Methods and compositions for somatic variant detection

Abstract

Some embodiments of the methods and systems provided herein relate to variant calling from sequence data obtained from a single sample.In some embodiments, somatic variants can be distinguished from germline variants based on the variant allele frequency in the sample and the position in the genome.

Description

Field of the Invention

[0001] Some embodiments of the methods and systems provided herein relate to variant calling from sequence data obtained from a single sample.In some embodiments, somatic variants can be distinguished from germline variants based on the variant allele frequency of the variant in the sample and its position in the genome. Background of the Invention

[0003] DNA mutation is a cause of cancer and is the focus of cancer research and treatment.Next generation sequencing (NGS) is a promising de novo mutation detection technology because modern sequencers can produce huge amounts of reads.In theory, all mutations or variants in a genomic sample, regardless of variant allele frequency (VAF) or genomic region, can be observed given sufficient read depth.However, due to the noise in the reads, it is not trivial to call variants with confidence.Several bioinformatics tools have been developed to reveal variants from sequencing reads, and such a process is usually composed of three components: read processing, mapping and alignment, and variant calling.

[0004] For read processing, read processing tools trim low-quality bases, typically located near the 3' end of the read, as well as exogenous sequences such as sequencing adapters, from the DNA sample. Second, the clean reads are mapped using mapping and alignment tools to determine where the variants may come from in the reference genome, and then aligned base by base. Third, the process of variant calling is used to separate true variants from artifacts derived from library preparation, sample enrichment, sequencing, and mapping / alignment. There is still a need for improved methods for calling variants from sequence data. SUMMARY OF THE INVENTION

[0006] Some embodiments include methods for identifying somatic variants among a plurality of variants, comprising: (a) obtaining a plurality of variants comprising somatic variants and germline variants; (b) applying a database filter to the plurality of variants, comprising: determining a first germline variant among the plurality of variants, wherein each of the first germline variants has an allele count greater than or equal to a threshold allele count in a first variant reference set; (c) applying a neighbor filter to the plurality of variants, comprising: (i) grouping variants of the plurality of variants into a plurality of bins, wherein variants located in the same region of a genome are grouped into the same bin, (ii) determining a database variant among the plurality of variants, wherein the database variant is present in a second variant reference set, and (iii) determining a second germline variant among the plurality of variants, wherein each of the second germline variants has an allele frequency within an approximate range of the allele frequency of at least one database variant that is in the same bin as the second germline variant; and (d) determining a somatic variant among the plurality of variants by removing the identified first and second germline variants from the plurality of variants.

[0007] In some embodiments, (b) and (c) are performed consecutively.

[0008] In some embodiments, (c) is performed before (b).

[0009] In some embodiments, the threshold allele count is 5. In some embodiments, the threshold allele count is 10.

[0010] In some embodiments, the first and second variant reference sets are the same reference set.

[0011] In some embodiments, the first or second variant reference set comprises a database of variants of a plurality of individuals. In some embodiments, the first or second variant reference set comprises at least one database selected from the Genome Aggregation Database (gnomAD) and the 1000 Genomes Database.

[0012] In some embodiments, the same region of the genome is located in the same chromosome. In some embodiments, the same region of the genome is located in the same chromosome arm. In some embodiments, the same region of the genome is located in the same chromosome cell band. In some embodiments, the same region of the genome is located in the 10Mb region.

[0013] In some embodiments, applying the neighbor filter further comprises identifying a second germline variant having an allele frequency greater than or equal to 0.9.

[0014] In some embodiments, applying the neighbor filter further comprises identifying a second germline variant in the plurality of variants, wherein the second germline variant is a database variant present in a second reference set of variants.

[0015] In some embodiments, the approximate range is a range having a maximum and a minimum that differ from the allele frequency of the second germline variant by 0.05.

[0016] In some embodiments, the approximate range is a range having a maximum and a minimum value that are two standard deviations from a binomial distribution of the allele frequency of the second germline variant and is centered around the allele frequency of the second germline variant.

[0017] In some embodiments, the second germline variant has an allele frequency within a threshold close to the allele frequencies of at least five database variants that are in the same bin as the second germline variant. In some embodiments, the second germline variant has an allele frequency within a threshold close to the allele frequencies of at least ten database variants that are in the same bin as the second germline variant.

[0018] In some embodiments, (a) comprises: obtaining sequence data from a biological sample comprising tumor cells. Some embodiments further comprise aligning the sequence data with a reference sequence and identifying variants in the sequence data.

[0019] In some embodiments, the biological sample comprising tumor cells is selected from a serum sample, a stool sample, a blood sample, a tumor sample. In some embodiments, the tumor sample is a fixed sample.

[0020] Some embodiments include a method for determining the tumor mutation load of a tumor, comprising: obtaining sequence data from a biological sample comprising tumor cells; determining a plurality of variants from the sequence data; and determining the number of somatic variants in the plurality of variants according to the method of any of the foregoing embodiments, wherein the number of somatic variants is the tumor mutation load of the tumor.

[0021] Some embodiments include a method of treating a tumor, comprising: determining a tumor having a tumor mutation load greater than or equal to 10 somatic variants according to a method for determining the tumor mutation load of a tumor; and treating the tumor by administering an effective amount of a checkpoint inhibitor.

[0022] In some embodiments, the tumor is selected from the group consisting of a colorectal tumor, a lung tumor, an endometrial tumor, a uterine tumor, a gastric tumor, a melanoma, a breast tumor, a pancreatic tumor, a kidney tumor, a bladder tumor, and a brain tumor.

[0023] In some embodiments, the checkpoint inhibitor is selected from a CTLA-4 inhibitor, a PD-1 inhibitor, and a PD-L1 inhibitor. In some embodiments, the checkpoint inhibitor is selected from ipilimumab, nivolumab, pembrolizumab, spartalizumab, atezolizumab, avelumab, and durvalumab.

