Systems and methods for single sample variant calling utilizing circulating tumor allele fractions

Abstract

Systems and methods for differentiating between somatic and germline variants using a single biological sample are disclosed. One method may include: receiving, at a computer system and from a genotyping workflow performed on a biological sample, genotype data for a plurality of variants identified in a sequence of the biological sample; receiving, at the computer system and from an orthogonal assay workflow performed on the biological sample, tumor fraction data; providing, as input to a multinomial model, the genotype data and the tumor fraction data; and receiving, from the multinomial model, output that identifies a first portion of the plurality of variants as germline. variants, and a second portion of the plurality of variants as somatic variants.

Description

Attorney Docket No. 00316-0017-00304 SYSTEMS AND METHODS FOR SINGLE SAMPLE VARIANT CALLING UTILIZING CIRCULATING TUMOR ALLELE FRACTIONS CROSS REFERENCE TO RELATED APPLICATION

[0001] The application claims priority to U.S. Provisional Application No. 63 / 586,928, filed on September 29, 2023, which is incorporated by reference herein in its entirety. TECHNICAL FIELD

[0002] The present disclosure relates generally to the field of bioinformatics and genomics and, more specifically, to systems and methods for differentiating between somatic and germline variants in a single biological sample. BACKGROUND

[0003] Advancements in genomics have better informed our understanding of diseases, particularly cancer. Genetic mutations play a crucial role in the development and progression of cancer, making the accurate identification of these mutations essential for identifying disease or determining treatment strategies. However, distinguishing somatic mutations (i.e., genetic mutations occurring only in the tumor) from germline mutations (i.e., genetic mutations inherited from parents) may be challenging, especially in cases where only tumor samples are available for analysis.

[0004] The present disclosure is accordingly directed to systems and methods for single sample somatic variant calling that leverage methylation patterns, genomic variants and orthogonally derived allele fraction data to accurately distinguish somatic variants from germline variants. This novel approach mayAttorney Docket No. 00316-0017-00304 eliminate the requirement for paired normal and tumor samples and may improve the accuracy of somatic variant calls. By using ctDNA for a biological sample, e.g., the need for paired normal and tumor samples may be eliminated. Accordingly, it may be possible to identify somatic variants when normal samples are unavailable. The background description provided herein is for the purpose of generally presenting context of the disclosure. Unless otherwise indicated herein, the materials described in this section are not prior art to the claims in this application and are not admitted to be prior art, or suggestions of the prior art, by inclusion in this section. SUMMARY OF THE DISCLOSURE

[0005] According to certain aspects of the disclosure, systems and methods are described that enable precise and efficient variant calling from a mixed DNA sample.

[0006] In summary, one aspect provides a computer-implemented method for differentiating between germline and somatic variants is provided. The computer- implemented method includes: receiving, at a computer system and from a genotyping workflow performed on a biological sample, genotype data for a plurality of variants identified in a sequence of the biological sample; receiving, at the computer system and from an orthogonal assay workflow performed on the biological sample, tumor fraction data; providing, as input to a multinomial model, the genotype data and the tumor fraction data; and receiving, from the multinomial model, output that identifies a first portion of the plurality of variants as germline variants, and a second portion of the plurality of variants as somatic variants.

[0007] In another aspect, a system for differentiating between germline and somatic variants is provided. The system includes: one or more processors; one orAttorney Docket No. 00316-0017-00304 more computer readable media storing instructions that are executable by the one or more processors to perform operations to: receive, from a genotyping workflow performed on a biological sample, genotype data for a plurality of variants identified in a sequence of the biological sample; receive, from an orthogonal assay workflow performed on the biological sample, tumor fraction data; provide, as input to a multinomial model, the genotype data and the tumor fraction data; and receive, from the multinomial model, output that identifies a first portion of the plurality of variants as germline variants, and a second portion of the plurality of variants as somatic variants

[0008] In yet another aspect, a non-transitory computer-readable medium is provided. The non-transitory computer-readable medium is configured to store computer-executable instructions which, when executed by a server, cause the server to perform operations including: receiving, at a computer system and from a genotyping workflow performed on a biological sample, genotype data for variant plurality of variants identified in a sequence of the biological sample; receiving, at the computer system and from an orthogonal assay workflow performed on the biological sample, tumor fraction data; providing, as input to a multinomial model, the genotype data and the tumor fraction data; and receiving, from the multinomial model, output that identifies a first portion of the plurality of variants as germline variants, and a second portion of the plurality of variants as somatic variants.

[0009] In yet another aspect, a computer-implemented method is provided. The computer-implemented method may include: receiving, at a computer system, genomic data derived from an assay associated with a biological sample; receiving, at the computer system and from a genotyping workflow performed on the genomic data, genotype data for a plurality of variants identified in a sequence of theAttorney Docket No. 00316-0017-00304 biological sample; receiving, at the computer system and from orthogonal features that are bioinformatically extracted from the genomic data, tumor fraction data; providing, as input to a multinomial model, the genotype data and the tumor fraction data; and receiving, from the multinomial model, output that identifies a first portion of the plurality of variants as germline variants, and a second portion of the plurality of variants as somatic variants

[0010] Additional objects and advantages of the disclosed embodiments will be set forth in part in the description that follows, and in part will be apparent from the description, or may be learned by practice of the disclosed embodiments. The objects and advantages of the disclosed embodiments will be realized and attained by means of the elements and combinations particularly pointed out in the appended claims.

[0011] It is to be understood that both the foregoing general description and the following detailed description are exemplary and explanatory only and are not restrictive of the disclosed embodiments, as claimed. BRIEF DESCRIPTION OF THE DRAWINGS

[0012] The accompanying drawings, which are incorporated in and constitute a part of this specification, illustrate several embodiments and together with the description, serve to explain the principles of the disclosure.

[0013] FIG.1A depicts an exemplary computer system for executing the methods described herein.

[0014] FIG.1B depicts an exemplary software platform for executing the methods described herein.Attorney Docket No. 00316-0017-00304

[0015] FIG.2 depicts an exemplary workflow for differentiating between somatic variants and germline variants in a biological sample, according to one or more embodiments of the present disclosure.

[0016] FIG.3 depicts an exemplary illustration of how a tumor fraction- informed somatic mutation caller may be integrated with methylation-based classification assays, according to one or more embodiments of the present disclosure.

[0017] FIG.4 depicts a graph illustrating the likelihood or bayes factor threshold that would result in balanced precision and recall performance for a called somatic variant, according to one or more embodiments of the present disclosure.

[0018] FIG.5 depicts a flowchart of an exemplary method of deconvoluting small variants in a biological sample, according to one or more embodiments of the present disclosure.

[0019] FIG.6 depicts a flowchart of another exemplary method of deconvoluting small variants in a biological sample, according to one or more embodiments of the present disclosure.

