or more biomarkers.

[0240] One or more data structures can be generated from a corpus of information that stores identifiers for a subset of additional populations of individuals and an indicator that a subset of additional populations of individuals corresponds to one or more biomarkers. One or more data structures can be stored in an intermediate data repository. One or more deidentification operations can be performed on the identifiers of a subset of additional populations of individuals before modifying the integrated data repository to store at least a portion of the additional information of the medical records of a subset of additional populations of individuals in association with that number of identifiers. After the information stored in one or more data structures is deidentified, the information stored in the integrated data repository can be added to the integrated data repository. In at least some examples, deidentified medical record information can be added to the integrated data repository in addition to, or instead of, medical insurance claim data. In various examples, one or more data structures in which deidentified medical record information is stored in association with biomarker data may have one or more logical connections with other data structures stored in the integrated data repository.For example, one or more data structures in which anonymized medical record information is stored in association with biomarker data may have one or more logical connections with at least one of the following: a first data table that can store information corresponding to genomics data, genomic region mutations, mutation types, copy numbers of genomic regions, coverage data indicating the number of nucleic acid molecules identified in a sample having one or more mutations, examination date, and a panel used to generate patient information; a second data that stores data relating to one or more patient visits by an individual to one or more healthcare providers; a third data table that stores information corresponding to each service provided to the individual relating to one or more patient visits to one or more healthcare providers indicated by the second data table; a fourth data table that stores personal information of a group of individuals; a fifth data table that stores information relating to a health insurance company or government agency that made payments for services provided to the group of individuals; a sixth data table that stores information corresponding to health insurance coverage information for a group of individuals, such as the type of health insurance plan for a group of individuals, or a seventh data table that stores information relating to medical treatments received by a group of individuals.

[0241] In various examples, medical record data can be added to an integrated data repository by generating a data file containing a first token generated using a first hash function. Each first token may correspond to each individual in a group of individuals whose data is stored in the molecular data repository. Furthermore, the data file can be sent to a medical record data management system, and medical record data corresponding to the group of individuals can be obtained from the medical record data management system in response to the data file. In addition, a number of identifiers can be generated using a second hash function different from the first hash function. Each identifier may correspond to one or more tokens associated with each individual in the group of individuals. Using this number of identifiers, second data can be obtained from the molecular data repository for the group of individuals. In various examples, for a group of individuals, each part of the first data can be determined to correspond to each part of the second data. In this way, an integrated data repository can be generated in which each part of the first data and each part of the second data are stored in association with each identifier of the number of identifiers.

[0242] After generating an integrated data repository for storing medical record data, it is possible to receive requests to determine data for a certain number of individuals whose data is stored in the integrated data repository. These requests may include one or more search criteria. In one or more examples, a subset of that number of individuals can be determined that possesses one or more features corresponding to one or more search criteria. Information from this subset of individuals can then be analyzed to determine a measure of the significance of one or more features in relation to their biological state.

[0243] In one or more exemplary cases, it can be determined that one or more genomic mutations are present in a subset of that number of individuals, and that multiple treatments have been administered to that subset of individuals. In various cases, the survival rate for each of the subsets of individuals, e.g., the real-world survival rate, can be determined. In at least some cases, the measure of significance may correspond to the survival rate for one of the multiple treatments and one of the genomic mutations. Based on the measure of significance, the effectiveness of the treatment for the subset of individuals can be determined. In one or more cases, it can be determined that the untreated individuals within the subset of individuals can be identified. One or more therapeutically effective doses of treatment can be administered to the untreated individuals within the subset of individuals.

[0244] Figure 8 is a data flow diagram of an example process 800 for generating a number of datasets used to analyze information stored in an integrated data repository that stores health insurance claims data and genomics data, according to one or more implementations. Process 800 may include, in operation 802, determining a first set of data processing instructions that can be executed in relation to first data stored in the integrated data repository. The integrated data repository can store health insurance claims data and molecular data for a common population of individuals. In one or more examples, the first set of data processing instructions may be comprised of multiple sets of data processing instructions that are part of a data processing pipeline. Each set of data processing instructions in the data processing pipeline can be executed to generate its respective analysis-ready dataset. For example, individual sets of data processing instructions in the data processing pipeline may be executable to generate a dataset containing some of the information and / or combinations of information stored in the integrated data repository. In one or more additional examples, individual sets of data processing instructions in the data processing pipeline may be executable to analyze and modify some of the information stored in the integrated data repository to generate their respective datasets. Furthermore, individual sets of data processing instructions may be executable on subsets of individual pieces of information stored in the integrated data repository.

[0245] Process 800 may also include, in operation 804, executing a first set of data processing instructions to generate a first dataset. The first dataset may represent a subset of individuals in which a biological condition exists. The first set of data processing instructions can be executed to analyze data stored in the integrated data repository to identify a cohort of individuals in which a biological condition exists. In one or more exemplary examples, the biological condition may include cancer. For example, the first set of data processing instructions can be executed to analyze data stored in the integrated data repository to identify a cohort of individuals in which lung cancer exists. In various examples, the data processing pipeline may include multiple sets of data processing instructions to identify cohorts of individuals in which various biological conditions exist.

[0246] In one or more examples, a first set of data processing instructions can be executed to analyze at least one of either health insurance claims data or molecular data to determine a cohort of individuals in which a biological condition exists. For example, a first set of data processing instructions can be executed to identify individuals with one or more health insurance codes present in health insurance claims data to determine a group of individuals in which a biological condition exists. Furthermore, a first set of data processing instructions can be executed to identify individuals with one or more mutations in genomic regions of nucleic acid molecules derived from samples obtained from individuals to determine a group of individuals in which a biological condition exists.

[0247] Furthermore, in operation 806, process 800 may include determining a second set of data processing instructions that can be executed in relation to a second set of data stored in the integrated data repository. The second set of data stored in the integrated data repository may be different from the first set of data stored in the integrated data repository and can be parsed in relation to the first set of data processing instructions. For example, the first data may correspond to a first column of one or more first data tables stored in the integrated data repository, and the second data may correspond to a second column of one or more second data tables stored in the integrated data repository.

[0248] In operation 808, process 800 may include executing a second set of data processing instructions to generate a second dataset showing one or more treatments provided to a second subset of a population of individuals. The second dataset may show a subset of individuals that have received one or more treatments. One or more treatments may be provided to individuals in which one or more biological conditions exist. By analyzing data stored in the integrated data repository, a second set of data processing instructions may be executed in one or more examples to identify a cohort of individuals that have received one or more treatments. For example, a second set of data processing instructions may be executed to determine a cohort of individuals that have received one or more treatments by analyzing at least one of either health insurance claims data or genomics data. In one or more exemplary examples, a second set of data processing instructions may be executed to determine a population of individuals that have received one or more treatments by identifying individuals with one or more health insurance codes present in the health insurance claims data.