[0024] Some embodiments include an electronic system for analyzing genetic variation data, comprising: an informatics module, which runs on a processor and is configured to identify a plurality of variants from sequence data from a biological sample comprising a tumor cell, wherein the plurality of variants comprises somatic variants and germline variants; a database filter module, which is configured to remove a first germline variant from the plurality of variants, wherein each of the first germline variants has an allele count greater than or equal to a threshold allele count in a first variant reference set; a neighbor filter module, which is configured to remove a second germline variant from the plurality of variants, the neighbor filter module comprising: a binning module; submodules configured to return a plurality of bins, each bin containing variants of the plurality of variants located in the same region of the genome; an identification submodule configured to return database variants among the plurality of variants, wherein the database variants are present in a second variant reference set; and a removal submodule configured to remove second germline variants from the plurality of variants, wherein each of the second germline variants has an allele frequency within an approximate range of the allele frequency of at least one database variant that is in the same bin as the second germline variant; and a display module configured to return variants that were not removed from the plurality of variants.

[0025] In some embodiments, the informatics module includes a variant annotation tool.

[0026] In some embodiments, the threshold allele count is 5. In some embodiments, the threshold allele count is 10.

[0027] In some embodiments, the first and second variant reference sets are the same reference set.

[0028] In some embodiments, the first or second variant reference set comprises a database of variants of a plurality of individuals. In some embodiments, the first or second variant reference set comprises at least one database selected from the Genome Aggregation Database (gnomAD) and the 1000 Genomes Database.

[0029] In some embodiments, the same region of the genome is located in the same chromosome. In some embodiments, the same region of the genome is located in the same chromosome arm. In some embodiments, the same region of the genome is located in the same chromosome cell band. In some embodiments, the same region of the genome is located in the 10Mb region.

[0030] In some embodiments, the removing submodule is configured to remove variants having an allele frequency greater than or equal to 0.9 from the plurality of variants.

[0031] In some embodiments, the removing submodule is configured to remove database variants present in the second variant reference set from the plurality of variants.

[0032] In some embodiments, the approximate range is a range having a maximum and a minimum value that differ from the allele frequency of the second germline variant by 0.05.

[0033] In some embodiments, the approximate range is a range having a maximum and a minimum value that are two standard deviations from a binomial distribution of the allele frequency of the second germline variant and centered around the allele frequency of the second germline variant.

[0034] In some embodiments, the second germline variant has an allele frequency within a threshold close to the allele frequencies of at least five database variants that are in the same bin as the second germline variant. In some embodiments, the second germline variant has an allele frequency within a threshold close to the allele frequencies of at least ten database variants that are in the same bin as the second germline variant.

[0035] In some embodiments, the biological sample comprising tumor cells is selected from a serum sample, a stool sample, a blood sample, a tumor sample. In some embodiments, the tumor sample is a fixed sample.

[0036] Some embodiments include a computer-implemented method for identifying a somatic variant among a plurality of variants, comprising: performing any of the foregoing methods.

[0037] Some embodiments include a computer-implemented method for identifying a somatic variant among a plurality of variants, comprising: (a) receiving a plurality of variants from sequence data from a biological sample comprising a tumor cell, the plurality of variants comprising somatic variants and germline variants; (b) applying a database filter to the plurality of variants, comprising: creating a file index for the plurality of variants, searching a first variant reference set with the index to identify a first germline variant in the index, wherein each of the first germline variants has an allele count in the first variant that is greater than or equal to a threshold allele count, and removing the identified first germline variant from the index to create an index of first filtered variants; (c) applying a neighbor filter to the index of first filtered variants, comprising: (i) filtering the first germline variants for different regions of the genome; and (ii) filtering the first germline variants for different regions of the genome. The invention relates to a method for creating a plurality of bins in a first filtered variants domain, (ii) binning variants in the index of first filtered variants, wherein variants located in the same region of the genome are grouped into the same bin, (iii) searching a second reference set of variants with the index of the first filtered variants to identify database variants in the index of the first filtered variants, (iii) generating an index of second germline variants from the index of the first filtered variants by identifying second germline variants, wherein each of the second germline variants has an allele frequency within an approximate range of the allele frequency of at least one database variant, the at least one database variant being in the same bin as the second germline variant, and (iv) removing the identified second germline variants from the index of the first filtered variants to generate an index of somatic variants, thereby identifying somatic variants among a plurality of variants.

[0038] In some embodiments, the threshold allele count is 5. In some embodiments, the threshold allele count is 10.

[0039] In some embodiments, the first and second variant reference sets are the same reference set.

[0040] In some embodiments, the first or second variant reference set comprises a database of variants of a plurality of individuals. In some embodiments, the first or second variant reference set comprises at least one database selected from the Genome Aggregation Database (gnomAD) and the 1000 Genomes Database.

[0041] In some embodiments, the same region of the genome is located in the same chromosome. In some embodiments, the same region of the genome is located in the same chromosome arm. In some embodiments, the same region of the genome is located in the same chromosome cell band. In some embodiments, the same region of the genome is located in the 10Mb region.

[0042] In some embodiments, generating the index of second filtered variants further comprises identifying a second germline variant having an allele frequency greater than or equal to 0.9.

[0043] In some embodiments, generating the index of second filtered variants further comprises identifying a second germline variant in the plurality of variants, wherein the second germline variant is a database variant present in the second reference set of variants.

[0044] In some embodiments, the approximate range is a range having a maximum and a minimum value that differ from the allele frequency of the second germline variant by 0.05.

[0045] In some embodiments, the approximate range is a range having a maximum and a minimum value that are two standard deviations from a binomial distribution of the allele frequency of the second germline variant and centered around the allele frequency of the second germline variant.