[0020] FIG.7 depicts an example computing system, according to one or more aspects of the present disclosure. DETAILED DESCRIPTION OF EMBODIMENTS

[0021] The terminology used below may be interpreted in its broadest reasonable manner, even though it is being used in conjunction with a detailed description of certain specific examples of the present disclosure. Indeed, certain terms may even be emphasized below; however, any terminology intended to be interpreted in any restricted manner will be overtly and specifically defined as such inAttorney Docket No. 00316-0017-00304 this Detailed Description section. Both the foregoing general description and the following detailed description are exemplary and explanatory only and are not restrictive of the features, as claimed.

[0022] In this disclosure, the term “based on” means “based at least in part on.” The singular forms “a,” “an,” and “the” include plural referents unless the context dictates otherwise. The term “exemplary” is used in the sense of “example” rather than “ideal.” The terms “comprises,” “comprising,” “includes,” “including,” or other variations thereof, are intended to cover a non-exclusive inclusion such that a process, method, or product that comprises a list of elements does not necessarily include only those elements, but may include other elements not expressly listed or inherent to such a process, method, article, or apparatus. Relative terms, such as “about,” “approximately,” “substantially,” and “generally,” are used to indicate a possible variation of ±10% of a stated or understood value. In addition, the term “between” used in describing ranges of values is intended to include the minimum and maximum values described herein. The use of the term “or” in the claims and specification is used to mean “and / or” unless explicitly indicated to refer to alternatives only or the alternatives are mutually exclusive, although the disclosure supports a definition that refers to only alternatives and “and / or.” As used herein “another” may mean at least a second or more.

[0023] As used herein, the term “user” generally encompasses any person or entity, such as a researcher and / or a care provider (e.g., a doctor, etc.), that may desire information, resolution of an issue, or engage in any other type of interaction with a provider of the systems and methods described herein (e.g., via an application interface resident on their electronic device, etc.). The term “electronic application” or “application” may be used interchangeably with other terms like “program,” or theAttorney Docket No. 00316-0017-00304 like, and generally encompasses software that is configured to interact with, modify, override, supplement, or operate in conjunction with other software.

[0024] One of the fundamental challenges in cancer genomics is the accurate identification of somatic mutations – genetic alterations that occur specifically in tumor cells – versus germline mutations, which are inherited from an individual’s parents and are present in both healthy and tumor cells. Existing approaches to differentiate somatic mutations from germline mutations often involve joint calling methods. These methods typically require both a tumor sample and a paired normal sample (e.g., a biological sample taken from a healthy tissue or cell type from the same individual that also provided the tumor sample). In joint calling, variants present in the matched normal sample are considered part of the individual’s germline genome, and variants found in the tumor sample but absent in the matched normal sample are considered somatic mutations. This approach helps filter out the genetic variations that are present in both the tumor and the individual’s normal genome, thus enabling the accurate identification of somatic mutations.

[0025] However, the need for paired normal samples can sometimes be a limitation in clinical or research settings. Specifically, obtaining paired normal samples may not always be feasible, especially in cases where only tumor samples are available, as is often the case with liquid biopsies, such as circulating tumor DNA (ctDNA) or cell-free DNA (cfDNA) samples. Tumor-only workflows have also been explored, but they face challenges such as relying on population single nucleotide polymorphism (SNP) filtering or machine learning. Moreover, the complexities introduced by lower tumor fractions in cfDNA samples, coupled with the lack of paired normal samples, make accurate somatic variant calling from these samples a particularly challenging problem.Attorney Docket No. 00316-0017-00304

[0026] Accordingly, embodiments of the present disclosure offer an innovative approach to address the challenges associated with somatic variant calling from a single biological sample (e.g., a single ctDNA or cfDNA sample). At the outset, a single biological sample may be processed using a genotyping workflow, which generates a call set that includes both germline and somatic variants. However, this workflow alone may not provide sufficient information to differentiate between germline and somatic variants accurately. Accordingly, operating in parallel, the sample may be processed using an orthogonal assay workflow, either experimentally (e.g., a targeted methylation assay) or bioinformatically extracting orthogonal features using the same assay as in the genotyping workflow) to estimate the proportion of tumor DNA within the DNA sample (i.e., the tumor fraction). The collective data from the genotyping workflow and the orthogonal assay may then be collectively processed (e.g., by a genotype mixture model) to differentiate germline and somatic variant calls and the likely genotypes associated with each variant.

[0027] Accordingly, the processes described above enables improved single sample somatic variant calling, resulting in more accurate and efficient identification of both germline and somatic variants in cancer samples, even in cases where paired normal samples are not available. The described concepts may mitigate the risk of false positive calls, for example, in individuals with non-European ancestry. In some aspects, the calibrated germline calls may be of a higher quality. In an aspect, results from these processes may be included as complementary findings to existing methylation-based cancer classification assays. The candidate somatic variants identified from these results may allow researchers to explore both the genetic and epigenetic components of a given sample.Attorney Docket No. 00316-0017-00304

[0028] The embodiments of the present disclosure address a technical challenge that arises in genomics in the field of somatic variant calling. Specifically, the disclosed concepts incorporate genomic variants and orthogonal assay-based tumor fraction estimation to serve as the computation engine driving the identification of somatic variants. By integrating these metrics, the functionality of the computer system may be expanded to perform complex genetic analyses and variant calling tasks. Furthermore, embodiments of the present disclosure are not confined to theoretical concepts or ideas, but rather, are directed to a systematic methodology that may be implemented and applied in real-world scenarios involving various types of diseases, e.g., one of which may be cancer. The use of empirical data, algorithms, and computations models showcases its practical utility in accurately identifying somatic variants from biological samples containing tumor DNA.

[0029] The subject matter of the present disclosure will now be described more fully hereinafter with reference to the accompanying drawings, which form a part hereof, and which show, by way of illustration, specific exemplary embodiments. An embodiment or implementation described herein as “exemplary” is not to be construed as preferred or advantageous, for example, over other embodiments or implementations; rather, it is intended to reflect or indicate that the embodiment(s) is / are “example” embodiment(s). Subject matter may be embodied in a variety of different forms and, therefore, covered or claimed subject matter is intended to be construed as not being limited to any exemplary embodiments set forth herein; exemplary embodiments are provided merely to be illustrative. Likewise, a reasonably broad scope for claimed or covered subject matter is intended. Among other things, for example, subject matter may be embodied as methods, devices, components, or systems. Accordingly, embodiments may, for example, take the formAttorney Docket No. 00316-0017-00304 of hardware, software, firmware, or any combination thereof. The following detailed description is, therefore, not intended to be taken in a limiting sense.

