Determining the genotype of CYP2D6

Abstract

Methods and systems for allele typing and variant calling are provided herein. Genotyping CYP2D6 is complicated in some cases by the fact that one or both alleles involve tandem rearrangements and / or copy number changes. Furthermore, some alleles share homologous regions, e.g., CYP2D7 and CYP2D8P, with 100% identical stretches. In some embodiments, the results of the systems and methods disclosed herein are used as input for generating a report.

Description

[Background technology]

[0001] background The phase I metabolic enzyme CYP2D6 is notoriously difficult to genotype accurately. Multiple studies have reported inconsistent results between sequencing and single-variant genotyping techniques. Its small size (approximately 4400 nucleotides from start ATG to stop codon), along with the polymorphic nature of CYP2D6 and its surrounding loci, contribute to the complexity of comprehensively and accurately genotyping it.

[0002] Therefore, there is a need for methods to accurately classify complex structures into alleles using sequence information. [Overview of the Initiative] [Means for solving the problem]

[0003] overview A method for allele typing is disclosed. A method for treatment based on allele typing is disclosed.

[0004] A method is disclosed that includes the steps of determining a plurality of known allele sequences, determining a plurality of sequence reading data for a target region of a target chromosome, wherein the target region of the chromosome includes one or more loci, aligning the plurality of sequence reading data with a plurality of known allele sequences, determining the number of sequence reading data aligned to each known allele sequence for each known allele sequence of the plurality of known allele sequences based on the alignment, and determining the number of known allele sequences present at one or more loci for one or more loci based on the number of sequence reading data aligned to each known allele sequence.

[0005] A method is disclosed that includes the steps of determining a plurality of known allele sequences, determining a plurality of sequence reading data for a target region of a target chromosome, wherein the target region of the chromosome includes one or more loci, aligning the plurality of sequence reading data with a plurality of known allele sequences, determining, based on the alignment, the number of sequence reading data families aligned to each known allele sequence (i.e., the number of nucleic acid molecules, where a sequence reading data family may be a group of sequence reading data corresponding to a single nucleic acid molecule), and determining, based on the number of sequence reading data families aligned to each known allele sequence, the number of known allele sequences present at one or more loci.

[0006] A method is disclosed that includes the steps of determining a plurality of known allele sequences, determining a plurality of sequence reading data for a target region of a target chromosome, wherein the target region of the chromosome includes one or more loci, aligning the plurality of sequence reading data with a plurality of known allele sequences, determining the number of sequence reading data aligned to each known allele sequence for each known allele sequence of the plurality of known allele sequences based on the alignment, generating one or more supersets of known allele sequences based on the number of sequence reading data aligned to each known allele sequence, and determining the number of known allele sequences present at one or more loci for one or more loci based on the number of distinctly different reading data in one or more supersets of known allele sequences.

[0007] A method is disclosed, comprising the steps of: determining a plurality of known allelic sequences; determining a plurality of sequence reads of a target region of a subject chromosome, wherein the target region of the chromosome comprises one or a plurality of loci; aligning the plurality of sequence reads with the plurality of known allelic sequences; based on the alignment, for each known allelic sequence of the plurality of known allelic sequences, determining the number of families of sequence reads that align to the respective known allelic sequence; generating one or a plurality of supersets of the known allelic sequences based on the number of families of sequence reads that align to each known allelic sequence; and determining, for the one or more loci, the known allelic sequences present at the one or more loci based on the number of distinct families of reads in the one or more supersets of the known allelic sequences.

[0008] In some embodiments, the results of the systems and methods disclosed herein are used as input to generate a report. The report can be in paper or electronic form. For example, the determination of an allelic type (e.g., an allelic sequence) determined by the methods and systems disclosed herein can be directly displayed in such a report.

[0009] Various steps of the methods disclosed herein, or steps performed by the systems disclosed herein, can be implemented at the same or different times, at the same or different geographical locations, such as countries, and / or by the same or different people.

[0010] The accompanying drawings, which are incorporated in and constitute a part of this specification, illustrate certain embodiments, and together with the written description, serve to explain certain principles of the methods, computer-readable media and systems disclosed in this specification. The description provided herein is better understood when read in conjunction with the accompanying drawings, which are included by way of example and not by way of limitation. Like reference numerals are understood to identify like components throughout the several drawings, unless the context indicates otherwise. It is also understood that some or all of the drawings may be schematic representations for illustrative purposes and do not necessarily show the actual relative size or location of the elements shown. BRIEF DESCRIPTION OF THE DRAWINGS

[0011] [Figure 1-1] FIG. 1A is a flowchart schematically showing exemplary method steps for allelic typing.

[0012] [Figure 1-2] FIG. 1B is a flowchart schematically showing another exemplary method step for allelic typing.

[0013] [Figure 2] FIG. 2 shows an example of a system for allelic typing.

[0014] [Figure 3] FIG. 3 shows an exemplary nucleic acid structure.

[0015] [Figure 4] FIG. 4 shows exemplary rearrangements and other complex structures.

[0016] [Figure 5] FIG. 5 shows exemplary sequence read data.

[0017] [Figure 6] FIG. 6 shows an exemplary graph data structure.

[0018] [Figure 7] Figure 7 shows exemplary different CYP2D6 alleles.

[0019] [Figure 8] Figure 8 shows an illustrative comparison.

[0020] [Figure 9] Figure 9 shows an illustrative comparison.

[0021] [Figure 10] Figure 10 shows an illustrative comparison.

[0022] [Figure 11] Figure 11 shows an exemplary comparison.

[0023] [Figure 12] Figure 12 shows an exemplary CNV call.

[0024] [Figure 13] Figure 13 shows an exemplary CNV call.

[0025] [Figure 14] Figure 14 shows an exemplary CNV call.

[0026] [Figure 15] Figure 15 shows an exemplary CNV call. [Modes for carrying out the invention]

[0027] I. Definition To facilitate understanding of this disclosure, certain terms are first defined below. Further definitions of the terms below and other terms may be provided throughout this specification. If the definitions of the terms provided below do not coincide with the definitions in any patent application or published patent incorporated by reference, the definitions provided in this application should be used to understand the meaning of the terms.

[0028] Where used herein and in the appended claims, the singular forms “one (a),” “one (an),” and “this (the)” include plural references unless the context explicitly indicates otherwise. Thus, for example, a reference to “one (a) method” includes one or more methods and / or steps of a type that would be apparent to those skilled in the art upon reading herein and / or this disclosure. It is also understood that there is an implied “about” before temperatures, concentrations, cycles, number of bases or base pairs, coverage, etc., discussed in this disclosure, so as to whether subtle and intangible equivalents are within the scope of this disclosure. In this application, the use of the singular form includes the plural form unless otherwise specifically stated. Furthermore, the use of “comprise,” “comprises,” “comprising,” “contain,” “contains,” “containing,” “include,” “includes,” and “including” is not intended to be limiting.

[0029] It should be understood that the terminology used herein is intended solely to describe specific embodiments and is not intended to be limiting. Furthermore, unless otherwise defined, all technical and scientific terms used herein have the same meaning as commonly understood by those skilled in the art to which this disclosure belongs. When describing and claiming methods, computer-readable media and systems, the following terms and their grammatical variations are used according to the definitions set forth below. Detailed explanation II. Introduction

[0030] Methods and systems for determining allele types from nucleic acid samples obtained from a test subject are provided herein. In some embodiments, the nucleic acid sample may be, but is not limited to, cell-free nucleic acid (cfNA), genomic DNA, or RNA. In some embodiments, the nucleic acid sample may originate from a specific chromosome and / or a specific region of a chromosome. In some embodiments, the nucleic acid sample may originate from all or part of a metabolic enzyme, such as CYP2D6. a. Metabolic enzyme CYP2D6

[0031] Since CYP2D6 can be duplicated, multiplied, or deleted, the first challenge we face is copy number variation (CNV). On a single allele, CYP2D6 has been described to date, ranging from zero to a wide range of 12 copies. A simple assessment of the number of CYP2D6 genes present in a particular sample is, at first glance, straightforward; the complicating factor lies in the chimeras and / or hybrids that exist with its nearest neighbor and the pseudogene CYP2D7. Based on their structures, CYP2D6 and CYP2D7 can form either two main types of hybrids—CYP2D6 / CYP2D7 hybrids or CYP2D7 / CYP2D6 hybrids.

[0032] Even with accurate calculations of the number of CYP2D6 genes and hybrids present in a particular sample, evaluating sequence variations is virtually impossible. CYP2D6 contains numerous sequence variations, including point mutations, insertions, and deletions. The problem lies in determining which CYP2D6 sequence variants should be investigated. While commercially available CYP2D6 genotyping panels are intentionally available, a clear drawback of genotyping panels designed to detect a single sequence variant is the possibility of known and unknown mutations in the remaining unexamined sequences of the gene. Here, even when sequencing the entire gene using next-generation sequencing techniques, along with software tools for analyzing and calling CYP2D6 genotypes, there is a tendency for misalignment of sequence read data to highly homologous CYP2D7 and failure to detect structural variants (hybrids and / or duplication and multiplication). III. Exemplary Systems and Methods

[0033] In step 104A, the data may be preprocessed. For example, step 104A may include constructing an allele k-mer data structure. The allele k-mer data structure may be a database. The allele k-mer data structure may be a flat file. The allele k-mer data structure may be any form of data structure. Constructing an allele k-mer data structure may include dividing a known allele sequence into k-mer quantities. For example, k-mer quantities having a length of about 100 to about 200 nucleotides. In one embodiment, a k-mer quantity may have a length of 143 nucleotides. Constructing an allele k-mer data structure may include associating each k-mer with metadata. The metadata may include, for example, an indication of the amount of allele containing the k-mer, as well as an allele identifier and the k-mer start position for each allele containing the k-mer.

[0034] In step 106A, sequencing may be performed. For example, step 106A may include obtaining (or otherwise determining, reading, receiving, etc.) sequence read data pairs (e.g., test sequence read data) from a cell-free nucleic acid (cfDNA) sample obtained from the test subject. Step 106A may include performing alignment between the test sequence read data and known allele sequences. For example, step 106A may include performing alignment between the test sequence read data and k-mers in an allele k-mer data structure. Sequencing may determine the alleles supported by the test sequence read data. An allele may be supported by more than one test sequence read data. A test sequence read data may support more than one allele. In one embodiment, if a test sequence read data aligns to an allele (e.g., an allele of a k-mer) with a percentage identity exceeding a certain threshold, the test sequence read data may be found to support that allele. For example, threshold percentage identity can be 80%, 85%, 90%, 95%, 96%, 97%, 98%, 99%, 100%, etc. In one embodiment, the threshold percentage identity may be 100%, requiring a "perfect" match between the test sequence reading data and the allele (e.g., the k-mer allele).

[0035] In one embodiment, step 106A may include determining the number of test sequence reading data families that support an allele (e.g., the number of nucleic acid molecules that support the allele). Each test sequence reading data may include a barcode. The barcode may identify the nucleic acid molecule (e.g., test sequence reading data family) to which the test sequence reading data is associated. In one embodiment, a test sequence reading data family may be found to support an allele if it aligns to the allele with a percentage identity above a certain threshold. For example, the threshold percentage identity may be 80%, 85%, 90%, 95%, 96%, 97%, 98%, 99%, 100%, etc. In one embodiment, the threshold percentage identity may be 100%, requiring a "perfect" match between the test sequence reading data family and the allele (e.g., the k-mer allele).

[0036] In step 108A, a clustering operation may be performed. Alleles may be selected by the number of supporting test sequence readings (or supporting test sequence reading data families) to construct one or more allele supersets. An allele superset may be constructed by determining the first allele associated with the highest number of supporting test sequence readings (or supporting test sequence reading data families). The first allele may form the basis of the allele superset. If a given allele is associated with supporting test sequence readings (or supporting test sequence reading data families) that are themselves a subset of the supporting test sequence readings (or supporting test sequence reading data families) of the first allele, then further alleles may be added to the allele superset. Alleles not included in the allele superset of the first allele may be used in a similar manner to construct one or more further allele supersets.

[0037] The allele superset can be a data structure. The allele superset can be a database. The allele superset can be a flat file. The allele superset can include a representation of a Hasse diagram. A Hasse diagram is a representation of the relationships between elements of a partially ordered set with implied upward directionality. Points or nodes may represent each element of the partially ordered set, and the nodes may be connected by line segments according to the following rules: 1) If p<q in the partially ordered set, the point corresponding to p appears below the point corresponding to q in the diagram; 2) If p covers q, the two points p and q are connected by a line segment. In certain embodiments, the Hasse diagram may be represented as a graph data structure, for example, a directed acyclic graph (DAG) or the like. For example, a DAG comprising a line from node A to node B, where node A strictly contains node B, and there is no node C such that node A strictly contains node C and node C strictly contains node B.