[0046] In some embodiments, the second germline variant has an allele frequency within a threshold close to the allele frequencies of at least five database variants that are in the same bin as the second germline variant. In some embodiments, the second germline variant has an allele frequency within a threshold close to the allele frequencies of at least ten database variants that are in the same bin as the second germline variant.

[0047] In some embodiments, the biological sample comprising tumor cells is selected from a serum sample, a stool sample, a blood sample, a tumor sample. In some embodiments, the tumor sample is a fixed sample.

[0048] BRIEF DESCRIPTION OF THE DRAWINGS

[0049] Figure 1 An exemplary embodiment of a workflow is depicted, which includes obtaining sequence data such as a VCF file, identifying and annotating variants in the data, identifying and filtering germline variants, and returning a variant table indicating the status of the variants.

[0050] Figure 2A The graph shows the variant allele frequency (VAF) of various variants according to the chromosomal location of each variant with somatic variants (black filled circles) and germline variants (grey filled circles).

[0051] Figure 2B The graph shows the VAF of various variants according to the chromosomal position of each variant with filter-determined somatic variants (black filled circles) and filter-determined germline variants (grey filled circles).

[0052] Figure 3The figure shows the VAF of various variants according to the chromosomal position of chromosomes 1-7 for each variant with filter-determined somatic variants (black filled circles) and filter-determined germline variants (grey filled circles), as well as a zoomed-in view of the variant located on chromosome 7, where a specific filter-determined somatic variant has been selected, and the range drawn from the selected variant.

[0053] Figure 4A The graph shows the VAF of various variants according to the chromosomal position of each variant for somatic variants determined by filters with only database filters (black filled circles) and germline variants determined by filters (grey filled circles).

[0054] Figure 4B The graph shows the VAF of various variants according to the chromosomal position of each variant for filter-determined somatic variants (black filled circles) and filter-determined germline variants (grey filled circles) with only database filters and neighbor filters.

[0055] Figure 5 An overview of an exemplary embodiment of a workflow is depicted, which includes obtaining formalin-fixed paraffin-embedded (FFPE) samples, obtaining sequence data, and analyzing the sequence data.

[0056] Figure 6 Depicted is an exemplary embodiment of a workflow that includes filtering germline variants from identified variants using database filters and neighbor filters, and calculating tumor mutation burden.

[0057] Figure 7 The line graphs show the distribution of remaining germline variant counts after filtering with the database only (peak at approximately 3 germline residues / Mb) and the hybrid strategy (peak at approximately 0 germline residues / Mb).

[0058] Fig. 8A The graph shows the comparison of tumor mutation burden (TMB) between tumor-only and tumor / normal assays.

[0059] Figure 8B The graph shows the comparison of tumor mutation burden (TMB) between tumor only and WES tumor normal assays. DETAILED DESCRIPTION OF THE INVENTION

[0061] Some embodiments of the method and system provided herein relate to variant calls from sequence data obtained from a single sample. In some embodiments, based on the allele frequency of the variant in the sample and the position of the variant in the genome, somatic variants can be distinguished from germline variants. As used herein, "variant" can include polymorphisms in nucleic acid molecules. Polymorphisms can include insertions, deletions, variable length tandem repeats, single nucleotide mutations and structural variants such as translocations, copy number variations or the above combination. As used herein, "germline variants" can include variants present in the germ cells and all cells of an individual. As used herein, "somatic variants" can include variants present in tumor cells of an individual and not present in other cells of the individual.

[0062] Traditionally, calling variants between somatic variants and germline variants relies on comparisons between data obtained from tumor samples and data obtained from matched normal samples. However, traditional variant calling requires matched samples to be available and requires obtaining two sets of data. Embodiments provided herein relate to calling variants from sequence data obtained from a single sample from an individual. Using a single sample can reduce the need for matched samples and reduce the cost required to obtain sequence data for tumor samples and matched normal samples.

[0063] Some embodiments relate to obtaining sequence data from a sample, such as a sample containing tumor cells from an individual, comparing the sequence data with a reference to identify multiple variants in the sequence data, and applying one or more filters to the variants to identify germline variants and somatic variants. In some embodiments, the filter may include a neighbor filter. In some embodiments, the neighbor filter includes dividing multiple variants into multiple boxes according to the position of the variant in the genome. The variants of some boxes can be identified as germline variants by the corresponding variants present in one or more variant reference sets. If the uncharacterized box variant has an allele frequency similar to the allele frequency of one or more identified germline variants in the same box as the uncharacterized variant, the uncharacterized box variant can be determined as a germline variant. Some embodiments also include applying a database filter to identify germline variants. The database filter can identify germline variants based on the allele counts of the corresponding variants in one or more variant reference sets. In some embodiments, database filters and neighbor filters can be applied to multiple variants to identify germline variants. In some embodiments, somatic variants are variants identified as germline variants. The number of somatic variants can indicate the tumor mutation load of the tumor.

[0064] Tumor mutation load has emerged as an important biomarker for cancer treatment selection after recent studies have shown a correlation between tumor mutation load and the effectiveness of checkpoint inhibitor immunotherapy. When calculating tumor mutation load, it is useful to identify and filter out germline variants. Germline variants can include variants that an individual is born with (or shared between tumor and normal cells) but are detected as variants compared to a reference genome. These variants do not help to distinguish tumor cells from normal cells, and therefore may lead to an overestimation of tumor mutation load if they are not correctly filtered out. The embodiment includes determining the tumor mutation load of a tumor, selecting a treatment for the tumor based on the tumor mutation load, and administering the treatment to a subject in need.