[0030] Throughout the specification and claims, terms may have nuanced meanings suggested or implied in context beyond an explicitly stated meaning. Likewise, the phrase “in one embodiment” or “in some embodiments,” or “in one aspect” or “in some aspects” as used herein does not necessarily refer to the same embodiment or aspect, and the phrase “in another embodiment” or “in another aspect” as used herein does not necessarily refer to a different embodiment or aspect. It is intended, for example, that claimed subject matter include combinations of exemplary embodiments in whole or in part.

[0031] It is important to note that although the concepts described throughout this disclosure are made in reference to cancer, these designations are for exemplary purposes only and are not intended to be limiting. Specifically, the concepts described herein may be applicable to other disease types.

[0032] FIG.1A depicts an exemplary system for differentiating germline from somatic variants using a single biological sample. Exemplary system 100 includes a data collection component 10, a database 20, and a device data intelligence component 30, operably connected to each other via network 40. Alternatively, or additionally, one or more of the components may be connected with another component locally without reliance on network connection; e.g., through a wired connection. In many aspects described herein, sequencing data of cell-free nucleic acids are used to illustrate the concepts. However, one of skill in the art would understand that the current method may be applied to sequencing data of DNA, RNA, or other materials as well from a variety of sample types, e.g., a blood sampleAttorney Docket No. 00316-0017-00304 (e.g., a serum sample, a plasma sample, a whole blood sample), a urine sample, a saliva sample, a tissue sample, a bone marrow sample, etc.

[0033] As disclosed herein, data collection component 10 may include a device or machine with which sequencing data may be generated. In some embodiments, data collection component 10 may include a sequencing device or a facility that uses a sequencing device to generate nucleic acid (e.g., DNA or RNA) sequence data of biological samples. In some aspects, data collection component 10 may be a database that receives sequencing information generated from one or more sequencing devices. Any suitable liquid or solid biological samples may be used for sequencing. In some embodiments, a biological sample may be cell-based, for example, one or more types of tissue. In some embodiments, a biological sample may be a sample that includes cell-free nucleic acid fragments. Examples of biological samples include, but are not limited to, a blood sample (e.g., a cell-free DNA (cfDNA) sample, a cell-free RNA (cfRNA) sample, a serum sample, a plasma sample, a whole blood sample), a urine sample, a saliva sample, a tissue sample, a bone marrow sample, etc. Further, although sequencing of DNA from these samples is discussed herein, RNA from these samples may alternatively or additionally be sequenced.

[0034] Examples of sequencing data may include, but are not limited to, sequence read data of targeted genomic locations, partial or whole genome sequencing data of the genome represented by nucleic acid fragments in cell-free or cell-based samples, partial or whole genome sequencing data including one or more types of epigenetic modifications (e.g., methylation), or combinations thereof.

[0035] Data acquired by the data collection component 10 may be transferred to database 20 via network 40. In some embodiments, the collected dataAttorney Docket No. 00316-0017-00304 may be analyzed by data intelligence component 30, via a local or network connection. FIG.1B depicts exemplary functional modules that may be implemented to perform tasks of data intelligence component 30.

[0036] FIG.1B depicts an exemplary computer system 110 for processing sequencing data associated with a biological sample to identify variant types and genotype likelihoods for variants present in the DNA sequence of the biological sample. Exemplary system 110 achieves such functionalities by implementing, on one or more computer devices, user input and output (I / O) module 120, memory or database 130, data processing module 140, data analysis module 150, classification module 160, network communication module 170, and any other functional modules that may be needed for carrying out a particular task (e.g., an error correction or compensation module, a data compression module, etc.). As disclosed herein, user I / O module 120 may further include an input sub-module, such as a keyboard, and an output sub-module, such as a display (e.g., a printer, a monitor, or a touchpad). In some embodiments, all functionalities may be performed by one computer system. In some embodiments, the functionalities are performed by more than one computer system.

[0037] Also disclosed herein, a particular task may be performed by implementing one or more functional modules. In particular, each of the enumerated modules itself may, in turn, include multiple sub-modules. For example, data processing module 140 may include a sub-module for data quality evaluation (e.g., for discarding very short sequence reads or sequence reads including obvious errors), a sub-module for normalizing numbers of sequence reads that align to different regions of a reference genome, a sub-module to compensate / correct GC biases, and etc.Attorney Docket No. 00316-0017-00304

[0038] In some embodiments, a user may use I / O module 120 to manipulate data that is available either on a local device or can be obtained via a network connection from a remote service device or another user device. For example, I / O module 120 may allow a user, e.g., via a keyboard, a mouse, or a touchpad, to perform data analysis via a graphical user interface (GUI). In some embodiments, a user may manipulate data via voice control. In some embodiments, user authentication may be required before a user is granted access to the data being requested. In some embodiments, user I / O module 120 may be used to manage various functional modules. For example, a user may request via user I / O module 120 input data while an existing data processing session is in process. A user may do so by selecting a menu option or type in a command discretely without interrupting the existing process. In another example, a user may utilize user I / O module 120 to set various thresholds (e.g., likelihood thresholds), configure genotype identification settings, and / or provide other instructions to computer system 110 that dictate how variants and their corresponding genotypes are identified and subsequently utilized. As disclosed herein, a user may use any type of input to direct and control data processing and analysis via I / O module 120.

[0039] In some embodiments, system 110 further comprises a memory or database 130. In some embodiments, database 130 comprises a local database that may be accessed via user I / O module 120. In some embodiments, database 130 comprises a remote database that may be accessed by user I / O module 120 via network connection. In some embodiments, database 130 is a local database that stores data retrieved from another device (e.g., a user device or a server). In some embodiments, memory or database 130 may store data retrieved in real-time from internet searches. In other embodiments, database 130 may store sequencingAttorney Docket No. 00316-0017-00304 information (e.g., the genetic information that is carried in a particular DNA segment as represented by amino acid sequences) obtained from one or more devices (e.g., a DNA sequencer, translational machinery configured to decode sequence reads produced by the DNA sequencer, etc.). In some embodiments, database 130 may send data to and receive data from one or more of the other functional modules, including, but not limited to, a data collection module (not shown), data processing module 140, data analysis module 150, classification module 160, network communication module 170, and etc. In some aspects, some or all of the mutation identifications and genotype likelihood determinations may be stored on database 130.

[0040] In some embodiments, database 130 may be a database local to the other functional modules. In some embodiments, database 130 may be a remote database that may be accessed by the other functional modules via wired or wireless network connection (e.g., via network communication module 170). In some embodiments, database 130 may include a local portion and a remote portion.