[0038] In step 110A, alleles may be classified. For example, in step 110A, an allele type may be determined for a given allele. Alleles may be classified based on one or more allele supersets. In certain embodiments, where only one allele superset is constructed, the first allele of the superset may be classified as an allele present at a chromosomal locus (e.g., a haploid locus). In certain embodiments, where a plurality of allele supersets are constructed, the first alleles of the two supersets having the cumulative maximum number of distinct supporting test sequence read data (or the cumulative maximum number of distinct supporting test sequence read data families) may be classified as alleles present at a chromosomal locus (e.g., a diploid locus).

[0039] Allele classification may be used to direct treatment of a subject. Whether a subject has a disease may have been previously unknown, or it may have been known. The disease may be cancer. The method may include a step of treating the disease by administering one or more treatments to the subject. Treatments may include administering immunotherapy, chemotherapy, radiation therapy, or performing surgery to remove all or part of a tumor. The method may include a step of assisting in communicating the determination of allele classification to the subject related to the test sample.

[0040] Figure 1B is a schematic flowchart illustrating exemplary techniques for allele typing and / or variant calling in cell-free nucleic acid (cfDNA) samples obtained from a test subject. Allele typing may be used to determine one or more alleles present at a chromosomal locus. Variant calling may be used to identify the presence of known or unknown variants. Variant calling may be used to characterize cancer progression. As shown, method 100B may include obtaining data in step 102B. The data may include sequence data, e.g., allele sequence data and / or decoy sequence data.

[0041] A decoy sequence is a sequence of genomic material (generally human) that is similar to the sequence the inventors wish to examine (e.g., the region the inventors wish to genotype). Since these encode an alternative form of a region or gene (hence the name "alt"), they are no longer part of the reference. The problem for the inventors is to deploy targeted sequencing, which is a method for selecting only molecules from the portion of the genome that matches certain identified regions (these "identified regions" are called probes or baits, which in the inventors' case are 120 base pairs long): Occasionally, a probe designed to capture molecules from the region of interest may instead capture molecules from one of these "alt" sequences. In these cases, the inventors can detect this because the read data (or read data pair) aligns better to the decoy than to the human reference.

[0042] In one embodiment, the decoy sequence may include a decoy sequence selected to identify contamination in the test sample. One or more decoy sequences may include one or more non-human reference sequences. For example, one or more decoy sequences may include a bovine reference sequence, a rat reference sequence, a microbial reference sequence, or a combination thereof. Any pair of test sequences aligned to a non-human decoy sequence may be used to support the conclusion that the test sample is contaminated with DNA from a source other than the subject of the test. The concept is the same as above, except that the inventors use the sequence of the inventors' suspected contaminant as “decoy” in this specification. For example, suppose the inventors suspect that some bovine DNA is present in their sample (this happened!), they add the entire bovine genome to their decoy list, and when they align the read data pairs to both the human reference and the bovine genome (decoy), they find that the contaminant molecules align better to the bovine than to the human.

[0043] In step 104B, the data may be preprocessed. For example, step 104B may include constructing an allele k-mer data structure. The allele k-mer data structure may be a database. The allele k-mer data structure may be a flat file. The allele k-mer data structure may be any form of data structure. Constructing an allele k-mer data structure may include dividing a known allele sequence into k-mer quantities. For example, k-mer quantities having a length of about 100 to about 200 nucleotides. In one embodiment, a k-mer quantity may have a length of 143 nucleotides. Constructing an allele k-mer data structure may include associating each k-mer with metadata. The metadata may include, for example, an indication of the amount of allele containing the k-mer, as well as an allele identifier and the k-mer start position for each allele containing the k-mer.

[0044] For example, step 104B may include constructing a decoy data structure. The decoy data structure may be a database. The decoy data structure may be a flat file. The decoy data structure may be any form of data structure. By structuring the algorithm in this way (i.e., using target sequences + decoy sequences), the inventors can maintain some flexibility. The idea is that the inventors can always add any number of previously unknown "problematic" sequences to the decoy, where problematic means sequences similar to one of the inventors' targets (in other words, sequences that the inventors might accidentally pick up instead of target regions using their targeted sequencing technology development).

[0045] In step 106B, sequencing may be performed. For example, step 106B may include obtaining (or otherwise determining, reading, receiving, etc.) sequence reading data (e.g., test sequence reading data) from a cell-free nucleic acid (cfDNA) sample obtained from the test subject. Step 106B may include performing alignment between the test sequence reading data and known allele sequences. For example, step 106B may include performing alignment between the test sequence reading data and k-mers in an allele k-mer data structure. Sequencing may determine the alleles supported by the test sequence reading data. An allele may be supported by more than one test sequence reading data. A test sequence reading data may support more than one allele. In one embodiment, if a test sequence reading data aligns to an allele (e.g., an allele of a k-mer) with a percentage identity exceeding a certain threshold, the test sequence reading data may be found to support that allele. For example, threshold percentage identity may be 80%, 85%, 90%, 95%, 96%, 97%, 98%, 99%, 100%, etc. In one embodiment, threshold percentage identity may be 100%, requiring a “perfect” match between the test sequence reading data and the allele (e.g., the k-mer allele), indicating no mismatches and no indels. In another embodiment, threshold percentage identity may be less than 100%, requiring an “imperfect” match between the test sequence reading data and the allele (e.g., the k-mer allele), indicating at least one mismatch and / or at least one indel. The percentage identity representation may be determined for each alignment and stored for later processing. The result of the alignment may be indicated by an alignment score, which is described in more detail with respect to the alignment component 215. The alignment score may be equal to the sum of the number of mismatches and the number of indels.

[0046] In one embodiment, step 106B may include determining the number of test sequence reading data families that support an allele (e.g., the number of nucleic acid molecules that support the allele). Each test sequence reading data may include a barcode. The barcode may identify the nucleic acid molecule (e.g., test sequence reading data family) to which the test sequence reading data is associated. In one embodiment, a test sequence reading data family may be found to support an allele if it aligns to the allele with a percentage identity above a certain threshold. For example, the threshold percentage identity may be 80%, 85%, 90%, 95%, 96%, 97%, 98%, 99%, 100%, etc. In one embodiment, the threshold percentage identity may be 100%, requiring a "perfect" match between the test sequence reading data family and the allele (e.g., the k-mer allele).

[0047] Step 106B may include performing alignment between the test sequence read data and the decoy sequence. For example, step 106B may include performing alignment between the test sequence read data and the decoy sequence in the decoy data structure. Sequence processing may determine the decoy sequence supported by the test sequence read data. In one embodiment, if the test sequence read data aligns to the decoy sequence with a percentage identity exceeding a certain threshold, the test sequence read data may be found to support the decoy sequence. For example, the threshold percentage identity may be 80%, 85%, 90%, 95%, 96%, 97%, 98%, 99%, 100%, etc. In one embodiment, the threshold percentage identity may be 100%, requiring a "perfect" match between the test sequence read data and the decoy sequence, indicating no mismatches and no indels. The percentage identity indication may be determined for each alignment and stored for later processing. In one embodiment, one or more test sequence reads that align with one or more decoy sequences with 100% identity may be discarded and not used for further processing. In one embodiment, insofar as the decoy sequences include one or more non-human sequences, any test sequence reads that match a non-human decoy sequence with 100% identity may be used to help identify a test sample as contaminated. Notifications associated with potential contaminated samples may be generated and / or sent.

[0048] The alignment results may be indicated by an alignment score, which is described in more detail with respect to the alignment component 215. The alignment score may be equal to the sum of the number of mismatches and the number of indels.

[0049] In step 108B, the clustering operation may be performed based on alignments between test sequence reading data and known allele sequences. Known alleles may be selected by the number of supporting test sequence reading data (or the number of supporting test sequence reading data families) to construct one or more allele supersets. An allele superset may be constructed by determining the first allele associated with the highest number of supporting test sequence reading data (or the highest number of supporting test sequence reading data families). The first allele may form the basis of the allele superset. If a given allele is associated with supporting test sequence reading data (or supporting test sequence reading data families) that are themselves a subset of the supporting test sequence reading data (or supporting test sequence reading data families) of the first allele, then further alleles may be added to the allele superset. Alleles not included in the allele superset of the first allele may be used in a similar manner to construct one or more further allele supersets.

[0050] The allele superset may be a data structure. The allele superset may be a database. The allele superset may be a flat file. The allele superset may include a representation of a Hasse diagram. A Hasse diagram is a representation of the relationships of elements of a partially ordered set, with implied upward directionality. Points or nodes may represent each element of the partially ordered set, and the nodes may be connected by line segments according to the following rules: 1) If p < q in the partially ordered set, the point corresponding to p appears below the point corresponding to q in the diagram; 2) If p covers q, the two points p and q are connected by a line segment. In certain embodiments, the Hasse diagram may be represented as a graph data structure, such as a directed acyclic graph (DAG). For example, a DAG that includes a line from node A to node B, where node A strictly contains node B, and there is no node C such that node A strictly contains node C and node C strictly contains node B.

[0051] In step 110B, alleles may be classified. For example, in step 110B, an allele type may be determined for a given allele. Alleles may be classified based on one or more allele supersets. In certain embodiments, where only one allele superset is constructed, the first allele of the superset may be classified as an allele present at a chromosomal locus (e.g., a haploid locus). In certain embodiments, where a plurality of allele supersets are constructed, the first alleles of the two supersets having the cumulative maximum number of distinct supporting test sequence read data (or the cumulative maximum number of distinct supporting test sequence read data families) may be classified as alleles present at a chromosomal locus (e.g., a diploid locus).

[0052] Allele classification may be used to direct treatment of a subject. Whether a subject has a disease may have been previously unknown, or it may have been known. The disease may be cancer. The method may include a step of treating the disease by administering one or more treatments to the subject. Treatments may include administering immunotherapy, chemotherapy, radiation therapy, or performing surgery to remove all or part of a tumor. The method may include a step of assisting in communicating the determination of allele classification to the subject related to the test sample.

[0053] In step 110B, test sequence read data pairs associated with germline alignment scores greater than the decoy alignment score may be analyzed to determine and / or identify the test sequence read data pairs as variants. Variant calling is the process of identifying the true differences between the sequence read data of the test sample and the reference sequence. Variant calling may be carried out as described in more detail with respect to the variant caller component 219 below. In one embodiment, a test sequence read data pair may be identified as a somatic variant. In one embodiment, a test sequence read data pair may be identified as a variant which is a candidate variant associated with a somatic event. In step 110B, a candidate variant may be identified in the test sequence read data pair. In one embodiment, a candidate variant may be identified by comparing the test sequence read data pair with a reference sequence of a target region in a reference genome (e.g., human reference genome hg19). The ends of the test sequence read data pair may be aligned to the reference sequence, and the genomic locations of the mismatched ends and mismatched nucleotide bases adjacent to the ends may be recorded as the locations of the candidate variant. In some embodiments, the genomic locations of mismatched nucleotide bases relative to the left and right ends are recorded as the locations of the called variants. Furthermore, candidate variants may be identified based on the sequencing depth of the target region. In particular, a greater number of sequence read data, for example, helps to resolve mismatches or other base pair variations between sequences (e.g., using redundancy), so higher confidence may be obtained when identifying variants in target regions with a greater sequencing depth.

[0054] In one embodiment, the reference sequence used for variant calling may include one or more reference sequences. One or more reference sequences may be selected to identify contamination in the test sample. One or more reference sequences may include one or more non-human reference sequences. For example, one or more reference sequences may include a bovine reference sequence, a rat reference sequence, a microbial reference sequence, a combination thereof, and the like. Any pair of test sequences identified as a non-human variant may be used to support the conclusion that the test sample is contaminated with DNA from a source other than the subject of the test.

[0055] Figure 2 shows an example of a system 200 for determining the allele type and / or variant of a test subject 211 according to one embodiment of the present disclosure. The system 200 may process one or more samples 201 from the subject 211 to generate sequence reading data. The system 200 may include a laboratory system 202, a computer system 210, and / or other components. It should be noted that the laboratory system 202 and the computer system 210 may be remote from each other and may be connected to each other via a computer network (not shown). The laboratory system 202 may include a sample collection and preparation pipeline 203, a sequencing pipeline 205, a sequence reading data data store 209, and / or other components. The sequencing pipeline 205 may include one or more sequencing devices 207 (shown as sequencing devices 207a...n in Figure 2).