[0065] Some methods

[0066] Some embodiments of the methods and systems provided herein relate to methods for identifying somatic variants in a plurality of variants including somatic variants and germline variants. In some embodiments, one or more filters can be used to filter germline variants from a plurality of variants. Examples of such filters include database filters and neighboring filters.

[0067] In some embodiments, the database filter can be applied to multiple variants. The database filter can be used to identify a variant as a germline variant and remove the variant from the multiple variants. For a specific variant in the multiple variants, the database filter can be associated with the allele count of the corresponding variant in the database.

[0068] For each variant in a variety of variants, the database can be searched for the corresponding variant in the reference database. The reference database can include a database of variants of multiple individuals. Examples of databases that can be used for the embodiments provided herein include a genome aggregation database (gnomAD), including gnomAD exome and gnomAD genome databases, and 1000 genome databases (International Genome Sample Resource). See, for example, Lek, M.et al., (2016) Nature 536: 285-292, which is incorporated herein by reference as a whole. The total allele count of the corresponding variant can be determined in one or more reference databases. The allele count can represent the total number of observations of the variant observed in the database. For example, the allele count in the database of the corresponding variant is 10, indicating that the corresponding variant is observed in at least 5 samples for the homozygous variant, or observed in up to 10 samples for the heterozygous variant. In some embodiments, the allele count can be the highest allele count observed in more than one database. The variant of the corresponding variant with an allele count greater than or equal to a certain threshold allele count can be identified as a germline variant. In some embodiments, the threshold allele count can be greater than or equal to 2, 3, 4, 5, 6, 7, 8, 9, 10, 11, 12, 13, 14, 15, 16, 17, 18, 19, and 20.

[0069] In some embodiments, contiguous filters can be applied to multiple variants.Database filters can be used for variant identification as germline variants, and remove the variant from multiple variants.Contiguous filters can be related to the proximity of the allele frequency of the germline variant identified in the same region of the genome with the allele frequency of a certain variant in multiple variants, the position of the variant in the genome region, and the allele frequency of the variant and the genome.In some embodiments, the variant of multiple variants can be classified or divided into multiple boxes so that the variants in the same region of the genome are classified or divided into the same box.In some embodiments, the same region of the genome can be located in the same chromosome, in the same chromosome arm, in the same chromosome cell band.In some embodiments, the same region of the genome can be located in identical continuous 100Mb, 50Mb, 40Mb, 30Mb, 20Mb, 10Mb, 5Mb, 1Mb, or in any range between any two of the aforementioned numbers.

[0070] In some embodiments, the proximity filter also includes determining which bin variants can be readily identified as germline variants. For example, bin variants can have corresponding variants present in one or more reference databases and be identified as germline variants.

[0071] In some embodiments, the proximity filter includes determining that a variant having an allele frequency greater than or equal to a threshold frequency in a sample is a germline variant. In some such embodiments, a variant having an allele frequency greater than or equal to 0.7, 0.8, 0.9, or 1.0 can be identified as a germline variant.

[0072] In some embodiments, the contiguous filter comprises the approximate range of the allele frequency of the variant that is not yet identified as germline variant.The approximate range of the allele frequency of the variant can include the scope of the allele frequency above and below the allele frequency of the variant.In some embodiments, the approximate range is to have the maximum and minimum of any numeral in the scope between 0.01,0.02,0.03,0.04,0.05,0.06,0.07,0.08,0.09 that differs from the allele frequency of the variant, or any two aforementioned numerals.For example, for the variant with the allele frequency of 0.2 and the approximate range of 0.05, the minimum and maximum of the approximate range will be respectively the allele frequency of 0.15 and 0.25.

[0073] In some embodiments, assuming that the supporting evidence for a given variant is generated by a binomial process, the approximate range is determined by the value of two (n) standard deviations of the binomial distribution. For example, for a variant with allele frequency (x) and coverage (y), the approximate range (z) can be:

[0074] z=n*sqrt(y*x*(1-x)) / y

[0075] For example, for a variant with an allele frequency of 0.2, a coverage / sequencing depth of 100, the approximate range would be 0.08, and the minimum and maximum values ​​of the approximate range would be allele frequencies of 0.12 and 0.28, respectively. In some embodiments, the approximate range is the higher of 0.05 above and below the variant allele frequency, or two (n) standard deviations from the binomial distribution of the variant allele frequency.

[0076] In some embodiments, a variant can be identified as a germline variant if the variant has an allele frequency that is within the approximate range of one or more identified germline variants in the same bin as the variant. In some embodiments, a variant can be identified as a germline variant if the variant has an allele frequency that is within the approximate range of more than 1, 2, 3, 4, 5, 6, 7, 8, 9, or 10 identified germline variants in the same bin as the variant. In some embodiments, a variant can be identified as a germline variant if the variant has an allele frequency that is within the range of more than 5 identified germline variants in the same bin as the variant. For example, in an embodiment, a variant will be identified as a germline variant if the variant has an allele frequency within the range of more than 5 identified germline variants in the same bin as the variant: a variant with an allele frequency of 0.2, with an approximate range of 0.05, and therefore with a minimum range of 0.15 and a maximum range of 0.25, and is assigned to a bin representing chromosome 7 will be identified as a germline variant, where more than 5 identified germline variants have an allele frequency within the approximate range of the variant and are assigned to a bin representing chromosome 7.

[0077] In some embodiments, the proximity filter identifies somatic variants, which are variants that are not identified as germline variants. In some embodiments, the number of somatic variants obtained from sequencing data from a tumor is the tumor mutation burden of the tumor.

[0078] In some embodiments, database filters or proximity filters can be applied to multiple variants to identify and remove germline variants from multiple variants. In some embodiments, database filters and proximity filters can be applied continuously. For example, the output of a database filter can be used for the input of a proximity filter. On the contrary, the output of a proximity filter can be used as the input of a database filter.