[0249] Furthermore, process 800 may include determining a third subset of the population of individuals in operation 810, which includes a portion of the first subset of the population of individuals that overlaps with a portion of the second subset of the population of individuals. As a result, the third subset of the population of individuals corresponds to individuals in which a biological state is present and one or more treatments are offered. In 812, process 800 may include analyzing the first and second datasets in relation to the third subset of the population of individuals to determine a measure of significance for the features of the third subset of the population of individuals. In one or more examples, one or more machine learning or statistical techniques may be applied to the information contained in at least one of the first and second datasets in relation to the third subset of the population of individuals. The measure of significance may correspond to a statistical measure of significance for the features. In one or more additional examples, the measure of significance may correspond to the probability of a feature being present in individuals in which a biological state is present.

[0250] In one or more exemplary examples, a feature may include one or more treatments provided to individuals with the biological condition. In one or more additional exemplary examples, a feature may include the presence of mutations in genomic regions of nucleic acid molecules derived from samples obtained from individuals with the biological condition. In various examples, the information contained in at least one of the first or second datasets can be analyzed to determine the effect of the feature with respect to one or more metrics. In one or more examples, the information contained in at least one of the first or second datasets can be analyzed to determine the degree to which the treatments have an effect on the survival rate of individuals with the biological condition. In one or more further examples, the information contained in at least one of the first or second datasets can be analyzed to determine the degree to which mutations in genomic regions have an effect on the survival rate of individuals with the biological condition. Furthermore, the information contained in the first and second datasets can be analyzed to determine the degree to which one or more treatments have an effect on individuals with the biological condition and also with one or more genomic mutations.

[0251] Figure 9 shows a schematic representation of machine 9900 in the form of a computer system that can execute a set of instructions causing machine 900 to perform one or more of the methodologies discussed herein, according to one example. Specifically, Figure 8 shows a schematic representation of machine 900 in the form of an example of a computer system that can execute instructions 902 (e.g., software, programs, applications, applets, apps, or other executable code) causing machine 900 to perform one or more of the methodologies discussed herein. For example, instructions 902 can cause machine 900 to implement the architectures and frameworks 100, 200, 300, 400, 500, and 600 described with respect to Figures 1, 2, 3, 4, 5, and 6, respectively, and to execute the methods 700 and 800 described with respect to Figures 7 and 8, respectively.

[0252] Instruction 902 translates a general, unprogrammed machine 900 into a specific machine 900 programmed to perform the functions described and illustrated in the described manner. In alternative implementations, machine 900 may operate as a standalone device or be coupled with other machines (e.g., network-connected). In network-connected deployments, machine 900 may operate as a server machine or client machine in a server-client network environment, or as a peer machine in a peer-to-peer (or distributed) network environment. Machine 900 may include, but is not limited to, server computers, client computers, personal computers (PCs), tablet computers, laptop computers, netbooks, set-top boxes (STBs), personal digital assistants (PDAs), entertainment media systems, mobile phones, smartphones, mobile devices, wearable devices (e.g., smartwatches), smart home devices (e.g., smart appliances), other smart devices, web appliances, network routers, network switches, network bridges, or any machine capable of sequentially or otherwise executing Instruction 902, which specifies the actions performed by machine 900. Furthermore, although only a single machine 900 is illustrated, the term “machine” shall be interpreted to also include a set of machines 900 that individually or collectively carry out Instruction 902 and implement one or more of the methodologies considered herein.

[0253] Examples of computer computing devices 900 may include logic, one or more components, circuits (e.g., modules), or mechanisms. A circuit is a tangible entity configured to perform a particular operation. In one example, the circuit may be arranged in a specified manner (e.g., internally or to external entities such as other circuits). In one example, one or more computer systems (e.g., standalone, client, or server computer systems) or one or more hardware processors (processors) may be configured by software (e.g., instructions, application portions, or applications) to operate as a circuit that performs a particular operation described herein. In one example, the software may reside (1) on a non-transient machine-readable medium, or (2) in a transmitted signal. In one example, the software, when executed by the hardware underlying the circuit, causes the circuit to perform a particular operation.

[0254] For example, a circuit may be implemented mechanically or electronically. For instance, a circuit may include a dedicated network or logic specifically configured to perform one or more techniques, such as a dedicated processor, a field-programmable gate array (FPGA), or an application-specific integrated circuit (ASIC). For example, a circuit may include programmable logic (e.g., a network contained within a general-purpose processor or other programmable processor) that can be temporarily configured (e.g., by software) to perform a particular operation. It will be understood that the decision of whether to implement a circuit mechanically (e.g., with a dedicated and permanently configured network) or with a temporarily configured network (e.g., configured by software) may be driven by cost and time considerations.

[0255] Therefore, the term “circuit” is understood to encompass tangible entities that are physically constructed, permanently configured (e.g., hardwired), or temporarily configured (e.g., transiently) (e.g., programmed) to operate in a specified manner or perform a specified operation. For example, assuming multiple temporarily configured circuits, each circuit does not need to be configured or instantiated at any given time. For example, if a circuit includes a general-purpose processor configured by software, the general-purpose processor may be configured as each of several different circuits at different times. Thus, software may configure the processor, for example, to constitute a particular circuit at one time and different circuits at different times.

[0256] In one example, a circuit can provide information to and receive information from other circuits. In this example, a circuit can be thought of as being communicatively coupled to one or more other circuits. If multiple such circuits exist simultaneously, communication can be achieved by signal transmission connecting the circuits (e.g., through appropriate circuits and buses). In an implementation where multiple circuits are configured or instantiated at different points in time, communication between such circuits can be achieved, for example, by storing and retrieving information in a memory structure accessed by multiple circuits. For example, one circuit may perform an operation and store the output of that operation in a memory device with which it is communicatively coupled. Another circuit may then access the memory device at a later point in time to retrieve and process the stored output. In one example, a circuit may be configured to initiate or receive communication using an input or output device and may operate on a resource (e.g., a collection of information).

[0257] Various operations of the examples of the methods described herein can be performed, at least in part, by one or more processors that are temporarily (e.g., by software) or permanently configured to perform the relevant operations. Such processors, whether temporarily or permanently configured, may constitute a processor-implemented circuit that operates to perform one or more operations or functions. In one example, the circuit referred to herein may include a processor-implemented circuit.