[0056] The methods of this disclosure may have a wide range of applications in the manipulation, preparation, identification, quantification, and / or analysis of cell-free nucleic acids. As shown in Figure 2, the sample collection and preparation pipeline 203 may include obtaining a cfDNA reference sample 201 from one or more reference subjects and obtaining a cfDN test sample 211 from a test subject. Where used herein, polynucleotides may include any type of nucleic acid, e.g., DNA and / or RNA. For example, if the polynucleotide is DNA, it may be genomic DNA, complementary DNA (cDNA), or any other deoxyribonucleic acid. Polynucleotides may also be cell-free nucleic acids, e.g., cell-free DNA (cfDNA). For example, polynucleotides may be circulating cfDNA. Circulating cfDNA may include DNA shed from somatic cells via apoptosis or necrosis. cfDNA shed via apoptosis or necrosis may originate from normal (e.g., healthy) somatic cells. a. Sample

[0057] The isolation and extraction of cell-free polynucleotides can be carried out through the collection of samples using various techniques. Samples may be any biological samples isolated from a subject. Samples may include body tissues, whole blood, platelets, serum, plasma, feces, red blood cells, white blood cells (or leucocytes), endothelial cells, tissue biopsies (e.g., biopsies from known or suspected solid tumors), cerebrospinal fluid, synovial fluid, lymph, ascites, interstitial fluid or extracellular fluid (e.g., fluid from the intercellular space), gingival fluid, crevicular fluid, bone marrow, pleural fluid, cerebrospinal fluid, saliva, mucous membranes, sputum, semen, sweat, and urine. Preferably, the samples are body fluids, particularly blood and its fractions, as well as urine. Nucleic acids may include DNA and RNA, and nucleic acids may be in double-stranded and single-stranded forms. The sample may be in the form initially isolated from the subject, or it may not have been subjected to further processing to remove or add components such as cells, to enrich one component by comparing it with another, or to convert one form of nucleic acid to another, for example, RNA to DNA, or single-stranded nucleic acid to double-stranded nucleic acid. Therefore, for example, a bodily fluid sample for analysis may be plasma or serum containing cell-free nucleic acids, such as cell-free DNA (cfDNA).

[0058] In some embodiments, the sample volume of body fluid obtained from the subject depends on the desired reading depth for the sequenced region. Exemplary volumes are about 0.4–40 ml, about 5–20 ml, and about 10–20 ml. For example, the volume may be about 0.5 ml, about 1 ml, about 5 ml, about 10 ml, about 20 ml, about 30 ml, about 40 ml, or more milliliters. The volume of sampled plasma is typically between about 5 ml and about 20 ml.

[0059] A sample may contain varying amounts of nucleic acids. Typically, the amount of nucleic acid in a given sample is considered equivalent to multiple genome equivalents. For example, a sample containing approximately 30 ng of sample DNA is equivalent to approximately 10,000 (10) genome equivalents. 4 ) May contain the equivalent of a haploid human genome, and in the case of cfDNA, approximately 200 billion (2 × 10⁻¹⁶). 11It may contain ) individual polynucleotide molecules. Similarly, a sample of about 100 ng of DNA may contain about 30,000 haploid human genome equivalents, and in the case of cfDNA, it may contain about 600 billion individual molecules.

[0060] In some embodiments, the sample includes nucleic acids from different sources, such as cells and cell-free sources (e.g., blood samples).

[0061] Exemplary amounts of cell-free nucleic acids in the sample before amplification are typically in the range of about 1 femtogram (fg) to about 1 microgram (μg), for example, about 1 picogram (pg) to about 200 nanograms (ng), about 1 ng to about 100 ng, and about 10 ng to about 1000 ng. In some embodiments, the sample contains up to about 600 ng, up to about 500 ng, up to about 400 ng, up to about 300 ng, up to about 200 ng, up to about 100 ng, up to about 50 ng, or up to about 20 ng of cell-free nucleic acid molecules. If necessary, the amount is at least about 1 fg, at least about 10 fg, at least about 100 fg, at least about 1 pg, at least about 10 pg, at least about 100 pg, at least about 1 ng, at least about 10 ng, at least about 100 ng, at least about 150 ng, or at least about 200 ng of cell-free nucleic acid molecules. In certain embodiments, the amount is up to about 1 fg, about 10 fg, about 100 fg, about 1 pg, about 10 pg, about 100 pg, about 1 ng, about 10 ng, about 100 ng, about 150 ng, or about 200 ng of cell-free nucleic acid molecules. In some embodiments, the method includes the step of obtaining between about 1 fg and about 200 ng of cell-free nucleic acid molecules from a sample.

[0062] Cell-free nucleic acids typically have a size distribution between approximately 100 and 500 nucleotide lengths, with molecules between approximately 110 and 230 nucleotide lengths accounting for approximately 90% of the molecules in the sample, having a mode of approximately 168 nucleotide lengths, and a second smaller peak in the range between approximately 240 and 440 nucleotide lengths. In certain embodiments, cell-free nucleic acids are approximately 160 to 180 nucleotide lengths, or approximately 320 to 360 nucleotide lengths, or approximately 440 to 480 nucleotide lengths.

[0063] In some embodiments, cell-free nucleic acids are isolated from body fluids via a partitioning step in which the cell-free nucleic acids found in solution are separated from intact cells and other insoluble components of the body fluid. In some of these embodiments, the partitioning involves techniques such as centrifugation or filtration. Alternatively, cells in the body fluid are lysed, and the cell-free and cellular nucleic acids are processed together. Generally, after buffer addition and washing steps, the cell-free nucleic acids are precipitated with, for example, alcohol. In certain embodiments, a further purification step, such as a silica-based column, is used to remove impurities or salts. For example, nonspecific bulk carrier nucleic acids are added via the reaction as needed, in certain aspects of the exemplary procedure, e.g., to optimize yield. After such processing, the sample typically contains various forms of nucleic acids, including double-stranded DNA, single-stranded DNA, and / or single-stranded RNA. If necessary, single-stranded DNA and / or single-stranded RNA are converted to double-stranded form for inclusion in subsequent processing and analysis steps. Further details relating to cfDNA distribution and related analyses of epigenetic modifications, which are adapted as necessary for use in carrying out the methods disclosed herein, are described, for example, in WO2018 / 119452, application filed December 22, 2017, which is incorporated by reference. b. Nucleic acid tags

[0064] In certain embodiments, a tag providing a molecular identifier or barcode is incorporated into or attached to the adapter by means of chemical synthesis, ligation, or overlap extension PCR, among other methods. In some embodiments, the assignment of unique or non-unique identifiers or molecular barcodes in the reaction utilizes the methods and systems described herein, for example, in accordance with U.S. Patent Applications Nos. 20010053519, 20030152490, 20110160078 and U.S. Patents Nos. 6,582,908, 7,537,898, and 9,598,731, which are incorporated by reference, respectively.

[0065] The tags are ligated to the sample nucleic acids, either randomly or non-randomly. In some embodiments, the tags are introduced in an expected ratio of identifiers (e.g., unique and / or non-unique barcode combinations) to the microwells. For example, identifiers may be loaded so that more identifiers than approximately 1, 2, 3, 4, 5, 6, 7, 8, 9, 10, 20, 50, 100, 500, 1000, 5000, 10000, 50,000, 100,000, 500,000, 1,000,000, 10,000,000, 50,000,000 or 1,000,000,000 are loaded per genome sample. In some embodiments, identifiers are loaded such that approximately 2, 3, 4, 5, 6, 7, 8, 9, 10, 20, 50, 100, 500, 1000, 5000, 10000, 50,000, 100,000, 500,000, 1,000,000, 10,000,000, 50,000,000, or fewer than 1,000,000,000 identifiers are loaded per genome sample. In a particular embodiment, the average number of identifiers loaded per sample genome is approximately 1, 2, 3, 4, 5, 6, 7, 8, 9, 10, 20, 50, 100, 500, 1000, 5000, 10000, 50,000, 100,000, 500,000, 1,000,000, 10,000,000, 50,000,000, or 1 per genome sample. Identifiers are either less than 1,000,000,000, or about 1, 2, 3, 4, 5, 6, 7, 8, 9, 10, 20, 50, 100, 500, 1000, 5000, 100,000, 500,000, 1,000,000, 10,000,000, 50,000,000, or more than 1,000,000,000. Identifiers are generally unique or non-unique.

[0066] One exemplary form uses approximately 2 to 1,000,000 different tags ligated to both ends of a target nucleic acid molecule, or approximately 5 to 150 different tags, or approximately 20 to 50 different tags. For 20 to 50 × 20 to 50 tags, a total of 400 to 2,500 tags are created. Such a number of tags is typically sufficient to ensure that different molecules with the same start and stop points receive different combinations of tags (e.g., at least 94%, 99.5%, 99.99%, 99.999%).

[0067] In some embodiments, the identifier is an oligonucleotide of a predetermined, random, or semi-random sequence. In other embodiments, multiple barcodes may be used such that the barcodes are not necessarily unique to one another. In these embodiments, the barcodes are generally attached to individual molecules (e.g., by ligation or PCR amplification) so that the combination of the barcode and the sequence to which it can be attached creates a unique sequence that can be individually tracked. As described herein, the detection of non-uniquely tagged barcodes combined with sequence data from the beginning (start) and end (stop) portions of the sequence reading data typically enables the assignment of a unique identity to a particular molecule. The length or number of base pairs of individual sequence reading data is also used, if necessary, to assign a unique identity to a given molecule. Thus, as described herein, a fragment from a single strand of nucleic acid to which a unique identity has been assigned may enable the subsequent identification of fragments from the parent and / or complementary strands.

[0068] In some embodiments, nucleic acid molecules (from a polynucleotide sample) may be tagged with a sample index and / or molecular barcode (commonly referred to as a “tag”). The tag may be incorporated into an adapter or otherwise bound to the adapter by, among other methods, chemical synthesis, ligation (e.g., blunt-end ligation or sticky-end ligation), or overlap-extension polymerase chain reaction (PCR). Such an adapter may ultimately be bound to a target nucleic acid molecule. In other embodiments, one or more rounds of amplification cycles (e.g., PCR amplification) are generally applied to introduce the sample index to the nucleic acid molecule using conventional nucleic acid amplification methods. Amplification may be carried out in one or more reaction mixtures (e.g., multiple microwells in an array). The molecular barcode and / or sample index may be introduced simultaneously or in any sequential order. In some embodiments, the molecular barcode and / or sample index are introduced before and / or after the sequence capture step is performed. In some embodiments, only the molecular barcode is introduced before probe capture, and the sample index is introduced after the sequence capture step is performed. In some embodiments, both the molecular barcode and the sample index are introduced before the probe-based capture step is performed. In some embodiments, the sample index is introduced after the sequence capture step is performed. In some embodiments, the molecular barcode is incorporated into a nucleic acid molecule (e.g., a cfDNA molecule) in the sample via an adapter through ligation (e.g., blunt-end ligation or sticky-end ligation). In some embodiments, the sample index is incorporated into a nucleic acid molecule (e.g., a cfDNA molecule) in the sample via overlap extension polymerase chain reaction (PCR). Typically, the sequence capture protocol involves introducing a single-stranded nucleic acid molecule complementary to a targeted nucleic acid sequence, e.g., the coding sequence of a genomic region, where mutations in such region are associated with a cancer type.

[0069] In some embodiments, the tag may be located at one or both ends of the sample nucleic acid molecule. In some embodiments, the tag is an oligonucleotide with a predetermined, random, or semi-random sequence. In some embodiments, the tag may be less than approximately 500, 200, 100, 50, 20, 10, 9, 8, 7, 6, 5, 4, 3, 2, or 1 nucleotide length. The tag may be ligated to the sample nucleic acid randomly or non-randomly.

[0070] In some embodiments, each sample is uniquely tagged with a sample index or a combination of sample indices. In some embodiments, each nucleic acid molecule in a sample or subsample is uniquely tagged with a molecular barcode or a combination of molecular barcodes. In other embodiments, multiple molecular barcodes may be used such that the molecular barcodes are not necessarily unique to one another (e.g., non-unique molecular barcodes). In these embodiments, molecular barcodes are generally bound to individual molecules (e.g., by ligation) such that the combination of molecular barcode and the sequence to which it can be bound creates a unique sequence that can be individually tracked. In some embodiments, techniques for distinguishing true genomic alterations from technical errors may be used, as described in Lee, et al., "Accurate Detection of Rare Mutant Alleles by Target Base-Specific Cleavage with the CRISPR / Cas9 System," ACS Synth. Biol. 2021, 10, 6, 1451-1464, May 19, 2021, which is incorporated entirely by reference herein. The detection of non-unique molecular barcodes, combined with endogenous sequence information (e.g., the starting and / or stopping genome locations / positions corresponding to the sequence of the original nucleic acid molecule in the sample, the start and stop genome locations corresponding to the sequence of the original nucleic acid molecule in the sample, the starting and stop genome locations / positions of the sequence read data mapped to the reference sequence, the start and stop genome locations of the sequence read data mapped to the reference sequence, the subsequences of the sequence read data at one or both ends, the length of the sequence read data, and / or the length of the original nucleic acid molecule in the sample), typically allows for the assignment of a unique identity to a particular molecule. In some embodiments, the starting region includes the first 1, first 2, first 5, first 10, first 15, first 20, first 25, first 30 or at least the first 30 base positions at the 5' end of the sequenced read data aligned to the reference sequence.In some embodiments, the terminal region comprises the last 1, last 2, last 5, last 10, last 15, last 20, last 25, last 30, or at least the last 30 base positions at the 3' end of sequencing read data that aligns to a reference sequence. The length of individual sequence read data or the number of base pairs is also optionally used to assign a unique identity to a given molecule. Thereby, as described herein, fragments from a single strand of a nucleic acid assigned a unique identity may enable subsequent identification of fragments from the parent strand and / or the complementary strand.