[0079] Certain electronic systems and computer-implemented methods

[0080] Some embodiments of the methods and systems provided herein include an electronic system for analyzing genetic variation data. In some such embodiments, the database filters described herein and / or the proximity filters described herein can be applied to the genetic variation data to identify germline variants.

[0081] Some embodiments may include an informatics module running on a processor and configured to identify a plurality of variants from sequence data from a biological sample comprising tumor cells, wherein the plurality of variants comprises somatic variants and germline variants.

[0082] Some embodiments include a database filter module configured to remove germline variants from the plurality of variants, wherein each of the germline variants has an allele count greater than or equal to a threshold allele count in a variant reference set. In some embodiments, the threshold allele count can be greater than or equal to 2, 3, 4, 5, 6, 7, 8, 9, 10, 11, 12, 13, 14, 15, 16, 17, 18, 19, and 20.

[0083] Some embodiments include a contiguous filter module, which is configured to remove germline variants from multiple variants. In some embodiments, the contiguous filter module may include a box module configured to return multiple boxes, and each box contains the variant of the multiple variants in the same region of the genome. In some embodiments, the variant of multiple variants may be classified or divided into multiple boxes so that the variants in the same region of the genome are classified or divided into the same box. In some embodiments, the same region of the genome may be located in the same chromosome, in the same chromosome arm, in the same chromosome cell band. In some embodiments, the same region of the genome may be located in identical continuous 100Mb, 50Mb, 40Mb, 30Mb, 20Mb, 10Mb, 5Mb, 1Mb, or in any range between any two of the aforementioned numbers.

[0084] In some embodiments, the proximity filter module may include an identification submodule configured to return a database variant among the plurality of variants, wherein the database variant is present in a reference set of variants.

[0085] In some embodiments, the neighboring filter module may include a removal submodule configured to remove a germline variant from a plurality of variants, wherein each of the germline variants has an allele frequency within the approximate range of the allele frequency of at least one database variant, and the at least one database variant is in the same box as the germline variant. In some embodiments, the neighboring filter includes determining the approximate range of the allele frequency of a variant that has not yet been identified as a germline variant. In some embodiments, the approximate range is the range of the maximum and minimum values ​​of any number in the range between 0.01, 0.02, 0.03, 0.04, 0.05, 0.06, 0.07, 0.08, 0.09, or any two of the aforementioned numbers with the variant allele frequency. In some embodiments, the approximate range is the range of the maximum and minimum values ​​with the binomial distribution of the variant allele frequency differing by two standard deviations. In some embodiments, the approximate range is the higher of 0.05 or the binomial distribution of the variant allele frequency differing by two (n) standard deviations above and below the variant allele frequency.

[0086] In some embodiments, if the allele frequency that variant has is in the approximate range of the germline variant of one or more identifications in the same box with variant, variant can be identified as germline variant.In some embodiments, if the allele frequency that variant has is in the approximate range of the germline variant of more than 1,2,3,4,5,6,7,8,9 or 10 kinds of identifications in the same box with variant, variant can be identified as germline variant.In some embodiments, remove submodule and be configured to remove the variant with the allele frequency greater than or equal to threshold frequency.In some such embodiments, the variant with the allele frequency greater than or equal to 0.7,0.8,0.9 or 1.0 can be identified as germline variant.In some embodiments, remove submodule and be configured to remove the database variant that is present in the variant reference set from multiple variants.

[0087] Some embodiments provided herein include computer-implemented methods for identifying somatic variants in a plurality of variants.Some such embodiments may include receiving a plurality of variants from sequence data from a biological sample comprising a tumor cell, the plurality of variants may include somatic variants and germline variants.

[0088] Some embodiments include applying a database filter to multiple variants. Some such embodiments include creating a file index for multiple variants, searching a variant reference set with the index to identify the germline variants in the index. In some embodiments, each of the germline variants has an allele count greater than or equal to a threshold allele count in the variant reference set. In some embodiments, the threshold allele count can be greater than or equal to 2, 3, 4, 5, 6, 7, 8, 9, 10, 11, 12, 13, 14, 15, 16, 17, 18, 19 and 20. Some embodiments also include removing the identified germline variants from the index to generate an index of the first filtered variant.

[0089] Some embodiments include applying a neighboring filter to the index of the first filtered variant.Some such embodiments include creating multiple boxes for different regions of a genome.Some embodiments include binning the variant of the index of the first filtered variant, wherein the variant positioned in the same region of the genome is divided into the same box.In some embodiments, the same region of the genome can be positioned in the same chromosome, in the same chromosome arm, in the same chromosome cell band.In some embodiments, the same region of the genome can be positioned in the same continuous 100Mb, 50Mb, 40Mb, 30Mb, 20Mb, 10Mb, 5Mb, 1Mb, or in any range between any two of the aforementioned numbers.

[0090] Some embodiments include searching a reference set of variants with an index of the first filtered variants to identify database variants in the index of the first filtered variants.

[0091] Some embodiments include generating the index of germline variant from the index of the first filtered variant by identifying germline variant.In some embodiments, each of the germline variants has an allele frequency within the approximate range of the allele frequency of at least one database variant, and the at least one database variant is in the same box with the second germline variant.In some embodiments, the approximate range is the range of the maximum and minimum values ​​of any number in the range between 0.01, 0.02, 0.03, 0.04, 0.05, 0.06, 0.07, 0.08, 0.09, or any two aforementioned numbers with the variant allele frequency.In some embodiments, the approximate range is the range of the maximum and minimum values ​​with the binomial distribution of the allele frequency of the variant differing by two standard deviations.In some embodiments, the approximate range is higher than and lower than 0.05 of the variant allele frequency or differs from the binomial distribution of the variant allele frequency by two (n) standard deviations.