[0041] In some embodiments, system 110 comprises a data processing module 140. Data processing module 140 may receive the real-time data, from I / O module 120 or database 130. In some embodiments, data processing module 140 may perform standard data processing algorithms, such as one or more of noise reduction, signal enhancement, normalization of counts of sequence reads, correction of GC bias, etc. In some embodiments, data processing module 140 may be configured to identify variants in a sample, distinguish between the variant types, and / or provide a tumor fraction estimation. For example, a single biological sample (e.g., ctDNA sample) may be processed using two separate workflows, and the results of each may be collectively processed to differentiate between variant typesAttorney Docket No. 00316-0017-00304 and to a likely genotype estimation for each variant. More particularly, a genotyping workflow may be performed on the sample to identify all mutations in the sample (e.g., both germline and somatic mutations, etc.) and an orthogonal assay (e.g., a targeted methylation assay) may separately be performed on the sample to estimate the fraction of tumor in the sample. This data may be introduced into a genotype mixture model (e.g., a multinomial model) that may be configured to arithmetically process relevant metrics from the collective data to differentiate the variant types and provide genotype likelihoods for each variant.

[0042] In some embodiments, system 110 comprises a data analysis module 150. In some embodiments, data analysis module 150 includes identifying and treating systematic errors in sequencing data, as described in connection with data processing module 140.

[0043] In some embodiments, system 110 comprises a classification module 160, which may embody a “machine-learning model” or “trained classifier.” As used herein, a “machine-learning model” or “trained classifier” generally encompasses instructions, data, and / or a model configured to receive input, and apply one or more of a weight, bias, classification, or analysis on the input to generate an output. The output may include, for example, a classification of the input, an analysis based on the input, a design, process, prediction, or recommendation associated with the input, or any other suitable type of output. A machine-learning model is generally trained using training data, e.g., experiential data and / or samples of input data, which are fed into the model in order to establish, tune, or modify one or more aspects of the model, e.g., the weights, biases, criteria for forming classifications or clusters, or the like. Aspects of a machine-learning model may operate on an inputAttorney Docket No. 00316-0017-00304 linearly, in parallel, via a network (e.g., a neural network), or via any suitable configuration.

[0044] The execution of the machine-learning model may include deployment of one or more machine-learning techniques, such as k-nearest neighbors, linear regression, logistic regression, random forest, gradient boosted machine (GBM), deep learning, a deep neural network, and / or any other suitable machine-learning technique that solves problems in the field of Natural Language Processing (NLP). Supervised, semi-supervised, and / or unsupervised training may be employed. For example, supervised learning may include providing training data and labels corresponding to the training data, e.g., as ground truth. Unsupervised approaches may include clustering, classification or the like. K-means clustering or K-Nearest Neighbors may also be used, which may be supervised or unsupervised. Combinations of K-Nearest Neighbors and an unsupervised cluster technique may also be used. Any suitable type of training may be used, e.g., stochastic, gradient boosted, random seeded, recursive, epoch or batch-based, etc.

[0045] In an exemplary use case, a machine-learning model may be trained to analyze data from a test sample from a test subject whose status with respect to a medical condition is unknown and subsequently classifies the unknown test sample from the test subject based on the likelihood of the subject fitting into a particular category. In some embodiments, the one or more parameters may include a binomial probability score that is calculated based on logistic regression analysis. As disclosed herein, the binomial probability score may correspond to the likelihood of a subject having a certain medical condition, such as cancer. For example, a score of over a predefined threshold may indicate that the subject associated with a test sample is more likely to have cancer than not have cancer. In some embodiments,Attorney Docket No. 00316-0017-00304 the one or more parameters may include a sequencing data distribution pattern correlating with the presence of cancer. A subject associated with a test sample having sequencing data with a pattern resembling the cancer pattern may be diagnosed as having cancer. In some embodiments, a sequencing data distribution pattern may be identified in connection with a specific type of cancer, thus allowing a test sample to be classified as indicative of a certain cancer type.

[0046] As disclosed herein, network communication module 170 may be used to facilitate communications between a user device, one or more databases, and any other suitable system or device through a wired or wireless network connection. Any communication protocol / device may be used, including without limitation a modem, an Ethernet connection, a network card (wireless or wired), an infrared communication device, a wireless communication device, and / or a chipset (such as a Bluetooth™ device, an 802.11 device, a WiFi device, a WiMax device, cellular communication facilities, etc.), a near-field communication (NFC), a Zigbee communication, a radio frequency (RF) or radio-frequency identification (RFID) communication, a PLC protocol, a 3G / 4G / 5G / LTE based communication, and / or the like. For example, a user device having a user interface platform for processing / analyzing tumor fraction data may communicate with another user device with the same platform, a regular user device without the same platform (e.g., a regular smartphone), a remote server, a physical device of a remote IoT local network, a wearable device, a user device communicably connected to a remote server, and etc.

[0047] The functional modules described herein are provided by way of example. It will be understood that different functional modules may be combined toAttorney Docket No. 00316-0017-00304 create different utilities. It will also be understood that additional functional modules or sub-modules may be created to implement a certain utility.

[0048] Referring now to FIG.2, an exemplary process for differentiating between somatic and germline variants in cancer samples is provided, according to one or more aspects of the present disclosure. Aspects of the exemplary process 200 may be performed in accordance with some or all components described in FIG. 1A and 1B.

[0049] At step 205, a biological sample may be obtained from a patient. Examples of biological samples include, but are not limited to, a blood sample (e.g., a serum sample, a plasma sample, a whole blood sample), a urine sample, a saliva sample, a tissue sample, a bone marrow sample, etc. For simplicity purposes, the biological sample described herein is a blood sample.

[0050] The biological sample may be obtained via invasive (e.g., via biopsy) or non-invasive means (e.g., using standard venipuncture techniques, etc.). The biological sample may be a single cfDNA sample, which may include ctDNA and normal cfDNA. With respect to the former, cfDNA corresponds to small fragments of DNA, originating from various cells in the body, which are present in the bloodstream as a result of turnover and cell death. cfDNA analysis may provide insights into general health, tissue damage, and genetic variations. In the context of cancer, cfDNA analysis may help monitor tumor burden, response to treatment, and the presence of genetic mutations. With respect to the latter, ctDNA is a subset of cfDNA that corresponds to fragments of DNA that are shed into the bloodstream by tumor cells. These fragments results from the natural processes of cell death, apoptosis, and necrosis in tumors and carry genetic information that is specific to the tumor from which they originate. It may contain mutations, deletions, rearrangements, andAttorney Docket No. 00316-0017-00304 other genetic alterations present in the tumor cells, and its analysis may provide insights into the genetic mutations within the tumor, which can facilitate one or more of cancer detection or diagnosis, or may help to guide treatment decisions or monitor treatment response. In general, ctDNA may be present in lower concentrations compared to total cfDNA. For example, ctDNA may make up about 0.1% to about 50% of a total cfDNA sample. Although a single ctDNA sample is described, the biological sample may be any sample that includes a mixture of tumor DNA and normal DNA, e.g., an impure biopsy sample.

[0051] At step 210, the biological sample may be processed via a genotyping workflow to identify all mutations in the sample. In an aspect, the genotyping workflow may be an analytical process that identifies genetic variants (e.g., somatic and germline variants) within a DNA (or RNA) sample. It is important to note that the following genotyping workflow process described below is only exemplary, and other methods for genotyping a sample may also be utilized.