[0071] In certain specific embodiments, the number of distinct tags used to uniquely identify the number z of molecules in a class is 2 * z, 3 * z, 4 * z, 5 * z, 6 * z, 7 * z, 8 * z, 9 * z, 10 * z, 11 * z, 12 * z, 13 * z, 14 * z, 15 * z, 16 * z, 17 * z, 18 * z, 19 * z, 20 * z or 100 * z (e.g., the lower limit) of any of these, and 100,000 * z, 10,000 * z, 1000 * z or 100 *z can be any of the upper bounds (e.g., the upper limit). In some embodiments, molecular barcodes are introduced in an expected ratio of a set of identifiers (e.g., combinations of unique or non-unique molecular barcodes) to the molecules in the sample. One exemplary form uses about 2 to about 1,000,000 different molecular barcode sequences, or about 5 to about 150 different molecular barcode sequences, or about 20 to about 50 different molecular barcode sequences, ligated to both ends of the target molecule. Alternatively, about 25 to about 1,000,000 different molecular barcode sequences may be used. For example, 20 to 50 × 20 to 50 molecular barcode sequences may be used (i.e., one of the 20 to 50 different molecular barcode sequences may be ligated to each end of the target molecule). Such a number of identifiers is typically sufficient to ensure that different molecules having the same starting and ending points have a high probability (e.g., at least 94%, 99.5%, 99.99%, or 99.999%) of receiving different combinations of identifiers. In some embodiments, about 80%, 90%, 95%, or 99% of molecules have the same combination of molecular barcodes. c. Nucleic acid amplification

[0072] The sample nucleic acid adjacent to the adapter is typically amplified by PCR and other amplification methods using nucleic acid primers that bind to primer binding sites in the adapter adjacent to the DNA molecule to be amplified, as part of the sample collection and preparation pipeline 203. In some embodiments, the amplification method includes an extension, denaturation, and annealing cycle obtained from thermocycling, or can be isothermal, for example, as in transcription-mediated amplification. Other exemplary amplification methods that may be utilized as needed include, among other approaches, ligase chain reaction, strand displacement amplification, nucleic acid sequence-based amplification, and auto-persistent sequence-based replication.

[0073] One or more rounds of amplification cycles are generally applied to introduce molecular tags and / or sample index / tags into nucleic acid molecules using conventional nucleic acid amplification methods. Amplification is typically carried out in one or more reaction mixtures. Molecular tags and sample index / tags are introduced simultaneously or in any sequential order, as necessary. In some embodiments, molecular tags and sample index / tags are introduced before and / or after the sequence capture step is performed. In some embodiments, only molecular tags are introduced before probe capture, and sample index / tags are introduced after the sequence capture step is performed. In certain embodiments, both molecular tags and sample index / tags are introduced before the probe-based capture step is performed. In some embodiments, sample index / tags are introduced after the sequence capture step is performed. Typically, the sequence capture protocol involves introducing a single-stranded nucleic acid molecule complementary to a targeted nucleic acid sequence, e.g., the coding sequence of a genomic region, where mutations in such region are associated with a cancer type. Typically, the amplification reaction produces multiple non-specifically tagged nucleic acid amplicons with molecular tags and sample index / tags, ranging in size from approximately 200 nucleotides (nt) to approximately 700 nt, 250 nt to approximately 350 nt, or approximately 320 nt to approximately 550 nt. In some embodiments, the amplicons have a size of approximately 300 nt. In some embodiments, the amplicons have a size of approximately 500 nt.

[0074] In some embodiments, amplification may be performed before and / or after enrichment. d. Nucleic acid enrichment

[0075] In some embodiments, sequences are enriched as part of a sample collection and preparation pipeline 203 before sequencing the nucleic acids. Enrichment is performed, if necessary, for specific target regions or non-specifically ("target sequences"). In some embodiments, the targeted regions of interest may be enriched using nucleic acid capture probes ("baits") selected for one or more bait set panels using a differential tiling and capture scheme. In some embodiments, the targeted regions of interest may be enriched using CRISPR-mediated enrichment. The differential tiling and capture scheme generally involves differentially tiling (e.g., at different "resolutions") across genomic sections associated with baits subject to a set of constraints (e.g., sequencer constraints, e.g., sequencing load, usefulness of each bait) and using bait sets of different relative concentrations to capture targeted nucleic acids at desired levels for downstream sequencing. These targeted genomic sections of interest may optionally include native or synthetic nucleotide sequences of nucleic acid constructs. In some embodiments, biotin-labeled beads having probes for one or more sections of interest may be used to capture the target sequence, followed, if necessary, by amplification of these sections to enrich the region of interest.

[0076] Sequence capture typically involves the use of oligonucleotide probes that hybridize to a target nucleic acid sequence. In certain embodiments, a probe set strategy involves tiling probes across a section of interest. Such probes may be, for example, about 60 to about 120 nucleotides long. Sets may have depths of about 2×, 3×, 4×, 5×, 6×, 8×, 9×, 10×, 15×, 20×, 50×, or greater. The effectiveness of sequence capture generally depends in part on the length of the sequence in the target molecule that is complementary (or nearly complementary) to the probe sequence.

[0077] In some embodiments, the probe may be designed to be specific to the allele of interest. Therefore, different alleles from the same gene have an equal chance of being captured.

[0078] In some embodiments, amplification (as described above) may be carried out after enrichment. e. Nucleic acid sequencing

[0079] As shown in Figure 2, after extraction and isolation of cfDNA from the sample via the sample collection and preparation pipeline 203, the cfDNA can be sequenced via a sequencing pipeline 205 including one or more sequencing devices 207. If necessary, the adjacent sample nucleic acids are sequenced with or without prior amplification. Sequencing methods or commercially available formats used as needed include, for example, Sanger sequencing, high-throughput sequencing, bisulfite sequencing, pyrosequencing, synthetic sequencing, single-molecule sequencing, nanopore-based sequencing, semiconductor sequencing, ligation sequencing, hybridization sequencing, RNA-Seq (Illumina), Digital Gene Expression (Helicos), next-generation sequencing (NGS), Single Molecule Sequencing by Synthesis (SMSS) (Helicos), massively parallel sequencing, clonal single-molecule array (Solexa), shotgun sequencing, Ion Torrent, Oxford Nanopore, Roche Genia, Maxim-Gilbert sequencing, primer walking, PacBio, SOLiD, and sequencing using Ion Torrent or nanopore platforms. The sequencing reaction may be carried out in various sample processing units, which may include means for processing multiple lanes, multiple channels, multiple wells, or multiple sample sets substantially simultaneously. The sample processing unit may also include multiple sample chambers to enable simultaneous processing of multiple runs.

[0080] A sequencing reaction may be performed on another nucleic acid fragment type or section known to contain the allele of interest. A sequencing reaction may also be performed on any nucleic acid fragment present in the sample. A sequencing reaction may provide sequence coverage of at least approximately 5%, 10%, 15%, 20%, 25%, 30%, 40%, 50%, 60%, 70%, 80%, 90%, 95%, 99%, 99.9%, or 100% of the genome. In other cases, the sequence coverage of the genome may be less than approximately 5%, 10%, 15%, 20%, 25%, 30%, 40%, 50%, 60%, 70%, 80%, 90%, 95%, 99%, 99.9%, or 100% of the genome.

[0081] Simultaneous sequencing reactions can be performed using multiple sequencing techniques. In some embodiments, cell-free polynucleotides are sequenced in at least about 1,000, 2,000, 3,000, 4,000, 5,000, 6,000, 7,000, 8,000, 9,000, 10,000, 50,000, or 100,000 sequencing reactions. In other embodiments, cell-free polynucleotides are sequenced in less than about 1,000, 2,000, 3,000, 4,000, 5,000, 6,000, 7,000, 8,000, 9,000, 10,000, 50,000, or 100,000 sequencing reactions. Sequencing reactions are typically performed sequentially or simultaneously. Subsequent data analysis is generally performed on all or some of the sequencing reactions. In some embodiments, data analysis is performed on at least about 1,000, 2,000, 3,000, 4,000, 5,000, 6,000, 7,000, 8,000, 9,000, 10,000, 50,000, or 100,000 sequencing reactions. In other embodiments, data analysis may be performed on fewer than about 1,000, 2,000, 3,000, 4,000, 5,000, 6,000, 7,000, 8,000, 9,000, 10,000, 50,000, or 100,000 sequencing reactions. An example read depth is about 1,000 to about 50,000 reads per locus (base location).

[0082] In some embodiments, nucleic acid populations are prepared for sequencing by enzymatically forming blunt ends on double-stranded nucleic acids having single-stranded overhangs at one or both ends. In these embodiments, the population is typically treated with an enzyme having 5'-3' DNA polymerase activity and 3'-5' exonuclease activity in the presence of nucleotides (e.g., A, C, G, and T or U). Optionally, exemplary enzymes and their catalytic fragments used include Klenow large fragments and T4 polymerase. At the 5' overhang, the enzyme typically extends the concave 3' end on the opposite strand until it lies flat with the 5' end to produce a blunt end. At the 3' overhang, the enzyme generally digests from the 3' end to the 5' end of the opposite strand, and occasionally beyond the 5' end. If this digestion proceeds beyond the 5' end of the opposite strand, the gap may be filled by an enzyme having the same polymerase activity as used for the 5' overhang. The formation of blunt ends on double-stranded nucleic acids facilitates, for example, adapter binding and subsequent amplification.

[0083] In some embodiments, the nucleic acid population undergoes further processing, such as conversion from single-stranded nucleic acids to double-stranded nucleic acids and / or conversion from RNA to DNA. These forms of nucleic acids are also, if necessary, ligated to adapters and amplified.

[0084] Nucleic acids that undergo the process of forming the blunt ends described above, with or without prior amplification, and other nucleic acids in the sample as needed, may be sequenced to produce sequenced nucleic acids. Sequenced nucleic acids may refer to either the sequence of the nucleic acid (i.e., sequence information) or the nucleic acid whose sequence has been determined. Sequencer may be performed to provide sequence data of individual nucleic acid molecules in the sample, either directly or indirectly, from the consensus sequences of the amplification products of individual nucleic acid molecules in the sample.

[0085] In some embodiments, after blunt end formation, double-stranded nucleic acids with single-stranded overhangs in the sample are ligated at both ends to an adapter containing a barcode, and sequencing determines the nucleic acid sequence as well as the inline barcode introduced by the adapter. The blunt-ended DNA molecule is ligated to the blunt ends of an adapter that is at least partially double-stranded (e.g., a Y-shaped or bell-shaped adapter), if necessary. Alternatively, the blunt ends of the sample nucleic acid and the adapter may be tailed with complementary nucleotides to facilitate ligation (e.g., sticky-end ligation).

[0086] Nucleic acid samples are typically brought into contact with a sufficient number of adapters so that there is a low probability (e.g., <1 or 0.1%) that any two copies of the same nucleic acid will receive the same combination of adapter barcodes from adapters linked at both ends. The use of adapters in this manner allows for the identification of families of nucleic acid sequences linked to the same combination of barcodes, having the same start and stop points on the reference nucleic acid. Such families represent the sequences of the amplified products of the nucleic acids in the sample before amplification. The sequences of family members, when modified by blunt end formation and adapter linking, can be collected to derive consensus nucleotides or complete consensus sequences for the nucleic acid molecules in the original sample. In other words, the nucleotides occupying identified positions in the nucleic acid in the sample are determined to be the consensus of the nucleotides occupying corresponding positions in the family member sequences. Families may include sequences from one or both strands of a double-stranded nucleic acid. If a family member includes sequences from both strands of a double-stranded nucleic acid, the sequences from one strand are converted to their complements for the purpose of collecting all sequences to derive consensus nucleotides or sequences. Some families contain only a single member sequence. In this case, this sequence can be considered the nucleic acid sequence in the sample before amplification. Alternatively, families with only a single member sequence can be excluded from subsequent analysis.