[0092] In some embodiments, if the allele frequency that variant has is within the approximate range of the germline variant of one or more identifications in the same box with variant, variant can be identified as germline variant.In some embodiments, if the allele frequency that variant has is within the approximate range of the germline variant of more than 1,2,3,4,5,6,7,8,9 or 10 kinds of identifications in the same box with variant, variant can be identified as germline variant.In some embodiments, germline variant can be identified as variant with allele frequency greater than or equal to threshold frequency.In some such embodiments, variant with allele frequency greater than or equal to 0.7,0.8,0.9 or 1.0 can be identified as germline variant.

[0093] Some embodiments include removing the identified germline variants from the index of first filtered variants to generate an index of somatic variants, thereby identifying somatic variants in the plurality of variants. In some embodiments, the number of somatic variants obtained from sequencing data from a tumor is the tumor mutation burden of the tumor.

[0094] Treatment

[0095] Some embodiments of the method and system include methods for treating tumors. In some such embodiments, the number of somatic variants present in a tumor can be determined by the method and system provided herein. For example, sequence data can be obtained from a tumor, multiple variants can be identified from the sequence data, and germline variants can be identified and removed from multiple variants, thereby identifying somatic variants in multiple variants. In some embodiments, germline variants can be identified and removed from multiple variants by applying one or more of database filters and / or neighboring filters, thereby identifying somatic variants that are not removed by applying one or more filters. In some embodiments, the number of somatic variants obtained from sequencing data from a tumor is the tumor mutation load of the tumor. In some embodiments, the tumor mutation load is calculated as the average number of somatic variants per genomic region, such as the average number of mutations per 50kb, 100kb, 1Mb, 10Mb, 100Mb, etc. The tumor mutation load can be sampled by sequencing the entire genome or a portion thereof. For example, a portion of the genome can be sequenced by enriching one or more target genomic regions, such as a tumor genome, a complete exon, a partial exon, etc.

[0096] Some embodiments of treating tumors may include determining that the tumor has a tumor mutation load greater than or equal to a tumor mutation load threshold, and contacting the tumor with an effective amount of a therapeutic agent. Some embodiments include treating a subject with a tumor, and may include determining that the tumor has a tumor mutation load greater than or equal to a TMB threshold, and administering an effective amount of a therapeutic agent to the subject. In some embodiments, the tumor mutation load threshold may be 2, 3, 4, 5, 6, 7, 8, 9, 10, 15, 20, 30, 40, 50, 60, 70, 80, 90, 100, 200, 300, 400, 500, 600, 700, 800, 900, 1000, or any number in the range between any two of the foregoing numbers. Examples of therapeutic agents include chemotherapeutic agents. In some embodiments, the therapeutic agent may include a checkpoint inhibitor. Examples of checkpoint inhibitors include CTLA-4 inhibitors, PD-1 inhibitors, and PD-L1 inhibitors. In some embodiments, checkpoint inhibitors may include ipilimumab, nivolumab, pembrolizumab, spartalizumab, atezolizumab, avelumab, and durvalumab. Examples of tumors include colorectal tumors, lung tumors, endometriomas, uterine tumors, gastric tumors, melanomas, breast tumors, pancreatic tumors, kidney tumors, bladder tumors, and brain tumors. More examples of cancers that can be treated with the methods and systems included herein are listed in US20180218789, which is expressly incorporated herein by reference in its entirety.

[0097] sample

[0098] Some embodiments include obtaining sequence data from a biological sample. In some embodiments, the biological sample may include a tumor cell. In some embodiments, the biological sample may include a serum sample, a stool sample, a blood sample, and a tumor sample. In some embodiments, the biological sample is a fixed sample.

[0099] In some embodiments, the object can provide a biological sample. The biological sample can be any substance produced by the object. Generally, the biological sample is any tissue taken from the object or any substance produced by the object. Examples of biological samples may include blood, plasma, saliva, cerebrospinal fluid (CSF), cheek tissue, urine, feces, skin, hair, organ tissue. In some embodiments, the biological sample is a biopsy of a solid tumor or a solid tumor. In some embodiments, the biological sample is a formalin-fixed paraffin-embedded (FFPE) tissue sample. The biological sample can be any biological sample containing nucleic acids. The biological sample can be derived from the object. The object can be a mammal, a reptile, an amphibian, a bird or a fish. Examples of mammals include humans, apes, orangutans, monkeys, chimpanzees, cattle, pigs, horses, rodents, birds, reptiles, dogs, cats, dolphins or other animals. Examples of reptiles include lizards, snakes, alligators, turtles, crocodiles, iguanas and tortoises. Examples of amphibians include toads, frogs, salamanders and salamanders. Examples of birds include chickens, ducks, geese, penguins, ostriches, puffins, and owls. Examples of fish include catfish, eels, sharks, goldfish, and swordfish. In some embodiments, the subject is a human.

[0100] Certain systems and methods

[0101] Some embodiments include computer-based systems and computer-implemented methods for performing methods described herein. In some embodiments, the system can be used to determine and report the presence or absence of variants in a sample, such as germline variants and / or somatic variants. The system can include one or more client components. One or more client components can include a user interface. The system can include one or more server components. The server component can include one or more memory locations. One or more memory locations can be configured to receive data input. The data input can include sequencing data. Sequencing data can be generated from a nucleic acid sample from an object. The system can also include one or more computer processors. One or more computer processors can be operably connected to one or more memory locations. One or more computer processors can be programmed to map sequencing data to a reference sequence. One or more computer processors can be further programmed to determine the presence or absence of multiple variants from sequencing data. One or more computer processors can be further programmed to apply at least one filter to genetic variants, thereby identifying germline variants. Examples of filters include database filters and neighboring filters. One or more computer processors can be further programmed to remove the identified germline variants from the index of the identified variants. One or more computer processors can be further programmed to generate output for display on a screen. The output may include one or more reports identifying germline variants and / or somatic variants among the plurality of variants.