[0052] In an aspect, through the genotyping workflow, extracted DNA may be amplified (e.g., via a cycle of denaturation, annealing, and extension) and then subsequently sequenced to determine the genetic variants in designated target regions of the genome. For example, Sanger or next-generation sequencing (NGS) platforms may be used to read the DNA sequence and generate DNA sequencing data. In an aspect, before variants can be called, the raw data generated from the DNA sequencing may be processed and prepared for analysis. This typically involves several preprocessing steps including data filtering (e.g., to filter out low- quality data points or sequences in the raw data resulting from errors or artifacts introduced during sample preparation, amplification, or sequencing), alignment (e.g., to a reference genome), and ultimately variant calling (e.g., via the utilization of oneAttorney Docket No. 00316-0017-00304 or more variant calling algorithms to identify potential variants by comparing the aligned sequence reads to the reference genomes). Variant calling may reveal variations such as SNPs, indels, and structural variants.

[0053] After variant calling, a set of candidate variants may be obtained. However, not all detected variants are of equal interest or quality. To refine the list, variant filtering may be performed. One or more different filtering processes may be performed, such as quality score filtering (e.g., wherein each variant may be assigned a quality score based on one or more of: sequencing depth, mapping quality, etc., and variants with low-quality scores may be filtered out), frequency filtering (e.g., wherein variants that are common in the population and well- documented in public databases may be excluded if a study is focused on rare or disease-specific variants), functional filtering (e.g., wherein variants are annotated to predict their functional impact and then those variants with potential functional consequences may be prioritized), and / or depth filtering (e.g., wherein variants with low sequencing depth (i.e., insufficient coverage) may be filtered out, as they may be less reliable). Each remaining variant after filtering may be annotated with additional information, such as one or more of: its genomic location, gene association, predicted effect on protein function, and allele frequency in various populations. Ultimately, based on the annotation and filtering steps described above, a group of both somatic variants (e.g., variants that are specific to the tumor cells and not present in normal tissue) and germline variants (e.g., variants that are present in all cells of an individual’s body and may have been inherited) may remain.

[0054] At step 215, an orthogonal assay workflow may be performed on the biological sample in step 205 to provide data that may be used to interpret the genotyping methods described in step 210. It is important to note that although stepAttorney Docket No. 00316-0017-00304 210 is described before step 215, this order is not intended to be limiting, and the process performed at step 210 may occur before, after, or in parallel with the process described in step 215.

[0055] In an aspect, the orthogonal assay may be, for instance, a methylation-based or a protein-based assay to determine tumor allele fraction, and can be the same physical assay as the one producing the genotyping results when additional orthogonal features are bioinformatically extracted. In an aspect, a chosen orthogonal assay may need to be robust to clonal hematopoiesis. The orthogonal assay may be used to interpret the results from the genotyping workflow and improve statistical power in determining the somatic genotype. In an aspect, if a methylation-based assay is used, the methylation assay may involve whole-genome methylation sequencing. In another aspect, the methylation assay may involve targeted DNA methylation sequencing using a plurality of nucleic acid probes (e.g., about 100 probes, about 500 probes, about 1000 probes, etc.). In operation, the methylation sequencing detects one or more 5-methylcytosine (5mC) and / or 5- hydroxymethylcytosine (5hmC) in respective nucleic acid fragments obtained from the biological sample. In some embodiments, the methylation sequencing is bisulfite sequencing in which nucleic acid samples are treated with bisulfite to convert unmethylated cytosines to uracils that are subsequently detected as thymines during sequencing analysis. In some embodiments, methylated cytosines undergo enzymatic treatment to be converted to uracils (or a derivative thereof such as dihydrouracils) that are subsequently detected as thymines during sequencing analysis. In an aspect, unmodified cytosines constitute about 95% of the total cytosines in the human genome. Conversion of methylated cytosines instead ofAttorney Docket No. 00316-0017-00304 unmethylated cytosines may lead to fewer alterations to the genome and offer more information for additional analysis such as variant analysis.

[0056] In an aspect, through the methylation assay, a tumor fraction estimation may be determined, through one or more processes described in commonly owned U.S. Patent App. Serial No.17 / 185,885 (corresponding to U.S. Patent App. Pub. No.2021 / 0285042A1), which is incorporated by reference herein in its entirety. This publication discloses systems and methods for calling variants using methylation sequencing data. More particularly, the systems and methods described therein may determine tumor fraction using counts of fragments supporting and not supporting each variant that was generated from WGS sequencing of corresponding cfDNA samples matched to the WGBS data (e.g., the calls for each allele in the plurality of allelic positions). In some aspects, posterior tumor fraction estimates were calculated using a grid search over tumor fraction candidates and by employing a per-variant likelihood defined as a mixture of binomial likelihoods.

[0057] The tumor fraction estimation represents the fraction of cfDNA that likely originates from the tumor, and this information, processed collectively with the variant data produced by the genotyping workflow, may help to distinguish somatic from germline variants and may be utilized to provide a genotype likelihood for each variant type.

[0058] At step 220, the tumor fraction estimation data and the results of the genotyping workflow may be processed by a genotype mixture model. In an aspect, the genotype mixture model may be a multinomial model.

[0059] Generally, a per-read likelihood model may function as follows. Given the observed allele counts (X) quantified from the biological sample and the tumor allele fraction (AF) measured from the same sample, the genotype mixture model,Attorney Docket No. 00316-0017-00304 e.g., a multinomial model, may be utilized to model the counts for each allele. More specifically, in an aspect, the model assumes reads at a given position were generated from the following models:

[0060] represents the counts associated with eachposition i and represents the germline genotypes and somatic haplotypesis the variant allele frequency (VAF) of the somatic variant.

[0061] Specifically, if a single methylation assay is simultaneously used as the genotyping and tumor fraction assay, the per-read likelihood model may function as follows. When working with bisulfite sequencing data, additional care may be needed because the bisulfite-conversion process may affect the ability of sequencing technologies to call cytosine “C” and thymine “T” alleles confidently. Accordingly, the model may leverage a directional bisulfite-sequencing protocol, as described in commonly owned and incorporated U.S. Patent App. Serial No.17 / 185,885 (corresponding to U.S. Patent App. Pub. No.2021 / 0285042A1), to decipher the correct genotype. More specifically, when working with bisulfite sequencing data, the per-read model ( ) takes where Fxand Rxrepresent read counts forread.