[0087] Further details regarding nucleic acid sequencing, including the forms and applications described herein, are incorporated by reference, for example, in Levy et al., Annual Review of Genomics and Human Genetics, 17: 95-115 (2016), Liu et al., J. of Biomedicine and Biotechnology, Volume 2012, Article ID 251364:1-11 (2012), Voelkerding et al., Clinical Chem., 55: 641-658 (2009), MacLean et al., Nature Rev. Microbiol., 7: 287-296 (2009), and Astier et al., J Am Chem Soc., 128(5):1705-10. (2006), also provided under U.S. Patent Nos. 6,210,891, 6,258,568, 6,833,246, 7,115,400, 6,969,488, 5,912,148, 6,130,073, 7,169,560, 7,282,337, 7,482,120, 7,501,245, 6,818,395, 6,911,345, 7,501,245, 7,329,492, 7,170,050, 7,302,146, 7,313,308 and 7,476,503. f. Sequence determination panel

[0088] To improve the likelihood of detecting a target genomic region, the sequenced DNA section may include a panel of genes or genomic sections containing known genomic regions. Selecting a limited section for sequencing (e.g., a limited panel) can reduce the total sequencing required (e.g., the total amount of nucleotides sequenced).

[0089] The genes included in the panel for sequencing may include fully transcribed regions, promoter regions, enhancer regions, regulatory elements, and / or downstream sequences. In some embodiments, only exons may be included in the panel. The panel may include all exons of a selected gene, or only one or more exons of a selected gene. The panel may include exons from each of several different genes. The panel may include at least one exon from each of several different genes.

[0090] At least one complete exon from each different gene in a panel of genes can be sequenced. In some embodiments, all exons of a gene can be sequenced. A sequenced panel may contain all or some exons from multiple genes. A panel may contain exons from 2 to 100 different genes, 2 to 70 genes, 2 to 50 genes, 2 to 30 genes, 2 to 15 genes, or 2 to 10 genes.

[0091] The selected panel may contain a variable number of exons. In some embodiments, the selected panel may contain all of the gene's exons. The panel may contain 2 to 3000 exons. The panel may contain 2 to 1000 exons. The panel may contain 2 to 500 exons. The panel may contain 2 to 100 exons. The panel may contain 2 to 50 exons. The panel may contain 300 or fewer exons. The panel may contain 200 or fewer exons. The panel may contain 100 or fewer exons. The panel may contain 50 or fewer exons. The panel may contain 40 or fewer exons. The panel may contain 30 or fewer exons. The panel may contain 25 or fewer exons. The panel may contain 20 or fewer exons. The panel may contain 15 or fewer exons. The panel may contain 10 or fewer exons. A panel may contain 9 or fewer exons. A panel may contain 8 or fewer exons. A panel may contain 7 or fewer exons.

[0092] The panel may contain one or more exons from multiple different genes. The panel may contain one or more exons from each of a certain proportion of multiple different genes. The panel may contain at least two exons from each of at least 25%, 50%, 75%, or 90% of different genes. The panel may contain at least three exons from each of at least 25%, 50%, 75%, or 90% of different genes. The panel may contain at least four exons from each of at least 25%, 50%, 75%, or 90% of different genes.

[0093] The size of a sequencing panel can vary. A sequencing panel can be larger or smaller (in terms of nucleotide size) depending on several factors, including, for example, the total amount of nucleotides sequenced for a particular region in the panel or the number of unique molecules sequenced. A sequencing panel may be 5kb–50kb in size. A sequencing panel may be 10kb–30kb in size. A sequencing panel may be 12kb–20kb in size. A sequencing panel may be 12kb–60kb in size. A sequencing panel may be 50kb–10Mb in size. A sequencing panel may be 500kb–5Mb in size. The sequencing panel may be at least 10kb, 12kb, 15kb, 20kb, 25kb, 30kb, 35kb, 40kb, 45kb, 50kb, 60kb, 70kb, 80kb, 90kb, 100kb, 110kb, 120kb, 130kb, 140kb, 150kb, 200kb, 250kb, 300kb, 350kb, 400kb, 450kb, or 500kb. The sequencing panel may be less than 100kb, 90kb, 80kb, 70kb, 60kb, or 50kb. The sequencing panel may be at least 1Mb, 2Mb, 3Mb, 4Mb, 5Mb, 6Mb, 7Mb, 8Mb, 9Mb, or 10Mb in size.

[0094] The panel selected for sequencing may contain at least 1, 5, 10, 15, 20, 25, 30, 40, 50, 60, 80, or 100 genomic locations (e.g., each containing a genomic region of interest). In some cases, genomic locations in the panel are selected if they are relatively small in size. In some cases, the regions in the panel have a size of approximately 10kb or less, approximately 8kb or less, approximately 6kb or less, approximately 5kb or less, approximately 4kb or less, approximately 3kb or less, approximately 2.5kb or less, approximately 2kb or less, approximately 1.5kb or less, or approximately 1kb or less or less. In some cases, genomic locations in the panel have sizes ranging from approximately 0.5kb to 10kb, 0.5kb to 6kb, 1kb to 11kb, 1kb to 15kb, 1kb to 20kb, 0.1kb to 10kb, or 0.2kb to 1kb. For example, regions in the panel may have sizes ranging from approximately 0.1kb to 5kb.

[0095] The panel may include one or more locations containing the target genomic region from each of one or more genes. In some cases, the panel may include one or more locations containing the target genomic region from each of at least 1, 2, 3, 4, 5, 6, 7, 8, 9, 10, 15, 20, 25, 30, 40, 50, or 80 genes. In some cases, the panel may include one or more locations containing the target genomic region from each of at most 1, 2, 3, 4, 5, 6, 7, 8, 9, 10, 15, 20, 25, 30, 40, 50, or 80 genes. In some cases, the panel may include one or more locations containing the target genomic region from each of about 1 to about 80, 1 to about 50, about 3 to about 40, 5 to about 30, or 10 to about 20 different genes.

[0096] The concentration of the probe or bait used in the panel may be increased (2 to 6 ng / μL) to capture more nucleic acid molecules in the sample. The concentration of the probe or bait used in the panel may be at least 2 ng / μL, 3 ng / μL, 4 ng / μL, 5 ng / μL, 6 ng / μL or higher. The probe concentration may be about 2 ng / μL to about 3 ng / μL, about 2 ng / μL to about 4 ng / μL, about 2 ng / μL to about 5 ng / μL, or about 2 ng / μL to about 6 ng / μL. The concentration of the probe or bait used in the panel may be 2 ng / μL or higher to 6 ng / μL or lower. In some cases, this may allow more molecules in the biologic to be analyzed, thereby allowing lower-frequency alleles to be detected.

[0097] In one embodiment, the sequencing pipeline 205 may be used to provide the panel with one or more of the following: whole-genome bisulfite sequencing (WGBS) to examine genome-wide methylation patterns, whole-genome sequencing (WGS), and / or targeted sequencing approaches to examine copy number variants (CNVs) and single nucleotide variants (SNVs).

[0098] g. Sequence analysis pipeline In one embodiment, after sequencing, the sequence read data and any associated data may be stored in the sequence data store 209. The sequence read data may be stored in any format. The sequence data store 209 may be local and / or remote to where the sequencing is performed. As shown in Figure 2, the stored read data may be fed into the sequence analysis pipeline 212. i. Sequence quality control

[0099] The sequence analysis pipeline 212 may include a sequence quality control (QC) component 213 that can filter sequence read data from the laboratory system 102. The sequence QC component 213 may assign a quality score to one or more sequence read data. The quality score may be an indication of whether the sequence read data may be useful in subsequent analysis based on a threshold. In some cases, some sequence read data is neither of sufficient quality nor length to perform the subsequent mapping steps. Sequence read data with a quality score of at least 60%, 70%, 80%, 90%, 95%, 99%, 99.9%, 99.99%, or 99.999% may be filtered out of the sequence read data dataset. In other cases, sequence read data assigned a quality score of at least 90%, 95%, 99%, 99.9%, 99.99%, or 99.999% may be filtered out of the dataset.

[0100] Sequence read data that meet the specified quality score thresholds can be mapped to a reference genome by the sequence QC component 213. After mapping the alignment, the sequence read data can be assigned a mapping score. The mapping score may be an indication of the sequence read data, after being backmapped to the reference sequence, indicating whether each position is uniquely mappable or not. Sequence read data with mapping scores of at least 60%, 70%, 80%, 90%, 95%, 99%, 99.9%, 99.99%, or 99.999% can be filtered out of the dataset. In other cases, sequence read data assigned mapping scores of less than 90%, 95%, 99%, 99.9%, 99.99%, or 99.999% can be filtered out of the dataset. ii. Preprocessor

[0101] The preprocessor 214 can read / receive data from the analysis data store 218. For example, the preprocessor 214 can read / receive data showing multiple known allele sequences, multiple test sequence read data, and / or multiple decoy sequences. The preprocessor 214 can also be configured to read sequence data from another source (e.g., an external source).

[0102] The preprocessor 214 may be configured to split a known allele sequence into multiple k-mer sequences. In other examples, k may be about 25 to about 250. For example, k may be 135 or 140. In one embodiment, k may be 125 to 175 nucleotides long, 130 to 160 nucleotides long, 135 to 155 nucleotides long, or 140 to 150 nucleotides long. In one embodiment, k may be 140, 141, 142, 143, 144, or 145 nucleotides long.

[0103] The preprocessor 214 may create a database containing k-mar sequences and further data. The preprocessor 214 may also create a data structure containing k-mar sequences and further data. The data structure may be, for example, a table or a flat file. iii. Alignment Components

[0104] The alignment component 215 can read / receive data from the analysis data store 218. For example, the alignment component 215 can read / receive data showing multiple known allele sequences, k-mar sequences generated from multiple known allele sequences, multiple test sequence reading data, and / or multiple decoy sequences.

[0105] In various embodiments, the alignment component 215 may be configured to align test sequence reading data to a reference sequence or another test sequence reading data. The alignment component 215 may be configured to align test sequence reading data to one or more k-mar sequences generated from a plurality of known allele sequences. The alignment component 215 may be configured to align test sequence reading data (e.g., pairs) to one or more decoy sequences.

[0106] The alignment score is a score indicating the similarity between two arrays determined using an alignment method. In some implementations, the alignment score accounts for the number of edits (e.g., deletions, insertions, and substitutions of characters in a string). In some implementations, the alignment score accounts for the number of matches. In some implementations, the alignment score accounts for both the number of matches and the number of edits. In some implementations, the number of matches and edits are weighted equally for the alignment score. For example, the alignment score may be calculated as number of matches - number of insertions - number of deletions - number of substitutions. In other implementations, the number of matches and edits may be weighted differently. For example, the alignment score may be calculated as number of matches × 5 - number of insertions × 4 - number of deletions × 4 - number of substitutions × 6.

[0107] Pairwise alignment generally involves placing one sequence along a portion of a target, introducing gaps according to an algorithm, scoring how well the two sequences match, and preferably repeating them at various positions along a reference. The best-scoring match is considered an alignment and indicates an inference of homology between the aligned portions of the sequences. In some embodiments, scoring the alignment of pairs of nucleic acid sequences involves assigning values ​​for substitutions and indels. When individual bases are aligned, matches or mismatches contribute to the alignment score by the amount of substitution probability, which may be, for example, 1 for matches and -0.33 for mismatches. Indels are subtracted from the alignment score by the amount of a gap penalty, which may be, for example, -1. The gap penalty and substitution probability may be based on empirical knowledge or a priori assumptions about how sequences evolve. Their values ​​affect the resulting alignment. In particular, the relationship between the gap penalty and substitution probability affects whether substitutions or indels are preferred in the resulting alignment.

[0108] For example, the alignment component 215 may utilize the Burrows-Wheeler Aligner (BWA). Generally, the length of the test sequence read data may be substantially less than the length of the k-mar sequence generated from multiple known allele sequences. The test sequence read data and the k-mar sequence may contain sequences of symbols. The alignment of the test sequence read data and the k-mar sequence may contain a limited number of mismatches between the symbols of the test sequence read data and the symbols of the k-mar sequence. Generally, the test sequence read data may be aligned to a portion of the k-mar sequence to minimize the number of mismatches between the test sequence read data and the k-mar sequence.

[0109] In certain embodiments, the symbols in the test sequence reading data and k-mar sequences may indicate the composition of a biomolecule. For example, the symbols may correspond to the identity of a nucleotide in a nucleic acid, such as RNA or DNA. In some embodiments, the symbols may have a direct correlation with these subcomponents of the biomolecule. For example, each symbol may represent a single base of a polynucleotide. In other embodiments, each symbol may represent two or more adjacent subcomponents of a biomolecule, such as two adjacent bases of a polynucleotide. Furthermore, the symbols may represent overlapping sets of adjacent subcomponents or distinctly different sets of adjacent subcomponents. For example, if each symbol represents two adjacent bases of a polynucleotide, two adjacent symbols representing an overlapping set may correspond to three bases in the polynucleotide sequence, while two adjacent symbols representing distinctly different sets may represent a sequence of four bases. Furthermore, the symbols may directly correspond to subcomponents, such as nucleotides, or to color calls or other indirect measures of subcomponents. For example, the symbols may correspond to incorporation or non-incorporation for a particular nucleotide flow.

[0110] In one embodiment, the alignment component 215 may be configured to determine test sequence reading data that has the same or substantially the same alignment as one or more k-mar sequences.