[0258] Similarly, the methods described herein can be implemented at least partially by processors. For example, at least some or all of the operation of the method can be carried out by one or more processors or circuits implemented by processors. The implementation of a particular operation may reside not only within a single machine but also distributed among one or more processors deployed across several machines. In one example, the processor(s) may be located in a single location (e.g., in a home environment, an office environment, or a server farm), while in another example, the processors may be distributed across several locations.

[0259] One or more processors may also operate to support the execution of related operations in a “cloud computing” environment or as “software as a service” (SaaS).

[0260] For example, at least part of the operations can be performed by a group of computers (for example, a machine including a processor), and these operations can be accessed via a network (e.g., the Internet) and via one or more suitable interfaces (e.g., application programming interfaces (APIs)).

[0261] Examples of implementation (e.g., devices, systems, or methods) may be implemented in digital electronic networks, computer hardware, firmware, software, or any combination thereof. Examples of implementation may be implemented using computer program products (e.g., computer programs explicitly embodied in information carriers or machine-readable media for execution by or control of the operation of data processing devices such as programmable processors, computers, or multiple computers).

[0262] Computer programs may be written in any form of programming language, including compiled or interpreted languages, and may be deployed in any form, including as standalone programs or as software modules, subroutines, or other units suitable for use in a computer computing environment. Computer programs may be deployed to run on one computer, on multiple computers in one location, or distributed across multiple locations and interconnected by a communication network.

[0263] In one example, the operation can be carried out by one or more programmable processors that execute a computer program to perform a function by acting on input data and producing an output. An example of the operation of the method may also be carried out by a special-purpose logic network (e.g., a field-programmable gate array (FPGA) or an application-specific integrated circuit (ASIC)), and an example of the device may be implemented as a special-purpose logic network (e.g., a field-programmable gate array (FPGA) or an application-specific integrated circuit (ASIC)).

[0264] A computer computing system may include clients and servers. Clients and servers are generally geographically separated and typically interact through a communication network. The client-server relationship arises from computer programs running on each computer that have a client-server relationship with each other. It will be understood that in an implementation where a programmable computer computing system is deployed, both hardware and software architectures must be considered. Specifically, it will be understood that the choice of whether to implement a particular functionality in permanently configured hardware (e.g., ASICs), temporarily configured hardware (e.g., a combination of software and a programmable processor), or a combination of permanently and temporarily configured hardware, can be a design choice. The following describes the hardware (e.g., computer computing device 900) and software architectures that may be deployed in an example implementation.

[0265] For example, the computer computing device 900 may operate as a standalone device, or it may be connected to other machines (e.g., via a network connection).

[0266] In a networked deployment, the computing device 900 may operate as either a server or a client machine in a server-client network environment. For example, the computing device 900 may function as a peer machine in a peer-to-peer (or other distributed) network environment. The computing device 900 may be a personal computer (PC), a tablet PC, a set-top box (STB), a mobile phone, a web appliance, a network router, a switch or bridge, or any machine (sequentially or otherwise) capable of executing instructions that specify the actions performed (e.g., carried out) by the computing device 900. Furthermore, although only a single computing device 900 is illustrated, the term “computing device” shall also be interpreted as including any set of machines that individually or collectively execute a set (or set) of instructions to perform one or more of the methodologies considered herein.

[0267] An example of the computer computing device 900 may include a processor 904 (e.g., a central processing unit CPU), a graphics processing unit (GPU) or both), main memory 906, and static memory 908, some or all of which may communicate with each other via a bus 910. The computer computing device 900 may further include a display unit 912, an alphanumeric input device 914 (e.g., a keyboard), and a user interface (UI) navigation device 916 (e.g., a mouse). In one example, the display unit 912, the input device 914, and the UI navigation device 916 may be touchscreen displays. The computer computing device 900 may further include a storage device (e.g., a drive unit) 918, a signal generating device 920 (e.g., a speaker), a network interface device 922, and one or more sensors 924, a Global Positioning System (GPS) sensor, e.g., a compass, an accelerometer, or another sensor.

[0268] The storage device 918 may include a machine-readable medium 926 on which one or more sets of data structures or instructions 902 (e.g., software) that embody or utilize any one or more of the methods or functions described herein are stored. The instructions 902 may also reside, all or at least partially, in the main memory 906, static memory 908, or processor 904 during their execution by the computer computing device 900. In one example, one or any combination of the processor 904, main memory 906, static memory 908, or storage device 918 may constitute the machine-readable medium.

[0269] While machine-readable medium 926 is exemplified as a single medium, the term “machine-readable medium” may include a single or multiple mediums configured to store one or more instructions 902 (e.g., a centralized or distributed database, and / or associated caches and servers). The term “machine-readable medium” may also be interpreted to include any tangible medium capable of storing, encoding, or carrying instructions executed by a machine, causing a machine to implement one or more of the methods of this disclosure, or storing, encoding, or carrying data structures utilized by or associated with such instructions. Thus, the term “machine-readable medium” may be interpreted to include, but is not limited to, solid-state memory, as well as optical and magnetic media. Specific examples of machine-readable mediums include, for example, semiconductor memory devices (e.g., electrically programmable read-only memory).

[0270] Examples include (EPROM), electrically erasable programmable read-only memory (EEPROM), and flash memory devices; magnetic disks such as internal hard disks and removable disks; magneto-optical disks; and non-volatile memory including CD-ROM and DVD-ROM disks.

[0271] Instruction 902 may be further transmitted or received by a communication network 828 using a transmission medium via a network interface device 822 that utilizes one of several transport protocols (e.g., Frame Relay, IP, TCP, UDP, HTTP, etc.). Examples of communication networks may include, among others, a local area network (LAN), a wide area network (WAN), a packet data network (e.g., the Internet), a mobile phone network (e.g., a cellular network), a Plain Old Telephone (POTS) network, and a wireless data network (e.g., the IEEE 802.11 series of standards known as Wi-Fi®, the IEEE 802.16 series of standards known as WiMax®), and a peer-to-peer (P2P) network. The term “transmission medium” shall be interpreted as including any intangible medium capable of storing, encoding, or carrying instructions executed by a machine, and containing digital or analog communication signals, or other intangible medium for facilitating the communication of such software.