[0062] Assuming no somatic mutations, the per-read likelihoods based on different germline genotypes are given by the multinomial probabilitymass function (PMF). This base in the aforementioned incorporated U.S. Patent App. Pub. No.2021 / 0285042A1. When considering possible somaticAttorney Docket No. 00316-0017-00304 mutations, the model is expanded to include the somatic haplotype and corresponding VAF along with sequencing error probability as follows:combinations may be calculated fromcomputer code by iterating through all 40 combinations of and deriving the likelihood based on the germline genotypes and somatic haplotype. Optionally, population allele frequencies of the germline genotypes and somatic haplotypes may be taken into consideration.

[0064] For the determination of germline and somatic genotypes, the model may employ different algorithms dependent on whether somatic mutations are present or not. For instance, the most likely germline genotype when no somatic mutations are present may be determined by the following equation: . Conversely, with the presence of asomatic mutation with a known allele fraction, the most likely somatic haplotype and germline genotype are given by the following equation:

[0065] A likelihood ratio, calculated by the following: ,and Bayes factor, calculated by the following: , may be reported along with the most likely genotypes and somatic haplotype.Attorney Docket No. 00316-0017-00304 Somatic variants for which λ or K exceed predetermined thresholds may be flagged as candidate somatic variants.

[0066] Biological samples used according to embodiments of the disclosure, such as cfDNA, are mixtures of somatic and germline DNA, and the proportion of each in the mixture may be approximated by the tumor fraction estimation generated by the orthogonal assay. The tumor fraction estimation determined using the orthogonal assay described at step 215 should then line up with the allele frequency determined using the genotyping methodology described at step 210. If the allele frequency is very different than the tumor fraction estimation, then the variant may be determined to be germline. If the allele frequency is in line with the tumor fraction estimation, then the variant may be determined to be somatic.

[0067] In one example, set forth in Table 1 below, a tumor fraction estimate of around 10% represents that there is approximately 10% ctDNA in the total cfDNA. Accordingly, a variant exhibiting an allele frequency of approximately 10% may be determined to be a somatic variant. Table 1 lists exemplary allelic counts for the forward and reverse strand from bisulfite-sequencing at a certain locus. Given the count data and sequencing error probability of 1e-3, the most likely germline genotype may be AC, if somatic mutations are not considered. Taking into account, however, a tumor fraction of 10%, the most likely germline genotype may be AA + haplotype C. The log likelihood ratio may be 73.6126443705945, and the Bayes Factor may be 73.07212868699798, which suggests strong or very strong evidence for AA + haplotype C based on interpretation guidelines. Table 1 A C G T Fwd 90 10 - 5 0 5Attorney Docket No. 00316-0017-00304 Rev 90 – 30 10 30 0

[0068] Accordingly, utilizing the techniques and algorithms described above, somatic variants may be differentiated from germline variants, and the germline and somatic genotype likelihoods may be deduced.

[0069] Integration of orthogonally derived tumor allele fraction (e.g., methylation- or protein-based allele fraction estimated from the same cfDNA (plasma) with an existing DNA analysis workflow may improve statistical power in determining the somatic genotype. For instance, diagram 300 in FIG.3 illustrates an exemplary configuration of how the tumor fraction-informed somatic mutation caller may be integrated with methylation-based classification assays. More particularly, various inputs may be introduced to the workflow, e.g., reference genome data 305, sequencing data (e.g., in a text-based format such as FASTQ, etc.) 310 including small variants revealed from one or more targeted methylation assays, and orthogonally derived tumor allele fraction estimates 315 (e.g., methylation-based circulating tumor allele fraction estimated from targeted methylation assays). These inputs may be processed by the various modules in diagram 300 before ultimately being utilized by the tumor fraction-informed somatic mutation caller 320. Through the processes described above, somatic and germline variants along with their genotype likelihoods may be output. Additionally, candidate somatic and germline variants at various specificities given a set of pre-trained thresholds may also be output.

[0070] Referring now to FIG.4, graph 400 is presented that depicts an example process for selecting the likelihood or bayes factor threshold

[0060] thatAttorney Docket No. 00316-0017-00304 would result in balanced precision and recall performance for a called somatic variant. The precision-recall curves shown in FIG.4 are generated by comparing ground-truth somatic variants with a collection of candidate somatic variants obtained by thresholding genotype likelihoods from diagram 320 at various cutoffs, and the cutoff corresponding to the optimal precision-recall performance can be used in diagram 320 for all future somatic calling.

[0071] The described processes offer several advantages over previous methods for identifying somatic variants in cancer samples. For instance, unlike most traditional methods, the approach described herein does not required paired normal samples, such as white blood cell sequencing data, and is more robust to clonal hematopoiesis. This allows the techniques and procedures described herein to be applicable to a wide range of diagnostic scenarios. Furthermore, the techniques described herein may reduce the false positive somatic variant calls arising from private germline variants not captured by public polymorphism databases. Further, the described concepts enable the ability to calibrate germline genotype based on inferred somatic calls.

[0072] Referring now to FIG.5, an exemplary workflow for differentiating somatic from germline variants is provided. At step 505, computer system 110 may receive genotype data for a plurality of variants identified in a sequence of a biological sample. In an aspect, the biological sample may be one of: a blood sample (e.g., a cfDNA sample, a cfRNA sample, a serum sample, a plasma sample, a whole blood sample), a urine sample, a saliva sample, a tissue sample, a bone marrow sample, etc. In an aspect, the biological sample may contain some proportion of tumor-related cells. In an aspect, the plurality of variants in the biological sample may have been determined from a genotyping workflow (e.g., PCR-RestrictionAttorney Docket No. 00316-0017-00304 Fragment Length Polymorphism (PCR-RFLP) genotyping, SNP Genotyping, Allele- Specific PCR genotyping, next-generation sequencing (NGS), etc.) which may be configured to identify specific genetic variations, such as SNPs, insertions, deletions, or other mutations in an individual’s DNA or in a population of organisms. In an aspect, the plurality of variants may be some combination of somatic and germline variants.

[0073] At step 510, computer system 110 may receive tumor fraction data from an orthogonal assay. More particularly, the orthogonal assay may be conducted on the same biological sample that was processed in the genotyping workflow in step 505. In an aspect, step 510 may be performed before, after, or substantially in parallel with the processes discussed in step 505. In an aspect, the orthogonal assay may be, for instance, a targeted methylation assay, which may measure the methylation status of specific regions within the subject’s genome. The methylation data may additionally be utilized to estimate the tumor fraction in the biological sample, as described in commonly owned U.S. Patent App. Serial No.17 / 185,885 (corresponding to U.S. Patent App. Pub. No.2021 / 0285042A1).