[0102] Some embodiments of the method and system may include one or more client components. One or more client components may include one or more software components, one or more hardware components, or a combination of the above. One or more client components may access one or more services through one or more server components. One or more services may be accessed by one or more client components through a network. "Service" as used herein refers to any product, method, function, or purpose of the system. For example, a user may place an order for genetic testing. An order may be placed through one or more client components of the system, and a request may be sent to one or more server components of the system through a network. The network may be the Internet, the Internet and / or an extranet, or an intranet and / or an extranet communicating with the Internet. In some cases, the network is a telecommunications and / or data network. The network may include one or more computer servers that are capable of implementing distributed computing, such as cloud computing. In some cases, with the help of a computer system, the network may implement a peer-to-peer network that may enable a device connected to the computer system to behave as a client or server.

[0103] Some embodiments of the system may include one or more memory locations, such as random access memory, read-only memory, flash memory; electronic storage units, such as hard disks; communication interfaces, such as network adapters, for communicating with one or more other systems and peripherals, such as caches, other memories, data storage, and / or electronic display adapters. The memories, storage units, interfaces, and peripherals communicate with the CPU via a communication bus, such as a motherboard. The storage unit may be a data storage unit or a data repository for storing data. In one example, one or more memory locations may store received sequencing data.

[0104] Some embodiments of the method and system may include one or more computer processors. One or more computer processors may be operably connected to one or more memory locations to, for example, access stored sequencing data. One or more computer processors can implement machine executable code to perform the methods described herein. For example, one or more computer processors can execute machine readable code to map sequencing data input to a reference sequence, and / or identify germline variants and / or somatic variants.

[0105] Some embodiments of the method and system provided herein may include machine executable code or machine readable code. In some such embodiments, machine executable code or machine readable code may be provided in the form of software. During use, the code may be executed by a processor. In some cases, the code may be retrieved from a storage unit and stored in a memory so that it is ready for access by a processor. In some embodiments, an electronic storage unit may be excluded, and machine executable instructions may be stored on a memory. Code may be precompiled and configured to be used with a machine having a processor configured to execute code, may be compiled during runtime, or may be interpreted during runtime. Code may be provided in a programming language, which may be selected so that the code may be executed in a precompiled, compiled (as-complied) or interpreted manner.

[0106] Some embodiments of the system and method provided herein such as computer systems can be implemented in programming. Various aspects of the technology can be considered as "products" or "articles" in the form of machine (or processor) executable code and / or related data that are usually executed or implemented on a type of machine-readable medium. Machine executable code can be stored in an electronic storage unit, such as a memory or a hard disk. "Storage" type media may include any or all tangible memories of a computer, processor, etc., or its associated modules, such as various semiconductor memories, tape drives, disk drives, etc., which can provide non-transitory storage for software programming at any time. All or part of the software can sometimes communicate through the Internet or various other telecommunications networks. For example, such communication can allow software to be loaded from one computer or processor to another computer or processor, for example, from a management server or host to a computer platform of an application server. Therefore, another type of medium that can carry software elements includes light waves, electric waves and electromagnetic waves, such as through wired and optical landline networks and through various air links on the physical interface between local devices. Physical elements that carry such waves, such as wired or wireless links, optical links, etc., can also be considered as media that carry software. As used herein, unless restricted to non-transitory, tangible "storage" media, terms such as computer or machine "readable media" refer to any medium that participates in providing instructions to a processor for execution.

[0107] Some embodiments of the method and system disclosed herein may include one or more electronic displays or communicate with one or more electronic displays. The electronic display may be a part of a computer system, or may be directly or via a network connected to a computer system. The computer system may include a user interface (UI) for providing various features and functions disclosed herein. Examples of UI include, but are not limited to, a graphical user interface (GUI) and a web-based user interface. UI may provide an interactive tool through which a user may utilize the method and system described herein. By way of example, the UI envisioned herein may be a web-based tool through which a medical practitioner may order a genetic test, customize a list of genetic variants to be tested, and receive and view a biomedical report.

[0108] Some embodiments of the methods and systems disclosed herein may include biomedical databases, genomic databases, biomedical reports, disease reports, case-control analyses, and rare variant discovery analyses based on data and / or information from one or more databases, one or more assays, one or more data or results, one or more outputs based on or derived from one or more assays, one or more outputs based on or derived from one or more data or results, or a combination of the above. Example

[0109] Example 1 - Identification of somatic variants by sample comparison

[0110] Sequence data of a tumor sample and a normal sample are obtained from an individual. Variants are identified in the sequence data. Germline variants in the tumor sample are identified by comparing variants present in the tumor sample and absent in the normal sample. Figure 2A The graph shows the variant allele frequencies (VAFs) of different variants according to the chromosomal position of each variant with somatic variants (black filled circles) and germline variants (grey filled circles). The method requires two samples from an individual.

[0111] Example 2 - Database Filtering of Variants

[0112] Sequence data of tumor samples were obtained only from Example 1. Variants were identified in the sequence data. In brief, variants called from the variant calling pipeline were annotated using the annotation tool Nirvana (Illumina, San Diego). Nirvana provides clinical-grade annotations of genomic variants such as single nucleotide variants, multinucleotide variants, insertions, deletions, and copy number variants. The input to Nirvana is in the variant call format (VCF), and the output is a structured JSON representation of all annotations and sample information.