[0272] As used herein, a component may mean a device, physical entity, or logic having boundaries defined by function or subroutine calls, branching points, APIs, or other techniques that provide compartmentalization or modularization of a particular processing or control function. A component may be combined with other components via an interface to perform machine processing. A component may be a packaged functional hardware unit designed to be used with other components, or it may be part of a program that typically performs a particular function of the associated function. A component may constitute a software component (e.g., code materialized on a machine-readable medium) or a hardware component. A “hardware component” is a tangible unit capable of performing a particular operation and may be configured or arranged in a particular physical form. In various implementation examples, one or more computer systems (e.g., a standalone computer system, a client computer system, or a server computer system) or one or more hardware components of a computer system (e.g., a processor or a group of processors) may be configured by software (e.g., an application or application portion) as a hardware component that operates to perform a particular operation described herein.

[0273] Cancer and other diseases This method can be used to diagnose a condition in a subject, particularly the presence of cancer; to characterize the condition (e.g., to stage cancer or determine cancer heterogeneity); to monitor the response to treatment of the condition; and to determine the risk of the condition developing or the prognosis of the subsequent course of the condition. This disclosure may also be useful in determining the efficacy of a particular treatment option. A successful treatment option may increase the amount of copy number variation or rare mutations detected in the subject's blood, as more cancers may be killed and DNA may be shed if the treatment is successful. In other cases, this may not occur. In another case, a particular treatment option may correlate over time with the genetic profile of the cancer. This correlation may be useful in selecting a treatment.

[0274] Furthermore, if the cancer is observed to be in remission after treatment, this method can be used to monitor residual disease or disease recurrence.

[0275] In some embodiments, the methods and systems disclosed herein can be used to identify customized or targeted therapies to treat a given disease or condition in a patient, based on the classification of nucleic acid variants as being of somatic or germline origin. Typically, the disease under consideration is one type of cancer. Non-limiting examples of such cancers include biliary tract cancer, bladder cancer, transitional cell carcinoma, urothelial carcinoma, brain cancer, glioma, astrocytoma, breast cancer, and metaplastic carcinoma. Carcinoma, cervical cancer, cervical squamous cell carcinoma, rectal cancer, colorectal cancer, colon cancer, hereditary nonpolyposis colorectal cancer, colorectal adenocarcinoma, gastrointestinal stromal tumor (GIST), endometrial cancer, endometrial stromal sarcoma, esophageal cancer, esophageal squamous cell carcinoma, esophageal adenocarcinoma, intraocular melanoma, uveal melanoma, gallbladder cancer, gallbladder adenocarcinoma, renal cell carcinoma, renal clear cell carcinoma, transitional cell carcinoma, urothelial carcinoma, Wilms' tumor, leukemia, acute lymphoblastic leukemia (ALL), acute myeloid leukemia (AML), chronic lymphoblastic leukemia (CLL), chronic myeloid leukemia (CML), chronic myelomonocyte Leukemia malformation (CMML), liver cancer, hepatoma, hepatocellular carcinoma, cholangiocarcinoma, hepatoblastoma, lung cancer, non-small cell lung cancer (NSCLC), mesothelioma, B-cell lymphoma, non-Hodgkin lymphoma, diffuse large B-cell lymphoma, mantle cell lymphoma, T-cell lymphoma, non-Hodgkin lymphoma, progenitor T-lymphoblastic lymphoma / leukemia, peripheral T-cell lymphoma, multiple myeloma, nasopharyngeal carcinoma (NPC), neuroblastoma, oropharyngeal cancer, oral squamous cell carcinoma, osteosarcoma, ovarian cancer, pancreatic cancer, pancreatic ductal adenocarcinoma, pseudopapillary neoplasm Examples include neoplasms, acinar cell carcinoma, prostate cancer, prostate adenocarcinoma, skin cancer, melanoma, malignant melanoma, cutaneous melanoma, small intestine cancer, stomach cancer, gastrointestinal stromal tumors (GIST), uterine cancer, or uterine sarcoma.Cancer type and / or stage can be detected from genetic variations, including mutations, rare mutations, indels, copy number variations, base transpositions, translocations, inversions, deletions, aneuploidy, partial aneuploidy, polyploidy, chromosomal instability, chromosomal structural changes, gene fusions, chromosome fusions, gene shortening, gene amplification, gene duplication, chromosomal damage, DNA damage, abnormal changes in nucleic acid chemical modifications, abnormal changes in epigenetic patterns, and abnormal changes in nucleic acid 5-methylcytosine.

[0276] Genetic data can also be used to characterize specific forms of cancer. Cancers are often heterogeneous in both composition and staging. Genetic profiling data may enable the characterization of specific subtypes of cancer, which may be important in the diagnosis or treatment of that particular subtype. This information may also provide subjects or practitioners with clues regarding prognosis for specific types of cancer, enabling either the subject or practitioner to adapt treatment options as the disease progresses. Some cancers may progress to become more invasive and genetically unstable. Other cancers may remain benign, inactive, or quiescent. The systems and methods of this disclosure may be useful in determining disease progression.

[0277] Furthermore, the methods of this disclosure can be used to characterize heterogeneity of an abnormal condition in a subject. Such a method may include, for example, a step of generating a gene profile of extracellular polynucleotides derived from the subject, wherein the gene profile includes multiple data obtained from the analysis of copy number variations and rare mutations. In some embodiments, the abnormal condition is cancer. In some embodiments, the abnormal condition may result in a heterogeneous genomic population. In the example of cancer, it is known that several tumors may contain tumor cells at different stages of cancer. In other examples, heterogeneity may include multiple lesions of the disease. Again, in the example of cancer, multiple tumor lesions may be present, and perhaps one or more lesions are the result of metastases that have spread from the primary site.

[0278] This method can be used to generate or profile a fingerprint or set of data that is a summary of genetic information from different cells in a heterogeneous disease. This set of data can include copy number variations, epigenetic variations, and analysis of mutations, either alone or in combination.

[0279] This method can be used to diagnose, prognosticate, monitor or observe cancer or other diseases. In some embodiments, the methods herein do not involve diagnosing, prognosticating or monitoring a fetus, and thus are not intended for non-invasive prenatal testing. In other embodiments, these methods can be used to diagnose, prognosticate, monitor or observe cancer or other diseases in a non-born subject in which its DNA and other polynucleotides can co-circulate with maternal molecules during pregnancy in a subject.