[0074] At step 515, the results from the genotyping workflow (i.e., the plurality of identified variants) along with the orthogonally derived tumor fraction may be provided to and computationally processed by a genotype mixture model to differentiate the identified variants (e.g., the identified variants may be classified as germline or somatic). In general, the allele frequencies that are observed in the genotyping data should closely mirror the tumor fraction in the orthogonal data. For instance, if the tumor fraction in the sample is determined to be 10% and the variant base cytosine “C” appears approximately 10% of the time, then that variant may be classified as a somatic variant. Conversely, if the tumor fraction does not mirror theAttorney Docket No. 00316-0017-00304 observed frequency for a particular variant, then that variant may be considered to be a germline variant. In an aspect, the genotype mixture model may be a multinomial model that may leverage various algorithms to differentiate the variant types, as previously described herein. Additionally, these algorithms may be utilized to generate the highest probability germline and somatic genotype likelihoods for each variant.

[0075] At step 520, output from the genotype mixture model may be received that identifies a first portion of the plurality of variants as germline variants and a second portion of the plurality of variants as somatic variants. This data may be utilized for various downstream processes. For instance, the received data may be utilized to annotate training data that may be input into a machine learning classifier configured to determine the presence of a disease in an individual.

[0076] Referring now to FIG.6, another exemplary workflow for differentiating somatic from germline variants is provided. At step 605, computer system 110 may receive genomic data derived from an assay associated with a biological sample (e.g., a blood sample (e.g., a cfDNA sample, a cfRNA sample, a serum sample, a plasma sample, a whole blood sample), a urine sample, a saliva sample, a tissue sample, a bone marrow sample, etc.) In an aspect, the assay may be a targeted methylation assay.

[0077] At step 610, computer system 110 may receive genotype data for a plurality of variants identified in a sequence of the biological sample. In an aspect, the plurality of variants in the biological sample may have been determined from a genotyping workflow (e.g., PCR-Restriction Fragment Length Polymorphism (PCR- RFLP) genotyping, SNP Genotyping, Allele-Specific PCR genotyping, next- generation sequencing (NGS), etc.) which may be configured to identify specificAttorney Docket No. 00316-0017-00304 genetic variations, such as SNPs, insertions, deletions, or other mutations in an individual’s DNA or in a population of organisms. In an aspect, the plurality of variants may be some combination of somatic and germline variants.

[0078] At step 615, computer system 110 may receive tumor fraction data from orthogonal features that are extracted from the genomic data. More particularly, the tumor fraction data may be obtained from the same data set that the genotype data was obtained from. In an aspect, the tumor fraction data may be bioinformatically extracted from the genomic data. In an aspect, the bioinformatics data extraction may be performed before, after or substantially in parallel with the processes described in step 610.

[0079] At step 620, the results from the genotyping workflow (i.e., the plurality of identified variants) along with the orthogonally derived tumor fraction may be provided to and computationally processed by a genotype mixture model to differentiate the identified variants (e.g., the identified variants may be classified as germline or somatic). In general, the allele frequencies that are observed in the genotyping data should closely mirror the tumor fraction in the orthogonal data. For instance, if the tumor fraction in the sample is determined to be 10% and the variant base cytosine “C” appears approximately 10% of the time, then that variant may be classified as a somatic variant. Conversely, if the tumor fraction does not mirror the observed frequency for a particular variant, then that variant may be considered to be a germline variant. In an aspect, the genotype mixture model may be a multinomial model that may leverage various algorithms to differentiate the variant types, as previously described herein. Additionally, these algorithms may be utilized to generate the highest probability germline and somatic genotype likelihoods for each variant.Attorney Docket No. 00316-0017-00304

[0080] At step 625, output from the genotype mixture model may be received that identifies a first portion of the plurality of variants as germline variants and a second portion of the plurality of variants as somatic variants. This data may be utilized for various downstream processes. For instance, the received data may be utilized to annotate training data that may be input into a machine learning classifier configured to determine the presence of a disease in an individual.

[0081] In general, any process discussed in this disclosure that is understood to be computer-implementable may be performed by one or more processors of a computer system, such as system 100, 110, as described above. A process or process step performed by one or more processors may also be referred to as an operation. The one or more processors may be configured to perform such processes by having access to instructions (e.g., software or computer-readable code) that, when executed by the one or more processors, cause the one or more processors to perform the processes. The instructions may be stored in a memory of the computer server. A processor may be a central processing unit (CPU), a graphics processing unit (GPU), or any suitable type of processing unit.

[0082] A computer system, such as system environment 100, may include one or more computing devices. If the one or more processors of the computer system are implemented as a plurality of processors, the plurality of processors may be included in a single computing device or distributed among a plurality of computing devices. If a system environment comprises a plurality of computing devices, the memory of the computer system may include the respective memory of each computing device of the plurality of computing devices.

[0083] FIG.7 is a simplified functional block diagram of a computer system 700 that may be configured as a computing device for executing the processesAttorney Docket No. 00316-0017-00304 described herein, according to exemplary embodiments of the present disclosure. FIG.7 is a simplified functional block diagram of a computer that may be configured according to exemplary embodiments of the present disclosure. In various embodiments, any of the systems herein may be an assembly of hardware including, for example, a data communication interface 720 for packet data communication. The platform also may include a central processing unit (“CPU”) 702, in the form of one or more processors, for executing program instructions. The platform may include an internal communication bus 608, and a storage unit 706 (such as ROM, HDD, SDD, etc.) that may store data on a computer readable medium 722, although the system 700 may receive programming and data via network communications via electronic network 725 (e.g., voice, video, audio, images, or any other data over the electronic network 725). The system 700 may also have a memory 704 (such as RAM) storing instructions 724 for executing techniques presented herein, although the instructions 724 may be stored temporarily or permanently within other modules of system 700 (e.g., processor 702 and / or computer readable medium 722). The system 700 also may include input and output ports 712 and / or a display 710 to connect with input and output devices such as keyboards, mice, touchscreens, monitors, displays, etc. The various system functions may be implemented in a distributed fashion on a number of similar platforms, to distribute the processing load. Alternatively, the systems may be implemented by appropriate programming of one computer hardware platform.

[0084] Program aspects of the technology may be thought of as “products” or “articles of manufacture” typically in the form of executable code and / or associated data that is carried on or embodied in a type of machine-readable medium. “Storage” type media include any or all of the tangible memory of theAttorney Docket No. 00316-0017-00304 computers, processors or the like, or associated modules thereof, such as various semiconductor memories, tape drives, disk drives and the like, which may provide non-transitory storage at any time for the software programming. All or portions of the software may at times be communicated through the Internet or various other telecommunication networks. Such communications, for example, may enable loading of the software from one computer or processor into another, for example, from a management server or host computer of the mobile communication network into the computer platform of a server and / or from a server to the mobile device. Thus, another type of media that may bear the software elements includes optical, electrical and electromagnetic waves, such as used across physical interfaces between local devices, through wired and optical landline networks, and over various air-links. The physical elements that carry such waves, such as wired or wireless links, optical links, or the like, also may be considered as media bearing the software. As used herein, unless restricted to non-transitory, tangible “storage” media, terms such as computer or machine “readable medium” refer to any medium that participates in providing instructions to a processor for execution.