[0111] Two nucleic acid sequences or polypeptide sequences are said to be “identical” if the sequences of nucleotides or amino acid residues in each of the two sequences are the same when aligned to obtain the maximum response as described herein. The terms “identical” or “percent “identity” refer to two or more sequences or subsequences that are the same or have a specified percentage of identical amino acid residues or nucleotides when compared and aligned to obtain the maximum response across a comparison window, as measured using one of the following sequence comparison algorithms or by manual alignment and visual inspection. The term “substantially identical,” when used in the context of two nucleic acid sequences or polypeptides, refers to sequences that have at least 50% sequence identity with a reference sequence. Percent identity can be any integer between 50% and 100%. Some embodiments include, for example, using a program described herein, BLAST, to compare with a reference sequence, at least: 50%, 55%, 60%, 65%, 70%, 75%, 80%, 85%, 90%, 91%, 92%, 93%, 94%, 95%, 96%, 97%, 98%, or 99%.

[0112] For sequence comparison, typically one sequence acts as the reference sequence against which the test sequence is compared. When using a sequence comparison algorithm, the test and reference sequences are input into a computer, subsequence coordinates are specified, and, if necessary, sequence algorithm program parameters are specified. Default program parameters may be used, or alternative parameters may be specified. The sequence comparison algorithm then calculates the percent sequence identity of the test sequence compared to the reference sequence based on the program parameters.

[0113] When used herein, "comparison window" refers to a segment of consecutive positions, selected from the group consisting of 20 to 600, typically about 50 to about 200, and more typically about 100 to about 150, at which point the same number of consecutive positions of the reference sequence and the sequence can be compared after the two sequences have been optimally aligned. Methods for aligning sequences for comparison are well known in the art. Optimal alignment of sequences for comparison can be performed, for example, by the local homology algorithm of Smith & Waterman, Adv. Appl. Math. 2:482 (1981), by the homology alignment algorithm of Needleman & Wunsch, J. Mol. Biol. 48:443 (1970), by the similarity search method of Pearson & Lipman, Proc. Nat'l. Acad. Sci. USA 85:2444 (1988), or by computerized implementations of these algorithms (GAP, BESTFIT, FASTA, and TFASTA in the Wisconsin Genetics Software Package, Genetics Computer Group, 575 Science Dr., Madison, Wis.).

[0114] Suitable algorithms for determining percent sequence identity and sequence similarity are the BLAST and BLAST 2.0 algorithms, described in Altschul et al. (1990) J. Mol. Biol. 215: 403-410 and Altschul et al. (1977) Nucleic Acids Res. 25: 3389-3402, respectively. Software for performing BLAST analysis is publicly available through the National Center for Biotechnology Information (NCBI) website. The algorithm involves initially identifying high-scoring sequence pairs (HSPs) by identifying short words of length W in the query sequence that match or satisfy a certain positive threshold score T when aligned with words of the same length in the database sequence. T is called the neighbor word score threshold (Altschul et al., above). These initial neighbor word hits act as seeds to initiate a search for longer HSPs that contain them. Word hits are then extended in both directions along each sequence as long as the cumulative alignment score can increase. For nucleotide sequences, the cumulative score is calculated using parameters M (reward score for matching residue pairs; always > 0) and N (penalty score for mismatched residues; always < 0). For amino acid sequences, a scoring matrix is ​​used to calculate the cumulative score. Word hit extension in each direction is stopped if: the cumulative alignment score falls by a factor X from its maximum achieved value; the cumulative score becomes zero or less due to the accumulation of one or more negative-scoring residue alignments; or the end of either sequence is reached. The BLAST algorithm parameters W, T, and X determine the sensitivity and speed of the alignment.The BLASTN program (for nucleotide sequences) uses a word size (W) of 28, an expected value (E) of 10, M=1, N=-2, and comparison of both strands by default. For amino acid sequences, the BLASTP program uses a word size (W) of 3, an expected value (E) of 10, and the BLOSUM62 scoring matrix by default (see Henikoff & Henikoff, Proc. Natl. Acad. Sci. USA 89:10915 (1989)).

[0115] The BLAST algorithm also performs a statistical analysis of the similarity between two sequences (see, for example, Karlin & Altschul, Proc. Nat'l. Acad. Sci. USA 90:5873-5787 (1993)). One measure of similarity provided by the BLAST algorithm is the smallest sum probability (P(N)), which represents the probability that a match between two nucleotide or amino acid sequences occurs by chance. For example, if the smallest sum probability in the comparison of the test nucleic acid to the reference nucleic acid is less than about 0.01, more preferably less than about 10 to 5, and most preferably less than about 10 to 20, the nucleic acid is considered similar to the reference sequence.

[0116] Nucleic acid or protein sequences that are substantially identical to a reference sequence include “conservatively modified variants.” For a given nucleic acid sequence, a conservatively modified variant refers to a nucleic acid that codes for the same or essentially identical amino acid sequence, or a case where the nucleic acid does not code for an amino acid sequence that would result in an essentially identical sequence. Due to the degeneracy of the genetic code, numerous functionally identical nucleic acids code for any given protein. For example, the codons GCA, GCC, GCG, and GCU all code for the amino acid alanine. Therefore, at all positions where alanine is identified by a codon, the codon can be changed to one of the corresponding codons described without altering the coded polypeptide. Such nucleic acid variations are “silent variations,” a type of conservatively modified variation. All nucleic acid sequences in this specification that code for polypeptides also describe all possible silent variations of the nucleic acid. Those skilled in the art will recognize that each codon in a nucleic acid (except AUG, which is usually the sole codon for methionine) can be modified to obtain a functionally identical molecule. Therefore, each silent variation of a polypeptide-coding nucleic acid is implicit in each described sequence.

[0117] When the alignment component 215 aligns multiple test sequence readings to one or more k-mers generated from known allele sequences, a list of test sequence readings aligned to the (supported) k-mer sequence of that allele may be generated for each allele. In one embodiment, only test sequence readings that align identically to the k-mer sequence (e.g., no mismatches and no indels) may be included in the list. In another embodiment, only test sequence readings that align substantially identically to the k-mer sequence (e.g., at least one mismatch and / or at least one indel) may be included in the list. In one embodiment, the alignment component may discard actual alignments. In another embodiment, test sequence readings may be aligned (identically or substantially identically) to multiple alleles. Each test sequence reading may be associated with a test sequence reading identifier. Thus, for each allele, a list of test sequence reading identifiers associated with supporting test sequence readings may be generated. A list of test sequence readings aligned to decoy sequences may also be generated. In one embodiment, the list includes only test sequence read data that align identically to the decoy sequence (e.g., no mismatches and no indels). In another embodiment, the list includes only test sequence read data that align substantially identically to the decoy sequence (e.g., at least one mismatch and / or at least one indel). The alignment component 215 may be configured to discard any test sequence read data that aligns with the decoy sequence without mismatches and without indels. iv. Superset Components

[0118] The cluster component 216 can read / receive data from the analysis data store 218. For example, the cluster component 216 can read / receive data showing multiple known allele sequences, k-mar sequences generated from multiple known allele sequences, multiple test sequence reading data, and results from the alignment component 215.

[0119] In one embodiment, one or more supersets of a plurality of known allele sequences can be computationally generated by constructing one or more graph data structures. The graph data structures may include nodes (also called vertices) representing known allele sequences, and edges connecting the nodes, indicating that the supporting read data of one node is a subset of the supporting read data of other nodes. The construction of the graph data structures can be parallelized, given the computationally intensive nature of such constructions.

[0120] In one embodiment, the graph data structure is stored in a memory subsystem (e.g., Figure 2, memory 222) which may contain pointers to identify the physical location in memory 222 where each vertex is stored. Typically, each node in the graph data structure represents an element in a set, while edges represent relationships between elements. Graph data structures can include directed graphs, trees, directed acyclic graphs (DAGs), and the like. A directed graph is a graph in which edges have a direction. A tree is one type of directed graph data structure, having a root node and several further nodes, each of which is either an internal node or a leaf node. The root node and internal nodes each have one or more “child” nodes, each referred to as the “parent” of its child nodes. Leaf nodes have no child nodes. Edges in a tree conventionally swirl from parent to child. In a tree, a node has exactly one parent. A generalization of trees, known as a directed acyclic graph (DAG), allows a node to have multiple parents, but does not allow edges to form cycles.

[0121] In one embodiment, the graph data structure (superset) may represent a Hasse diagram. For a given locus of a chromosome, alleles may be selected based on the number of supporting test sequence reads. The graph data structure may be constructed by determining the first allele associated with the highest number of supporting test sequence reads. The first allele may form the basis (e.g., top-level node) of the graph data structure. The supporting test sequence reads of the first allele may define a set of supporting test sequence reads. Further alleles may be added to the graph data structure if a given allele is associated with supporting test sequence reads that are themselves a subset of the set of supporting test sequence reads of the first allele. Alleles not included in the allele superset of the first allele may be used in a similar manner to construct one or more further allele supersets. For example, a given allele may have the highest number of supporting test sequence reads, and each supporting test sequence read may be associated with a test sequence read identifier. In this case, the set may consist of test sequence data identifiers for supporting test sequence reads for the first allele. In a basic example, the first allele may be supported by test sequence reads with identifiers "1", "2", "3", and "4". The set of test supporting reads can be represented as A={1,2,3,4}. The power set P(A) of A is the set of all subsets of A. For A={1,2,3,4}: P(A)={φ,{1},{2},{3},{4},{1,2},{1,3},{1,4},{2,3},{2,4},{3,4},{1,2,3},{1,2,4},{1,3,4},{2,3,4},{1,2,3,4}}.

[0122] In some embodiments, a graph data structure (e.g., representing a superset) is stored in a memory subsystem (e.g., Figure 2, memory 222) which may contain pointers to locate the physical location in memory 222 where each vertex is stored, using an adjacency technique. In some embodiments, the graph data structure is stored in memory 222 using an adjacency list. In some embodiments, an adjacency list exists for each vertex.

[0123] In some embodiments, high-speed random access is supported, and graph object storage is implemented with index-free adjacency, in which all elements contain direct pointers to their neighboring elements, eliminating the need for index lookups and making traversal very fast. Index-free adjacency is another example of low-level or hardware-level memory references for data retrieval. Specifically, index-free adjacency can be implemented such that the pointers contained within an element are references to physical locations in memory.

[0124] Technical implementations using physical memory addressing, such as native pointers, allow access to and use of data in such a lightweight manner without requiring separate index tables or other intervening search steps, so that the capabilities of a given computer, e.g., any modern consumer-grade desktop computer, can be extended to enable full manipulation of genome-scale graphs (e.g., graph data structures representing a superset of known allele sequences). Therefore, storing graph elements (e.g., nodes and edges) using a library of objects along with native pointers or other implementations that provide index-free adjacency does indeed improve the ability of this technique to provide storage, retrieval, and alignment of sequence information, because it uses the computer's physical memory in a particular way. v. Allele callers - Selection of alleles from a superset

[0125] The allele caller 217 can read / receive data from the analysis data store 218. For example, the allele caller 217 can read / receive data showing multiple known allele sequences, k-mar sequences generated from multiple known allele sequences, multiple test sequence reading data, results from the alignment component 215, and / or one or more graph data structures (supersets) generated by the cluster component 216.

[0126] The allele caller 217 may be configured to determine the allele type for a given allele. The allele caller 217 may be configured to classify alleles based on one or more graph data structures (supersets).

[0127] In one embodiment, if only one graph data structure (superset) is constructed (for example, all known allele sequences are contained within a single graph data structure), then the allele associated with the root node of the graph data structure (the first allele) can be classified as an allele located at a chromosomal locus (for example, a haploid locus).

[0128] In one embodiment, when multiple graph data structures (supersets) are constructed, an allele (first allele) associated with the root node of two supersets having distinctly different cumulative maximum numbers of supporting test sequence read data can be classified as an allele located at a chromosomal locus (e.g., a diploid locus).