[0280] Non-limiting examples of other gene-based diseases, disorders, or conditions that may be evaluated using the methods and systems disclosed herein as needed include achondroplasia, alpha1-antitrypsin deficiency, antiphospholipid syndrome, autism, autosomal dominant polycystic kidney disease, Charcot-Marie-Tooth (CMT), cat cry syndrome, Crohn's disease, cystic fibrosis, Duchenne muscular dystrophy, factor V Leiden thrombophilia, familial hypercholesterolemia, familial Mediterranean fever, fragile X syndrome, Gaucher disease, hemochromatosis, hemophilia, holoprosencephaly, Huntington disease, Klinefelter syndrome, Marfan syndrome, myotonic dystrophy, neurofibromatosis, Noonan syndrome, osteogenesis imperfecta, Parkinson's disease, phenylketonuria, Poland anomaly, porphyria, progeria, retinitis pigmentosa, severe combined immunodeficiency (SCID), sickle cell disease, spinal muscular atrophy, Tay-Sachs, thalassemia, trimethylaminuria, Turner syndrome, velocardiofacial syndrome, WAGR syndrome, Wilson's disease, etc.

[0281] In some embodiments, the method described herein includes detecting the presence or absence of DNA originating from or derived from tumor cells at a preselected time point after a previous cancer treatment of a subject previously diagnosed with cancer, using a set of sequence information obtained as described herein. The method may further include determining, for the test subject, a cancer recurrence score that is an indicator of the presence or absence of DNA originating from or derived from tumor cells. Once the cancer recurrence score is determined, the score can be further used to determine the status of cancer recurrence. The status of cancer recurrence can be, for example, a state with a risk of cancer recurrence if the cancer recurrence score exceeds a predetermined threshold. The status of cancer recurrence can be, for example, a state with a low or lower risk of cancer recurrence if the cancer recurrence score exceeds a predetermined threshold. In certain embodiments, if the cancer recurrence score is equal to a predetermined threshold, a status of cancer recurrence with a risk of cancer recurrence or a low or lower risk of cancer recurrence can result.

[0282] In some embodiments, the cancer recurrence score is compared to a predetermined cancer recurrence threshold, and the test subject is classified as a candidate for subsequent cancer treatment if the cancer recurrence score exceeds the cancer recurrence threshold, or not a candidate for treatment if the cancer recurrence score is below the cancer recurrence threshold. In certain embodiments, if the cancer recurrence score is equal to the cancer recurrence threshold, a classification of either a candidate for subsequent cancer treatment or not a candidate for treatment can result.

[0283] The above method may further include any compatible feature(s) described elsewhere in this specification, including sections related to a method for determining the risk of cancer recurrence in a test subject and / or classifying the test subject as a candidate for subsequent cancer treatment.

[0284] A method for determining the risk of cancer recurrence in a test subject and / or classifying the test subject as a candidate for subsequent cancer treatment. In some embodiments, the methods provided herein are for determining the risk of cancer recurrence in a test subject. In some embodiments, the methods provided herein are for classifying a test subject as a candidate for subsequent cancer treatment.

[0285] Any such method may include the step of collecting DNA (e.g., originating from or derived from tumor cells) from a test subject diagnosed with cancer at one or more pre-selected time points after one or more previous cancer treatments to the test subject. The subject may be any of the subjects described herein. The DNA may be cfDNA. The DNA may be obtained from a tissue sample.

[0286] Any such method may include a step of capturing multiple sets of target regions from DNA derived from a subject, wherein the multiple sets of target regions include a sequence-variable target region set and an epigenetic target region set, thereby resulting in a set of captured DNA molecules. The capture step can be carried out according to any of the embodiments described elsewhere in this specification. In any such method, prior cancer treatment may include surgery, administration of a therapeutic composition, and / or chemotherapy.

[0287] Any such method may involve the step of sequencing the captured DNA molecule, thereby generating a set of sequence information. Captured DNA molecules of a sequence-variable target region set can be sequenced to a higher sequencing depth than captured DNA molecules of an epigenetic target region set.

[0288] Any such method may include the step of detecting the presence or absence of DNA originating from or derived from tumor cells at a pre-selected point in time using a set of sequence information. The detection of the presence or absence of DNA originating from or derived from tumor cells can be carried out according to any of the embodiments described elsewhere in this specification.

[0289] A method for determining the risk of cancer recurrence in a test subject may include the step of determining a cancer recurrence score for the test subject, which is an indicator of the presence or absence or amount of DNA originating from or derived from tumor cells. The cancer recurrence score can be further used to determine the status of cancer recurrence. The status of cancer recurrence may be, for example, a state of being at risk of cancer recurrence if the cancer recurrence score is above a predetermined threshold. The status of cancer recurrence may be, for example, a state of being at risk of cancer recurrence or at low or low risk of cancer recurrence if the cancer recurrence score is above a predetermined threshold. In certain embodiments, a cancer recurrence status of being at risk of cancer recurrence or at low or low risk of cancer recurrence may result if the cancer recurrence score is equal to a predetermined threshold.

[0290] A method for classifying a test subject as a candidate for subsequent cancer treatment may include the step of comparing the test subject's cancer recurrence score to a predetermined cancer recurrence threshold, thereby classifying the test subject as a candidate for subsequent cancer treatment if the cancer recurrence score is above the cancer recurrence threshold, or as not a candidate for treatment if the cancer recurrence score is below the cancer recurrence threshold. In certain embodiments, if the cancer recurrence score is equal to the cancer recurrence threshold, the classification may be either candidate for subsequent cancer treatment or not a candidate for treatment. In some embodiments, the subsequent cancer treatment includes chemotherapy or administration of a therapeutic composition.

[0291] Any such method may include a step of determining the disease-free survival (DFS) period for the study subjects based on a cancer recurrence score. For example, the DFS period could be 1 year, 2 years, 3 years, 4 years, 5 years, or 10 years.

[0292] In some embodiments, the set of sequence information includes a sequence variable target region sequence, and the step of determining the cancer recurrence score may include determining at least a first subscore that is an indicator of the amount of SNVs, insertions / deletions, CNVs and / or fusions present within the sequence variable target region sequence.

[0293] In some embodiments, the number of mutations in the sequence variable target region, selected from one, two, three, four, or five, is sufficient to produce a cancer recurrence score that classifies a patient as positive for cancer recurrence, based on a first subscore. In some embodiments, the number of mutations is selected from one, two, or three.