[0085] Furthermore, while some embodiments described herein include some but not other features included in other embodiments, combinations of features of different embodiments are meant to be within the scope of the invention, and form different embodiments, as would be understood by those skilled in the art. For example, in the following claims, any of the claimed embodiments can be used in any combination.

[0086] Thus, while certain embodiments have been described, those skilled in the art will recognize that other and further modifications may be made thereto without departing from the spirit of the invention, and it is intended to claim all suchAttorney Docket No. 00316-0017-00304 changes and modifications as falling within the scope of the invention. For example, functionality may be added or deleted from the block diagrams and operations may be interchanged among functional blocks. Steps may be added or deleted to methods described within the scope of the present invention.

[0087] The above disclosed subject matter is to be considered illustrative, and not restrictive, and the appended claims are intended to cover all such modifications, enhancements, and other implementations, which fall within the true spirit and scope of the present disclosure. Thus, to the maximum extent allowed by law, the scope of the present disclosure is to be determined by the broadest permissible interpretation of the following claims and their equivalents, and shall not be restricted or limited by the foregoing detailed description. While various implementations of the disclosure have been described, it will be apparent to those of ordinary skill in the art that many more implementations are possible within the scope of the disclosure. Accordingly, the disclosure is not to be restricted except in light of the attached claims and their equivalents.

Claims

Attorney Docket No. 00316-0017-00304 WHAT IS CLAIMED IS:

1. A computer-implemented method comprising: receiving, at a computer system and from a genotyping workflow performed on a biological sample, genotype data for a plurality of variants identified in a sequence of the biological sample; receiving, at the computer system and from an orthogonal assay workflow performed on the biological sample, tumor fraction data; providing, as input to a multinomial model, the genotype data and the tumor fraction data; and receiving, from the multinomial model, output that identifies a first portion of the plurality of variants as germline variants, and a second portion of the plurality of variants as somatic variants.

2. The computer-implemented method of claim 1, wherein the biological sample is one of: cell-free DNA (cfDNA), bone marrow, urine, tissue, or plasma.

3. The computer-implemented method of claim 1, wherein the plurality of variants are single nucleotide variations (SNVs).

4. The computer-implemented method of claim 1, wherein the orthogonal assay is a targeted methylation sequencing assay.Attorney Docket No. 00316-0017-00304 5. The computer-implemented method of claim 1, wherein the receiving the output further comprises receiving a highest likelihood genotype for each of the plurality of variants.

6. The computer-implemented method of claim 5, wherein the receiving the highest likelihood genotype comprises receiving at least one of a likelihood ratio and a Bayes factor value.

7. The computer-implemented method of claim 6, wherein the multinomial model is configured to identify the first portion of the plurality of variants as somatic variants when the likelihood ratio is greater than a first predetermined threshold or the Bayes factor value is greater than a second predetermined threshold.

8. The computer-implemented method of claim 1, wherein the identification of the germline variants and the somatic variants is achieved by leveraging a relationship between an estimated tumor fraction, obtained from the tumor fraction data, and a variant allele frequency, obtained from the genotype data.

9. The computer-implemented method of claim 1, further comprising: providing, as additional input to a trained machine learning classifier, at least a portion of the output; and receiving, from the trained machine learning classifier, an indication of a presence of a disease state associated with a subject associated with the biological sample.Attorney Docket No. 00316-0017-00304 10. The computer-implemented method of claim 9, wherein the disease state is cancer.

11. A system, comprising: one or more processors; one or more computer readable media storing instructions that are executable by the one or more processors to perform operations to: receive, from a genotyping workflow performed on a biological sample, genotype data for a plurality of variants identified in a sequence of the biological sample; receive, from an orthogonal assay workflow performed on the biological sample, tumor fraction data; provide, as input to a multinomial model, the genotype data and the tumor fraction data; and receive, from the multinomial model, output that identifies a first portion of the plurality of variants as germline variants, and a second portion of the plurality of variants as somatic variants.

12. The system of claim 11, wherein the biological sample is one of: cell-free DNA (cfDNA), bone marrow, urine, tissue, or plasma.

13. The system of claim 11, wherein the plurality of variants are single nucleotide variations (SNVs).Attorney Docket No. 00316-0017-00304 14. The system of claim 11, wherein the orthogonal assay is a targeted methylation sequencing assay.

15. The system of claim 11, wherein the instructions to receive the output further comprise instructions to receive a highest likelihood genotype for each of the plurality of variants.

16. The system of claim 15, wherein the instructions to receive the highest likelihood genotype comprise instructions to receive at least one of a likelihood ratio and a Bayes factor value.

17. The system of claim 16, wherein the multinomial model is configured to identify the first portion of the plurality of variants as somatic variants when the likelihood ratio is greater than a first predetermined threshold or the Bayes factor value is greater than a second predetermined threshold.

18. The system of claim 11, wherein the identification of the germline variants and the somatic variants is achieved by leveraging a relationship between an estimated tumor fraction, obtained from the tumor fraction data, and a variant allele frequency, obtained from the genotype data.

19. The system of claim 11, wherein the instructions are further executable by the one or more processors to: provide, as additional input to a trained machine learning classifier, at least a portion of the output; andAttorney Docket No. 00316-0017-00304 receive, from the trained machine learning classifier, an indication of a presence of a disease state associated with a subject associated with the biological sample.

20. A non-transitory computer-readable medium storing computer-executable instructions which, when executed by a system, cause the system to perform operations comprising: receiving, at a computer system and from a genotyping workflow performed on a biological sample, genotype data for variant plurality of variants identified in a sequence of the biological sample; receiving, at the computer system and from an orthogonal assay workflow performed on the biological sample, tumor fraction data; providing, as input to a multinomial model, the genotype data and the tumor fraction data; and receiving, from the multinomial model, output that identifies a first portion of the plurality of variants as germline variants, and a second portion of the plurality of variants as somatic variants.

21. A computer-implemented method comprising: receiving, at a computer system, genomic data derived from an assay associated with a biological sample; receiving, at the computer system and from a genotyping workflow performed on the genomic data, genotype data for a plurality of variants identified in a sequence of the biological sample;Attorney Docket No. 00316-0017-00304 receiving, at the computer system and from orthogonal features that are bioinformatically extracted from the genomic data, tumor fraction data; providing, as input to a multinomial model, the genotype data and the tumor fraction data; and receiving, from the multinomial model, output that identifies a first portion of the plurality of variants as germline variants, and a second portion of the plurality of variants as somatic variants.

22. The method of claim 21, wherein the assay is a targeted methylation assay.

23. The method of claim 21, wherein the biological sample is one of: cell-free DNA (cfDNA), bone marrow, urine, tissue, or plasma.

24. The method of claim 21, wherein the plurality of variants are single nucleotide variations (SNVs).

Images