[0129] A set operation may be performed on a combination of root nodes to determine two root nodes that have distinctly different cumulative maximum numbers of supporting test sequence reads. In one embodiment, a merged set operation (U) may be used. vi. Variant Caller

[0130] The variant caller 219 can read / receive data from the analysis data store 218. For example, the variant caller 219 can read / receive data showing multiple sequence readings. The variant caller 219 can read test sequence readings that align to a decoy sequence and to a known allele having at least one mismatch and / or at least one indel, and have a higher alignment score to the known allele. The test sequence readings can be analyzed to determine one or more variants. Variants may include, for example, single nucleotide variants (SNVs), indels, fusions, and / or copy number variations. Any known technique for variant calling can be used. In one embodiment, nucleotide variations in a sequenced nucleic acid may be determined by comparing the sequenced nucleic acid with a reference sequence. The reference sequence is often a known sequence from a subject, e.g., a known whole or partial genome sequence (e.g., the whole genome sequence of a human subject). The reference sequence may be, for example, hG19 or hG38. The sequenced nucleic acids, as described above, may represent a consensus of sequences directly determined for nucleic acids in a sample, or sequences of amplified products of such nucleic acids. Comparisons may be performed at one or more designated positions on the reference sequence. If each sequence is maximally aligned, a subset of sequenced nucleic acids containing the designated positions and corresponding positions on the reference sequence can be identified. Within such a subset, it may be determined which sequenced nucleic acids, if present, contain nucleotide variations at the designated positions, the length of a given cfDNA fragment based on where its endpoints (i.e., its 5' and 3' terminal nucleotides) map to the reference sequence, the offset of the midpoint of the given cfDNA fragment from the midpoint of the genomic region in the cfDNA fragment, and, if necessary, which contain the reference nucleotide (i.e., the same as in the reference sequence), if present. If the number of sequenced nucleic acids in the subset containing nucleotide variants exceeds a selected threshold, the variant nucleotides may be called at the designated positions.Among other possibilities, the threshold may be a simple number within a subset containing nucleotide variants, e.g., at least 1, 2, 3, 4, 5, 6, 7, 9, or 10 sequenced nucleic acids, or a ratio of sequenced nucleic acids within a subset containing nucleotide variants, e.g., at least 0.5, 1, 2, 3, 4, 5, 10, 15, or 20. The comparison may be repeated for any designated position of interest in the reference sequence. Occasionally, the comparison may be performed for designated positions occupying at least approximately 20, 100, 200, or 300 consecutive positions on the reference sequence, e.g., approximately 20–500 or approximately 50–300 consecutive positions.

[0131] Any data analyzed, determined, and / or output by the sequence analysis pipeline 212 may be stored in the analysis data store 218. Generally speaking, the processor 220 may implement (and be programmed by) various components of the sequence analysis pipeline 212, such as the sequence quality control component 213, the preprocessor 214, the alignment component 215, the cluster component 216, the allele caller 217, the variant caller 219, and / or other components. Alternatively, it should be noted that these components of the sequence analysis pipeline 212 may include hardware modules. Although shown separately for simplicity, one or more of the various components or instructions, such as the sequence quality control component 213, the preprocessor 214, the alignment component 215, the cluster component 216, the allele caller 217, and / or the variant caller 219, may be integrated with each other.

[0132] Computer system 210 may exchange data with computer system 224 using network 223. For example, computer system 224 may read data from analytics data store 218. Computer system 224 may be configured to determine / classify alleles present at a particular locus.

[0133] Determining known alleles present at one or more loci for one or more gene loci based on the number of sequence reading data aligned to each known allele sequence may include determining one or more known allele sequences with the highest number of aligned sequence reading data. Determining known alleles present at one or more loci for one or more gene loci based on the number of sequence reading data families aligned to each known allele sequence may include determining one or more known allele sequences with the highest number of aligned sequence reading data families.

[0134] Generating germline alignments of multiple pairs of sequence read data to multiple known allele sequences may include, based on the germline alignment, determining, for each pair of sequence read data, one or more known allele sequences that align without mismatches or indels.

[0135] Generating decoy alignments of multiple pairs of sequence read data to multiple decoy allele sequences may include, based on the decoy alignments, determining, for each pair of sequence read data, one or more decoy allele sequences that align without mismatch or indel, and discarding the pairs of sequence read data.

[0136] Generating decoy alignments of multiple pairs of sequence read data to multiple decoy allele sequences may include, based on the decoy alignments, determining, for each pair of sequence read data, one or more non-human decoy sequences that align without mismatches or indels, and identifying that the multiple pairs of sequence read data originate from a contaminated sample.

[0137] Generating germline alignments of multiple pairs of sequence read data to multiple known allele sequences may include, based on the germline alignment, determining, for each pair of sequence read data, one or more known allele sequences to which each read data in the pair aligns with at least one mismatch or indel, and generating germline alignment scores.

[0138] Generating decoy alignments of multiple pairs of sequence read data to multiple decoy allele sequences may include, based on the decoy alignments, determining one or more decoy allele sequences that align each pair of sequence read data with at least one mismatch or indel, and generating a decoy alignment score.

[0139] Generating germline alignments of multiple pairs of sequence reading data to multiple known allele sequences may involve determining pairs of sequence reading data to align to at least two of the multiple known allele sequences, and selecting one of the at least two known allele sequences.

[0140] Generating decoy alignments of multiple pairs of sequence read data to multiple decoy allele sequences may involve determining pairs of sequence read data to align to at least two of the multiple decoy allele sequences, and selecting one of the at least two decoy allele sequences. IV. Systems and Computer-Readable Media

[0141] The various processing operations and / or methods shown in the drawings may be achieved using some or all of the system components described in detail herein, and in some implementations, the operations may be performed in different sequences, and some operations may be omitted. Further operations may be performed in conjunction with some or all of the operations shown in the flow diagrams provided. One or more operations may be performed simultaneously. Thus, the operations shown (described in more detail herein) are provided as examples and should not be considered limiting.

[0142] The methods of the present invention can be implemented by computer, such that any or all of the operations described herein or in the appended claims, other than the wet chemical steps, can be performed on a suitablely programmed computer. The computer may be a mainframe, personal computer, tablet, smartphone, cloud, online data storage, remote data storage, etc. The computer may be operated in one or more locations.

[0143] Various operations of the method of the present invention may utilize information and / or programs and produce results that are stored on computer-readable media (e.g., hard drives, auxiliary memory, external memory, servers; databases, portable memory devices (e.g., CD-R, DVD, ZIP disk, flash memory card), etc.).

[0144] The disclosure also includes a product for analyzing nucleic acid populations, comprising a machine-readable medium containing one or more programs that, if performed, implement steps of the method of the present invention.

[0145] This disclosure may be implemented in hardware and / or software. For example, different aspects of this disclosure may be implemented in either client-side logic or server-side logic. This disclosure or its components may be embodied in a fixed-media program component containing logical instructions and / or data that, when loaded onto a properly configured computing device, cause the device to operate in accordance with this disclosure. The fixed media containing the logical instructions may be conveyed to the viewer on the fixed media for physical loading onto the viewer's computer, or the fixed media containing the logical instructions may reside on a remote server accessed by the viewer via a communication medium for downloading the program component.

[0146] This disclosure provides a computer control system programmed to implement the methods of this disclosure. Returning to Figure 2, the processor 220 may include a single-core or multi-core processor, or multiple processors for parallel processing. The storage device 222 may include random-access memory, read-only memory, flash memory, hard disks, and / or other types of storage. The computer system 210 may include one or more other systems, as well as a communication interface (e.g., a network adapter) for communicating with peripheral devices, such as caches, other memory, data storage, and / or electronic display adapters. Components of the computer system 210 may communicate with each other via an internal communication bus, such as a motherboard. The storage device 222 may be a data storage unit (or data repository) for storing data. The computer system 210 may be operably connected to a network 223 ("the network") with the help of the communication interface. The network 223 may be the Internet, the Internet and / or an extranet, or an intranet and / or extranet communicating with the Internet. In some cases, the network 223 is a telecommunications and / or data network. Network 223 may include a local area network. Network 23 may include one or more computer servers that can enable distributed computing, such as cloud computing. Network 223 may implement a peer-to-peer network that, in some cases, with the help of computer system 210, can enable devices connected to computer system 220 to act as clients or servers. Computer system 210 may use network 223 to exchange data with computer system 224. For example, computer system 224 may read data from analytics data store 218.

[0147] The processor 220 may execute a sequence of machine-readable instructions that can be embodied in a program or software. Instructions may be stored in a memory location, for example, in a storage device 222. Instructions that can subsequently be programmed or otherwise configured to implement the method of this disclosure may be directed to the processor 220. Examples of operations performed by the processor 220 may include fetching, decoding, executing, and writing back.

[0148] The processor 220 may be part of a circuit, such as an integrated circuit. One or more other components of the system 200 may be included in the circuit. In some cases, the circuit may include an application-specific integrated circuit (ASIC).

[0149] The storage device 222 may store files, such as drivers, libraries, and saved programs. The storage device 222 may also store user data, such as user preferences and user programs. In some cases, the computer system 210 may include one or more additional data storage units located on a remote server that communicates with the computer system 210 externally, for example, via an intranet or the internet.

[0150] Computer system 210 may communicate with one or more remote computer systems via a network. For example, computer system 210 may communicate with a user's remote computer system. Examples of remote computer systems include personal computers (e.g., portable PCs), slate or tablet PCs (e.g., Apple® iPad®, Samsung® Galaxy Tab), telephones, smartphones (e.g., Apple® iPhone®, Android® enabled devices, Blackberry®), or personal digital assistants. Users can access computer system 210 via the network.

[0151] The methods described herein may be implemented by machine-executable code (e.g., a computer processor) stored on an electronic storage location of the computer system 210, for example, on a storage device 222. The machine-executable or machine-readable code may be provided in the form of software (e.g., on a computer-readable medium). During use, the code may be executed by the processor 220. In some cases, the code may be read from the storage device 222 and stored on the storage device 222 for immediate access by the processor 220.

[0152] The code may be pre-compiled and configured for use with a machine having a processor adapted to run the code, or it may be compiled during runtime. The code may be provided in a programming language that can be chosen to allow the code to be executed in a pre-compiled or compiled form.

[0153] Embodiments of the systems and methods provided herein, for example, computer system 210, can be embodied in programming. Various embodiments of the technology can typically be considered “products” or “manufactured products” in the form of machine-executable code and / or related data carried on or embodied on a certain type of machine-readable medium. Machine-executable code can be stored on an electronic storage unit, for example, memory (e.g., read-only memory, random-access memory, flash memory) or a hard disk.

[0154] "Storage" type media may include computers, processors, or their associated modules, or any or all of tangible memory such as various semiconductor memory, tape drives, and disk drives, which can provide non-temporary storage for software programming at any time. All or part of the software may be communicated from time to time over the Internet or various other telecommunication networks. For example, such communication may enable the loading of software from one computer or processor to another, for example, from a management server or host computer to an application server computer platform. Thus, other types of media that can carry software elements include optical waves, radio waves, and electromagnetic waves used, for example, across physical interfaces between local devices, over wired and optical fixed telephone networks, and through various air-links. Physical elements that carry such waves, such as wired or wireless links, optical links, etc., may also be considered media that carry software. As used herein, "medium" may include other types of (intangible) media, unless it is not limited to non-temporary tangible storage media.

[0155] The term “storage” medium, for example, computer or machine “readable medium,” refers to any tangible (e.g., physical) non-transient medium that is involved in providing instructions to a processor for execution.

[0156] Therefore, machine-readable media, such as computer-executable code, can take many forms, including but not limited to tangible storage media, carrier media, or physical transmission media. Non-volatile storage media include any storage devices in any computer, such as optical or magnetic disks, which may be used to implement databases, etc., as shown in the drawings. Volatile storage media include the dynamic memory of such computer platforms, such as main memory. Tangible transmission media include wires, such as coaxial cables, copper wires, and optical fibers, including buses within computer systems. Carrier media can take the form of electrical or electromagnetic signals, or sound or light waves, such as waves generated during radio frequency (RF) and infrared (IR) data communications. Therefore, common forms of computer-readable media include, for example,: floppy disks, flexible disks, hard disks, magnetic tapes, any other magnetic media, CD-ROMs, DVDs or DVD-ROMs, any other optical media, punched card paper tapes, any other physical storage media having a pattern of holes, RAM, ROMs, PROMs and EPROMs, FLASH®-EPROMs, any other memory chips or cartridges, carriers for transporting data or instructions, cables or links for transporting such carriers, or any other media from which a computer can read programming code and / or data. Many of these forms of computer-readable media may be involved in transporting one or more sequences of one or more instructions to a processor for execution.

[0157] The computer system 210 may include, or communicate with, an electronic display 935 that includes a user interface (UI) for, for example, providing reports. Examples of UIs include, but are not limited to, graphical user interfaces (GUIs) and web-based user interfaces.

[0158] The methods and systems of this disclosure may be implemented by one or more algorithms. The algorithms may be implemented by software when executed by the processor 220. [Examples]

[0159] V. Examples (Example 1) Challenges in CYP2D6 genotyping In some cases, allele calling in CYP2D6 may be similar to allele calling in highly homologous genes, such as HLA or KIR. However, in some cases, genotyping CYP2D6 is complicated by several factors, such as the unique tandem structure described and high homology to neighboring regions. Firstly, CYP2D7 is nearly identical to CYP2D6. Secondly, in some cases (for example, * 10.002+ * 36.004) In many cases, the two alleles are nearly identical, and the majority of supporting read data actually support both alleles. Thirdly, in the case of CYP2D6, a filter can remove a read data pair if that pair maps (perfectly) to more than one gene, although filter removal induces false positives due to the fact that, in the usual conventional case, read data belonging to CYP2D7 randomly supports an odd number of CYP2D6 alleles. However, the inability to remove the filter is, * Different from 10 (Perfect match with CYP2D7) *This could result in the inability to fully explain the small amount of read data within the 36 small regions.