[0294] In some embodiments, the set of sequence information includes epigenetic target region sequences, and the step of determining the cancer recurrence score includes determining a second subscore that is an indicator of the amount of molecules (derived from the epigenetic target region sequences) that represent an epigenetic state different from the DNA found in a corresponding sample from a healthy subject (e.g., cfDNA found in a blood sample from a healthy subject, or DNA found in a tissue sample from a healthy subject; where the tissue sample is of the same type of tissue as that obtained from the test subject). These abnormal molecules (i.e., molecules having an epigenetic state different from the DNA found in a corresponding sample from a healthy subject) may correspond to epigenetic changes associated with cancer, e.g., methylation of a hypermethylated variable target region and / or perturbed fragmentation of a fragmented variable target region, where “perturbed” means different from the DNA found in a corresponding sample from a healthy subject.

[0295] In some embodiments, a second subscore being classified as positive for cancer recurrence is sufficient if the proportion of molecules corresponding to the hypermethylated variable target region set and / or fragmented variable target region set exhibiting hypermethylation within the hypermethylated variable target region set and / or abnormal fragmentation within the fragmented variable target region set is greater than or equal to a value in the range of 0.001% to 10%. The range may be 0.001% to 1%, 0.005% to 1%, 0.01% to 5%, 0.01% to 2%, or 0.01% to 1%.

[0296] In some embodiments, any such method may include the step of determining, from a set of molecular fractions of sequence information, the fraction of tumor DNA exhibiting one or more features that indicate its origin in tumor cells. This can be done, for example, for molecules corresponding to some or all of an epigenetic target region, including one or both of a hypermethylated variable target region and a fragmentation variable target region (hypermethylation of a hypermethylated variable target region and / or abnormal fragmentation of a fragmentation variable target region can be considered indicators of tumor cell origin). This can be done for molecules corresponding to sequence variable target regions, e.g., molecules containing modifications consistent with cancer, e.g., SNVs, indels, CNVs, and / or fusions. Based on combinations of molecules corresponding to epigenetic target regions and molecules corresponding to sequence variable target regions, the fraction of tumor DNA can be determined.

[0297] The determination of the cancer recurrence score may, at least in part, be based on the fraction of tumor DNA, where a fraction of tumor DNA greater than a threshold in the range of 10⁻¹¹ to 1 or 10⁻¹⁰ to 1 is sufficient for the cancer recurrence score to be classified as positive for cancer recurrence. In some embodiments, a fraction of tumor DNA greater than or equal to a threshold in the range of 10⁻¹⁰ to 10⁻⁹, 10⁻⁹ to 10⁻⁸, 10⁻⁸ to 10⁻⁷, 10⁷ to 10⁻⁶, 10⁶ to 10⁻⁵, 10⁵ to 10⁻⁴, 10⁴ to 10⁻³, 10⁻³ to 10⁻², or 10⁻² to 10⁻¹ is sufficient for the cancer recurrence score to be classified as positive for cancer recurrence. In some embodiments, a fraction of tumor DNA greater than at least a threshold of 10⁻⁷ is sufficient for the cancer recurrence score to be classified as positive for cancer recurrence. The determination that the tumor DNA fraction is greater than a threshold, for example, the corresponding threshold in any of the embodiments described above, can be made based on cumulative probability. For example, a sample was considered positive if the cumulative probability that the tumor fraction is greater than any of the thresholds in the range described above exceeds a probability threshold of at least 0.5, 0.75, 0.9, 0.95, 0.98, 0.99, 0.995, or 0.999. In some embodiments, the probability threshold is at least 0.95, for example, 0.99.

[0298] In some embodiments, the set of sequence information includes a sequence variable target region sequence and an epigenetic target region sequence, and the step of determining the cancer recurrence score includes determining a first subscore that indicates the amount of SNVs, insertions / deletions, CNVs and / or fusions present in the sequence variable target region sequence and a second subscore that indicates the amount of abnormal molecules in the epigenetic target region sequence, and combining the first and second subscores to obtain the cancer recurrence score. When combining the first and second subscores, they can be combined by independently applying thresholds to each subscore (e.g., more than a predetermined number of mutations in the sequence variable target region (e.g., >1) and higher than a predetermined fraction of abnormal molecules (i.e., molecules having a different epigenetic state than the DNA found in the corresponding sample from a healthy subject; e.g., tumors) in the epigenetic target region), or by training a machine learning classifier to determine the status based on multiple positive and negative training samples.

[0299] In some embodiments, a combined score value in the range of -4 to 2 or -3 to 1 is sufficient for the cancer recurrence score to be classified as ...

Claims

1. A step to determine the state of biomolecules obtained from a sample derived from a human subject, Steps for examining minimal residual disease (MRD), The steps include determining the likelihood of recurrence based on the aforementioned MRD examination, Based on the determination of the likelihood of recurrence, the steps include creating a schedule for one or more additional MRD tests. Methods that include...

2. The method according to claim 1, wherein the biomolecule is one or more of DNA, methylated DNA, RNA, methylated RNA, protein, and peptide.

3. The steps for testing for MRD are: The process involves preparing a nucleic acid-MBD protein solution by combining multiple nucleic acid molecules derived from the target with a solution containing a certain amount of methyl-binding domain (MBD) protein, and A step of preparing a certain number of nucleic acid fractions by performing multiple washes of the nucleic acid-MBD protein solution with a salt solution, wherein each nucleic acid fraction has a threshold number of methylated cytosines within the region of the plurality of nucleic acids having at least a threshold cytosine-guanine content. The method according to claim 1, including the method described in claim 1.

4. The method according to claim 3, wherein one of the aforementioned washes is performed using a solution containing sodium chloride (NaCl) at a certain concentration, and a nucleic acid fraction from the aforementioned number of nucleic acid fractions having a certain range of binding strength to the MBD protein is produced.

5. A step of determining that a first nucleic acid fraction is associated with a first segment of a plurality of segments of nucleic acid, wherein the first segment corresponds to a first range of binding strength to MBD protein, A step of attaching a first molecular barcode to the nucleic acid of the first nucleic acid fraction, wherein the first molecular barcode is included in a first set of molecular barcodes associated with the first division. A step of determining that a second nucleic acid fraction is associated with a second segment of the plurality of segments of nucleic acid, wherein the second segment corresponds to a second range of binding energy to the MBD protein, which is different from the first range of binding strength to the MBD protein. A step of attaching a second molecular barcode to the nucleic acid of the second nucleic acid fraction, wherein the second molecular barcode is included in a second set of molecular barcodes associated with the second portion. The method according to claim 3, including the method described in claim 3.

6. Step 1: Combine at least a portion of the aforementioned number of nucleic acid fractions with a certain amount of restriction enzyme that cleaves one or more molecules having unmethylated cytosine to prepare at least a portion of a plurality of samples used to generate sequencing reads. Includes, The method according to claim 3, wherein the threshold amount of methylated cytosine corresponds to the minimum frequency of methylated cytosine within the region having at least the threshold cytosine-guanine content.