[0160] Rather than attempting to solve general problems, these phenomena can be specifically sought out. For example, there are a limited number of known tandem arrangements and complex CNV structures. Furthermore, the precise identification of a tandem arrangement or CNV structure is merely a means to an end: the clinically relevant aspect is the function of the gene (whether it is normal, increased, or decreased). For example, * 17+ * Not 17.001 * Calling 17 does not affect clinical function (in other words, it may be decided not to attempt to identify this particular organization because it does not alter the clinical impact). (Example 2) design mechanism

[0161] The process for detecting complexly organized CYP2D6 alleles involves gene-based filtering, unique read data pairs, and ratios between unique read data pairs.

[0162] (1) A gene-based filter is the name of a logic that removes read data pairs if they map perfectly to more than one gene (e.g., both CYP2D6 and CYP2D7).

[0163] (2) When considering heterozygous genes, the unique read data pairs for two alleles are two sets of read data pairs unique to each allele (this is appropriate because the read data pairs often support both alleles).

[0164] (3) When considering heterozygous genes, the unique ratio is the ratio between the largest and smallest number of unique read data pairs for the two alleles (and therefore it is a real number greater than or equal to 1). If one of the two alleles has no unique read data pairs, this is defined as zero. (Example 3) List of known cases

[0165] Generate a list of known tandem rearrangements and copy number changes. Generate a list of special alleles composed of the list of known tandem rearrangements and copy number changes. For example, if the known tandem rearrangement allele is CYP2D6 * 10.002+CYP2D6 * If the value is 36.004, then CYP2D6 is in the list of special alleles. * Add 36.004. (Example 4) Run kmerizer

[0166] Run the allele caller kmerizer, specifically unfolding gene-based filters (this is the default behavior). In parallel, track all special alleles and remove (i.e., turn off) gene-based filters for supporting read data (this means that read data pairs supporting multiple genes can support special alleles). For example, hybrid * 10.002+ * To identify 36.004, * We need to track 36.004. Turn off gene-based filtering for specific alleles. (Example 5) Match alleles against a known list.

[0167] If one or both of the alleles called by kmerizer match one of the alleles in the list of known cases, check whether one of the combinations is supported by the read data and whether the intrinsic ratio between the two is positive and less than or equal to a user-defined threshold. More specifically, if kmerizer calls a germline event, check whether one or more of the germline alleles could potentially fit into one of the known combinations. For example, * If 10.002 is one of the alleles, * 10.002+ * Check if 36.004 is supported. If the combination is found, call it only if the intrinsic ratio between the two alleles involved is non-zero and <= 10. (Example 6) Performance evaluation

[0168] To evaluate the above performance characteristics, the inventors sequenced several samples from Coriell's cell line. In particular, known CYP2D6 status * 1 and * 10.002+ * Two samples from cell line NA23090 with a value of 36.004. The original algorithm was: * 1 and * Identify 10.002 as the "normal" allele. * Since 10.002 is one of the alleles in the inventors' list, the algorithm is a special allele. * Check for support for 36.004 and also check that this allele is supported. Then the algorithm, * 10.002 and * The algorithm calculates the unique ratio (approximately 4.5) between 36.004 and 36.004. ⇒ The algorithm uses the second allele as * 10.002+ *Call 36.004. Thus, the algorithm accurately identified the case of two hybrids. The algorithm also called the two samples from cell line NA17248 as a tandem arrangement.

[0169] What was expected: * 6 and * 10. Called: * 6 and * 10.002+ * 36.004. (Example 7) clinical function

[0170] While it is possible to accurately identify tandem formations and complex CNV structures in CYP2D6, in some cases, a unambiguous classification may be difficult.

[0171] In the same spirit as kmerizer (which relies on a list of known alleles to call genes), this logic simply matches the status of the sample against a list of known configurations.

[0172] In the same case, even if it is difficult / impossible to accurately classify the allele status of a sample, it is likely possible to accurately predict the clinical function of the alleles in the sample. As shown in Figure 7, the alleles and confirmed clinical function can deteriorate. In such cases, the output from the kmerizer is combined with the output from the CNV caller. (Example 8)

[0173] A method is disclosed that includes the steps of determining a plurality of known allele sequences, determining a plurality of sequence reading data for a target region of a target chromosome, wherein the target region of the chromosome includes one or more loci, aligning the plurality of sequence reading data with a plurality of known allele sequences, determining the number of sequence reading data aligned to each known allele sequence for each known allele sequence of the plurality of known allele sequences based on the alignment, and determining the number of known allele sequences present at one or more loci for one or more loci based on the number of sequence reading data aligned to each known allele sequence.

Claims

1. A method implemented by a computer, A step of receiving multiple pairs of sequence reading data from the target sample; A step of aligning the aforementioned pairs of sequence read data with references; The steps of keeping all possible alignments of the aforementioned pairs of sequence read data; and Steps to call one or more alleles Methods that include...

2. The step of obtaining the sample from the subject; and The step of determining the sequence of the sample to obtain the plurality of pairs of sequence reading data. The method according to claim 1, further comprising:

3. The method according to claim 1, wherein the references include all known alleles of the cytochrome P450 family 2 subfamily D (CYP2D) family of genes.

4. The method according to claim 3, wherein the genes of the CYP2D family include at least CYP2D6, CYP2D7, and CYP2D78P.

5. The method according to claim 3, wherein the references include all known alleles of the cytochrome P450 enzyme.

6. The cytochrome P450 enzymes include at least: CYP2D6, CYP2D7, CYP2D78P, CYP3A4, CYP3A5, CYP1A1, CYP1A2, CYP2C9, CYP2C19, CYP2E1, CYP2F1, CYP2J2, CYP1B1, CYP2A6, CYP2B6, CYP2C8, CYP2C18, CYP2C29, CYP2D8P, CYP2D78, CYP2R1, CYP2S1, CYP2U1, CYP2W1, CYP2Y1, CYP2Y11, CYP2Y12, CYP2Y13, CYP2Y8, C YP2Y9, CYP4A11, CYP4A22, CYP4B1, CYP4F11, CYP4F12, CYP4F22, CYP4V2 , CYP4X1, CYP4Z1, CYP4F2, CYP4F3, CYP4F8, CYP4F12, CYP4F22, CYP4F3, CYP4F8, CYP4F12, CYP4F22, CYP4V2, CYP4X1, CYP4Z1, CYP4F2, CYP4F3, C YP4F8, CYP4F12, CYP4F22, CYP4V2, CYP4X1, CYP4Z1, CYP4F2, CYP4F3, CY P4F8, CYP4F12, CYP4F22, CYP4V2, CYP4X1, CYP4Z1, CYP4F2, CYP4F3, CY P4F8, CYP4F12, CYP4F22, CYP4V2, CYP4X1, CYP4Z1, CYP4F2, CYP4F3, CYP 4F8, CYP4F12, CYP4F22, CYP4V2, CYP4X1, CYP4Z1, CYP4F2, CYP4F3, CYP4 F8, CYP4F12, CYP4F22, CYP4V2, CYP4X1, CYP4Z1, CYP4F2, CYP4F3, CYP4F The method according to claim 5, comprising 8, CYP4F12, CYP4F22, CYP4V2, CYP4X1, CYP4Z1, CYP4F2, CYP4F3, CYP4F8, CYP4F12, CYP4F22, CYP4V2, CYP4X1, CYP4Z1, CYP4F2, CYP4F3, CYP4F8, CYP4F12, CYP4F22, CYP4V2, CYP4X1, CYP4Z1, CYP4F2, CYP4F3, CYP4F8, CYP4F12, CYP4F22, CYP4V2, CYP4X1, CYP4Z1, and CYP4F2.

7. The method according to any one of claims 1 to 6, wherein the reference is constructed by concatenating the sequences of all known alleles of the CYP2D family of genes.

8. The step of generating a list of all possible alignments of the plurality of pairs of sequence read data for the aforementioned reference. The method according to claim 1, further comprising:

9. The method according to claim 8, further comprising the step of sorting all possible alignments into weak alignments and strong alignments.

10. The method according to claim 9, wherein a strong alignment includes a pair of read data that aligns to a single gene, and a weak alignment includes a pair of read data that aligns to different genes.

11. The method according to any one of the claims, wherein the calling step includes (1) determining a combination from the references that best match the allele, and (2) calling a heterozygous allele if the unique ratio between the reading data supporting the first allele and the reading data supporting the second allele does not exceed a set threshold.

12. The method according to claim 11, wherein the unique ratio includes the ratio between the largest and smallest number of unique read data pairs of the two alleles.

13. The method according to claim 11, wherein the intrinsic ratio is non-zero and less than or equal to 10.

14. The method according to claim 12, wherein the unique read data pairs for the two alleles include a set of read data pairs unique to each allele.

15. The method according to any one of the above claims, wherein the aforementioned reference includes at least a portion of chromosome position 22q13.

2.

16. The method according to any one of the above claims, wherein the sample is cell-free DNA (cfDNA).

17. The method according to any one of the above claims, wherein the sample is selected from the group consisting of blood, serum, plasma, bone marrow aspirate, bile, cerebrospinal fluid (CSF), saliva, and urine.

18. The method according to any one of the above claims, wherein the subject is a cancer patient.

19. The aforementioned cancers include biliary tract cancer, bladder cancer, transitional cell carcinoma, urothelial carcinoma, brain cancer, glioma, astrocytoma, breast cancer, metaplastic 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, ocular melanoma, uveal melanoma, gallbladder cancer, gallbladder adenocarcinoma, renal cell carcinoma, clear cell renal cell carcinoma, transitional cell carcinoma, urothelial carcinoma, Wilms' tumor, leukemia, acute lymphoblastic leukemia (ALL), acute myeloid leukemia (AML), chronic lymphocytic leukemia (CLL), chronic myeloid leukemia (CML), chronic myelomonocytic leukemia (CMML), liver cancer, liver The method according to claim 17, selected from the group consisting of 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 cancer (NPC), neuroblastoma, oropharyngeal cancer, oral squamous cell carcinoma, osteosarcoma, ovarian cancer, pancreatic cancer, pancreatic ductal adenocarcinoma, pseudopapillary neoplasm, acinar cell carcinoma, prostate cancer, prostate adenocarcinoma, skin cancer, melanoma, malignant melanoma, cutaneous melanoma, small intestine cancer, gastric cancer, gastric cancer, gastrointestinal stromal tumor (GIST), uterine cancer, and uterine sarcoma.

20. The method according to any one of the above claims, wherein the called allele determines the efficacy and / or toxicity of the drug.

21. The method according to any one of the above claims, wherein the called allele may influence the cancer treatment outcome in the patient.

22. The method according to any one of the claims, wherein the called allele may influence the patient's response to cancer treatment.

23. The method according to any one of the claims, wherein the called alleles guide the selection of treatment for the patient.

24. A method for determining the genotype of a nucleic acid sequence, A step of receiving multiple pairs of sequence reading data from the target sample; A step of aligning the aforementioned pairs of sequence read data with references; A step of keeping all possible alignments of the aforementioned pairs of sequence read data; The steps of categorizing alignments into strong alignments and weak alignments; and Steps to call one or more alleles A method comprising, wherein the nucleic acid sequence includes one or more confounding features.

25. The method according to claim 24, wherein the confounding feature includes at least one tandem arrangement, copy number amplification (CNA) and / or complex CNV structure.

26. The method according to claim 25, wherein the tandem configuration includes duplication of CYP2D6.

27. The aforementioned duplication of CYP2D6 is the allele CYP2D6 ＊ 36 and CYP2D6 ＊ The method according to claim 25, comprising 10 in a tandem configuration.

28. A step of determining multiple sequence reading data pairs of a target region of a target chromosome, wherein the target region of the chromosome includes one or more gene loci, The step of aligning the aforementioned plurality of sequence reading data pairs with a plurality of known allele sequences, Based on the alignment, the step of determining, for each known allele sequence of the plurality of known allele sequences, a strong read data pair that aligns to a single gene and a weak pair that aligns to different genes, A step of determining a combination of strong read data pairs and weak read data pairs from the reference that best matches the allele, A step in which a heterozygous allele is called if the unique ratio between the reading data supporting one allele and the reading data supporting a second allele does not exceed a set threshold. Methods that include...

29. The method according to claim 28, wherein the intrinsic ratio is positive.

30. The method according to claim 28, wherein the intrinsic ratio is non-zero and less than or equal to 10, i.e., 0 < intrinsic ratio ≤ 10.

31. The method according to claim 28, wherein one or more gene loci include at least one tandem arrangement, copy number amplification (CNA), and / or complex CNV structure.

32. The method according to claim 31, wherein the tandem configuration includes duplication of CYP2D6.

Images