7. A step of preparing at least a portion of multiple samples used to generate sequencing reads by combining at least a portion of the aforementioned number of nucleic acid fractions with a certain amount of restriction enzyme that cleaves molecules having one or more methylated cytosines. Includes, The method according to claim 3, wherein the threshold amount of unmethylated cytosine corresponds to the maximum frequency of uncleaved methylated cytosine within the region having at least the threshold cytosine-guanine content.

8. The steps for testing for MRD are: The steps include: determining the sequence of nucleic acid molecules derived from a sample obtained from the target, The steps include: analyzing sequence reads derived from the sequencing step to identify one or more driver mutations in the nucleic acid molecule; A step of identifying a tumor in the subject using information regarding the presence, absence, or amount of one or more driver mutations in the nucleic acid molecule. The method according to claim 1, including the method described in claim 1.

9. The method according to claims 3 to 8, wherein the nucleic acid molecule includes cell-free DNA.

10. The method according to any of the preceding claims, wherein the sample is at least one of blood, serum, plasma, or tissue.

11. The method according to any of the preceding claims, including a determination of a course of action for the subject.

12. The method according to any of the preceding claims, wherein the detection limit of the model for determining the tumor fraction of a sample is 0.05% or less.

13. The method according to any one of the preceding claims, wherein the one or more driver mutations include a somatic variant detected at a mutational allele frequency (MAF) of 0.05% or less.

14. The method according to any one of the preceding claims, wherein the one or more driver mutations include a fusion detected at a mutational allele frequency (MAF) of 0.1% or less.

15. The method according to any one of the preceding claims, further comprising the step of detecting the mutation distribution of one or more driver mutations, wherein the mutation distribution of one or more driver mutations is detected with a correlation of at least 0.99 to the mutation distribution of the driver mutations detected in the target cohort by tissue genotyping.

16. The method according to any one of the prior claims, for detecting the tumor in the subject with at least 85% sensitivity, at least 99% specificity, and at least 99% diagnostic accuracy.

17. The method according to any of the preceding claims, comprising the step of identifying circulating tumor DNA (ctDNA) and one or more driver mutations in the ctDNA.

18. A computer computing system having one or more hardware processors and memory obtains test sequence data from a target, wherein the test sequence data includes test sequence determination reads derived from the sample of the target. The steps include: analyzing the test sequencing reads using the aforementioned computer calculation system to determine a first quantitative measure derived from the test sequencing reads for a genomic region of a reference genome; The steps include: analyzing the test sequencing reads using the aforementioned computer calculation system to determine a second quantitative measure derived from the test sequencing reads for a genomic region of the reference genome; The steps include determining a measurement standard based on the first quantitative scale and the second quantitative scale using the aforementioned computer calculation system, The computer calculation system generates an input vector including the measurement criterion, The steps of determining an indicator of cancer status in a subject by providing the input vector to a model that implements one or more machine learning techniques using the computer computing system to generate an indicator of cancer status in the subject, wherein the model includes weights for individual classification regions of a plurality of classification regions, and at least a portion of the weights for the individual classification regions are different from each other. Methods that include...

19. Each of the aforementioned sequencing reads for testing includes a nucleotide sequence corresponding to a nucleic acid fragment contained in the sample, and each of the aforementioned sequencing reads for testing corresponds to a molecule having a threshold amount of methylated cytosine contained within a region of the nucleotide sequence having at least the threshold cytosine-guanine content. The first quantitative measure is derived from the test sequencing reads corresponding to individual classification regions of a plurality of classification regions, and at least a portion of the individual classification regions of the plurality of classification regions corresponds to a genomic region of a reference genome having the threshold amount of methylated cytosine and at least the threshold cytosine-guanine content in the subject in which cancer is detected. The method according to claim 18, wherein the second quantitative measure is derived from the test sequencing reads corresponding to individual control regions of a plurality of control regions, and each individual control region of the plurality of control regions corresponds to an additional genomic region of the reference genome having at least the threshold cytosine-guanine content and having at least the threshold amount of methylated cytosine in subjects in which cancer is detected and additional subjects in which cancer is not detected.

20. A computer computing system having one or more hardware processors and memory obtains training sequence data including training sequence reads derived from a plurality of samples of a plurality of training targets, wherein each training sequence read includes a nucleotide sequence corresponding to a nucleic acid fragment contained in one of the plurality of samples, and each training sequence read corresponds to a molecule having a threshold amount of methylated cytosine contained within a region of the nucleotide sequence having at least a threshold cytosine-guanine content. The steps include: analyzing the training sequence determination reads using the computer calculation system to determine an additional first quantitative measure derived from the training sequence determination reads corresponding to each of the multiple classification regions; The computer calculation system analyzes the training sequencing reads to determine an additional second quantitative measure derived from the training sequencing reads corresponding to a plurality of control regions. The steps include: determining additional measurement criteria for each of the multiple classification areas based on the additional first quantitative measure for each of the multiple classification areas and the additional second quantitative measure for each of the multiple classification areas using the computer calculation system; A computer computing device generates training data including the additional metrics for each of the classification regions of the plurality of classification regions for the training sequence reads derived from the plurality of training target samples, The steps include: using the training data, the computer computing system implements one or more machine learning algorithms to generate a model that determines the indicator of cancer status in a subject based on the amount of methylated cytosine in at least a portion of the plurality of classification regions; The method according to claims 18 and 19, including the method described in claims 18 and 19.

21. The method according to claim 20, wherein one or more machine learning algorithms include one or more classification algorithms.

22. The aforementioned one or more machine learning algorithms include one or more regression algorithms, The method according to claim 20, wherein the index corresponds to an estimated value of the tumor fraction of the sample.

23. The training sequence determination read includes a first portion of the training sequence data, and an additional training sequence determination read includes a second portion of the training sequence data, and the additional training sequence determination read differs from the training sequence determination read. The method described above is The computer calculation system performs the steps of: analyzing at least one of the first portion or the second portion of the training sequence data to determine the individual frequencies of the multiple variants present in each of the multiple samples; The computer calculation system determines, with respect to each sample, one of the plurality of variants which has the maximum frequency corresponding to the individual frequency that has the maximum value among the individual frequencies originating from each sample; The computer calculation system performs the steps of determining individual scales of tumor fractions for each sample based on the maximum value of each frequency derived from each sample. The method according to claim 18, including the method described in claim 18.

24. The training data includes individual measures of the tumor fraction for each of the plurality of samples, The method according to claim 20, wherein the model is generated based on individual measures of the tumor fraction for each of the plurality of samples.

25. A computer computing system including a processing network and memory generates a data file containing first tokens generated using a first hash function, wherein each first token corresponds to each individual in a group of individuals whose data is stored in a molecular data repository. The computer calculation system transmits the data file to the medical insurance claims data management system, The computer calculation system obtains health data corresponding to the group of individuals from the medical insurance claim data management system in response to the data file, The steps include: generating a number of identifiers using a second hash function different from the first hash function with the computer computing system, wherein each identifier corresponds to one or more tokens associated with each individual in the group of individuals; The computer calculation system obtains second data from the molecular data repository for the group of individuals using the identifier of the number, The computer calculation system determines, for each group of individuals, each part of the first data that corresponds to each part of the second data, The steps include: generating an integrated data repository using the computer calculation system to store the respective parts of the first data and the respective parts of the second data in association with each identifier of the numerical identifier; The method according to claim 1, including the method described in claim 1.

26. The steps include: determining a first set of data processing instructions that can be executed in relation to the first data stored in the integrated data repository using the computer calculation system; The steps include: causing the computer calculation system to execute the first set of data processing instructions to analyze the first medical insurance claim code contained in the first data to determine a first subset of the group of individuals in which a biological state exists; The steps include: generating a first dataset using the computer calculation system that shows a subset of the group of individuals in which the biological state exists; The method according to claim 25, including the method described in claim 25.

27. The steps include determining a second set of data processing instructions that can be executed in relation to the second data stored in the integrated data repository using the computer computing system, The steps include: causing the computer calculation system to execute the second set of data processing instructions to analyze the second medical insurance claim code contained in the second data to determine one or more treatments to be provided to a second subset of the group of individuals; The steps include: generating a second dataset using the computer calculation system that shows one or more treatments provided to a second subset of the group of individuals; The method according to claim 26, including the method described in claim 26.

28. The steps include determining a third subset of the group of individuals using the computer calculation system, which includes a portion of the first subset of the group of individuals that overlaps with a portion of the second subset of the group of individuals, The computer computing system receives a request to perform an analysis of the first dataset and the second dataset in relation to the third subset of the group of individuals. The computer computing system, in response to the request, analyzes the first dataset and the second dataset against a third subset of the group of individuals to determine a measure of the significance of the features of the third subset of the group of individuals regarding the biological state. The method according to claim 27, including the method described in claim 27.

29. The steps include determining one or more genomic mutations present in the third subset of the group of individuals using the aforementioned computer calculation system, The steps include determining a plurality of treatments to be provided to the third subset of the group of individuals using the computer calculation system, The computer calculation system performs the steps of determining the survival rate for each of the third subsets of the group of individuals. The method according to claim 28, including the method described in claim 28.

30. The method according to claim 29, wherein the measure of significance corresponds to the survival rate for one of the plurality of treatments and one of the one or more genomic mutations.

31. The method according to claim 30, further comprising the step of determining, by computer calculation system, the effectiveness of the treatment on a third subset of the group of individuals based on a measure of significance.

32. The method according to claim 31, further comprising the step of determining, by computer calculation system, individuals in a third subset of the group of individuals that have not undergone the treatment.

33. The method according to claim 32, further comprising the step of administering one or more therapeutically effective amounts of the treatment to individuals within the third subset that have not received the treatment.

34. The integrated data repository is arranged according to a data repository schema that includes multiple data tables and multiple logical links between the multiple data tables, The method according to any one of claims 25 to 33, wherein each logical link of the plurality of logical links indicates one or more rows in a data table among the plurality of data tables that correspond to one or more additional rows in an additional data table among the plurality of data tables.

35. The aforementioned multiple data tables, A first data table that stores genomic data of the group of individuals, A second set of data that stores data related to one or more patient visits by an individual to one or more healthcare providers, A third data table that stores information corresponding to each service provided to an individual regarding one or more patient visits to one or more healthcare providers as shown in the second data table, A fourth data table storing personal information of the aforementioned group of individuals, A fifth data table that stores information relating to a health insurance company or government agency that made a payment for the services provided to the group of individuals, A sixth data table that stores information corresponding to the scope of medical insurance coverage for the group of individuals, A seventh data table that stores information related to the medical treatment obtained by the group of individuals, The method according to claim 34, including the method described in claim 34.

36. The identifier of the number generated using the second hash function includes an intermediate identifier, and the method is The computer calculation system applies a salt function to the intermediate identifier to generate the final set of identifiers. The method according to any one of claims 25 to 35, including the method described above.

37. The computer computing system obtains information from an additional data repository, including electronic medical records of an additional group of individuals. The steps include determining, using the computer computing system, an additional subset of the group of individuals corresponding to the group of individuals whose data is stored in the genomics data repository, The steps include modifying the integrated data repository using the computer computing system to store at least a portion of the medical record information of the subset of the additional group of individuals in association with the numerical identifiers. The method according to any one of claims 25 to 36, including the method described above.

38. The computer system performs one or more optical character recognition operations with respect to additional information. The computer system analyzes the additional information obtained from the additional data repository, determines one or more parts of the additional information to be deleted, and creates a corpus of information. The method according to claim 37, including the method described in claim 37.

39. The steps include: analyzing the corpus of information using the aforementioned computer calculation system to determine a subset of the additional group of individuals corresponding to one or more biomarkers; The steps include: using the computer calculation system to store identifiers of the subset of the additional group of individuals, and generating one or more data structures that store an index indicating that the subset of the additional group of individuals corresponds to one or more biomarkers; The method according to claim 38, including the method described in claim 38.

40. The computer system performs the steps of storing one or more data structures in an intermediate data repository, The steps include: performing one or more de-identification operations with respect to the identifiers of a subset of the additional group of individuals using the computer calculation system, and then modifying the integrated data repository to store at least a portion of the additional information of the medical records of a subset of the additional group of individuals in association with the numerical identifiers; The method according to claim 39, including the method described in claim 39.

41. The method according to any one of claims 25 to 40, wherein the molecular data repository stores at least one or more of the following: genomic information, genetic information, metabolomics information, transcriptomics information, fragmentomics information, immune receptor information, methylation information, epigenomic information, or proteomics information.

42. The method according to any of the prior claims, wherein the step of determining the likelihood of recurrence includes MRD testing, real-world evidence (RWE), or both.

43. A system configured to carry out the method described in any of the prior claims.

44. A computer-readable medium comprising the method described in any of the prior claims.

Images