[0113] For identified variants, the total allele counts and variant allele frequencies and coverage of given variants are analyzed in the genome aggregation database (gnomAD) exome, gnomAD genome and 1000 genome databases. These total allele counts represent the total number of observations of different subgroups in the database. For each variant, the maximum allele count observed in all three databases is adopted to consider the region not covered in the exome database, while utilizing its larger sample size compared with the genome database. The filtering strategy will be marked as a potential germline variant with a maximum allele count greater than or equal to 10. For a given variant, the allele count in the database is 10, which means that if they are all homozygous, it is observed in at least 5 samples, or if they are all heterozygous, it is observed in up to 10 samples. Figure 2B The graph shows the variant allele frequency (VAF) of various variants according to the chromosomal position of each variant with filter-determined somatic variants (black filled circles) and filter-determined germline variants (grey filled circles). This demonstrates that only database filtering miscalled the variants.

[0114] Example 3 - Proximity Filtering of Variants

[0115] Sequence data of tumor samples were obtained only from individuals. Variants were identified in the sequence data. The database filter of Example 2 was applied to the variants. Variants not found in the database were further filtered out using a neighbor filter.

[0116] The neighboring filter uses the information of the variants filtered by the database in close proximity in position. For a given variant not found in the database and with an allele frequency lower than 0.9, the variants on the same chromosome are retrieved within the range of the allele frequency of the given variant of the unfiltered variant. Variants with an allele frequency greater than 90% are marked as germline without any further processing. The range is determined as the maximum value of 0.05 and 2 standard deviations of the binomial distribution, which assumes that the supporting evidence of the given variant is generated by the binomial process. For example, if the allele frequency of the unfiltered variant is 0.2 and the coverage is 100, the range is the maximum value between 0.05 and 2*sqrt(100*0.2*(1-0.2)) / 100=0.08, i.e. 0.08. This is converted into a range of 0.08 in both directions, and all variants are retrieved from the same chromosome with an allele frequency between 0.12-0.28. Subsequently, the number of variants retrieved that exceed a fixed threshold is checked and set to 5. If the required number of variants is met, we then check if a large fraction of those variants set to 0.95 are filtered by the database filter. If a variant meets both conditions, it is flagged by the neighbor filter. In other words, if a variant is surrounded by a sufficient number of variants in the allele frequency space found in the database, it is also considered a germline variant. This filter removes germline variants in normal regions with an expected variant allele frequency of about 50% or 100% and in copy number variant regions where the allele frequency distribution may be altered.

[0117] Figure 3 (Left) Graph showing the variant allele frequency (VAF) of various variants according to chromosomal position on chromosomes 1-7 for each variant with filter-determined somatic variants (black filled circles) and filter-determined germline variants (grey filled circles) filtered with database filters only. Figure 3 (Right) is a zoomed in view of a variant located on chromosome 7, where a particular filter-determined somatic variant (black circle) has been selected, and a range drawn from that variant includes several filter-determined germline variants (grey circles). It can be determined that the selected filter-determined somatic variant (black circle) should be referred to as a germline variant based on the proximity of the allele frequency of the selected variant to the allele frequencies of a certain number of identified germline variants.

[0118] Figure 4AThe graph shows the variant allele frequency (VAF) of various variants according to the chromosomal position of each variant for somatic variants determined by filters with only database filters (black filled circles) and germline variants determined by filters (grey filled circles). Figure 4B The graph shows the variant allele frequency (VAF) of various variants according to the chromosomal position of each variant for somatic variants determined by filters with only database filters and neighbor filters (black filled circles) and germline variants determined by filters (grey filled circles). Figure 4B Shown in Figure 4A Some putative false positives for somatic variants are shown in Figure 4B For example, when a neighbor filter is applied, the identified somatic variants located on chromosome 7 with allele frequencies of approximately 0.4 and 0.3 ( Figure 4A ) was identified as a germline variant ( Figure 4B ).

[0119] Example 4 - Measuring Tumor Mutation Burden Using Targeted Sequencing

[0120] This example relates to a targeted next generation sequencing assay for measuring tumor mutation burden (TMB) in formalin-fixed paraffin-embedded (FFPE) tumor samples. Figure 5 An exemplary workflow for the assay is shown. Sequence data for 523 genes with a panel size of 1.94 Mb and an exon size of 1.33 Mb were obtained from tumor samples. Sequencing utilized unique molecular identifiers (UMIs) and used Illumina NextSeq TM 500 / 550 platform. Data analysis was performed using a pipeline for detecting variants with a variant allele frequency (VAF) of 5%. For technical noise removal, a variant calling algorithm was used that utilized information from UMIs and sample-specific error signatures to ensure uniform variant calling performance across samples of different FFPE quality. In order to accurately remove germline variants from TMB calculations, a hybrid strategy was used that combined information from large-scale public databases with measured coverage and variant allele frequencies for each variant measured, and which was substantially similar to the database filters and neighbor filters of the above-described embodiments.

[0121] Briefly, sequence data were obtained, aligned to a reference, and variants were identified. Germline variants were filtered from the identified variants using database filters and neighbor filters, and in a manner substantially similar to Figure 6 TMB was calculated in the workflow of the pipeline shown. A total of 170 pairs of tumor-normal samples were analyzed to evaluate germline filtering and TMB performance (Table 1). A subset of 108 sample pairs was also analyzed using whole exome sequencing (WES).

[0122] Table 1

[0123]

[0124]

[0125] For technical noise removal, the number of false positive variants in a collection of normal FFPE samples (N=176) was evaluated. An average of 0.63 false positives per sample was observed, regardless of sample quality (R 2 =0.001), where 92.6% of samples contained ≤2 false positive variants (VAF<20%). In addition, a panel of FFPE and cell line mixed samples was tested, where variants approached 5% and achieved a sensitivity of 98.7%.

[0126] The performance of germline filtering was evaluated using 170 tumor / normal sample pairs described in Table 1. In filtering of small variants (SNVs, indels), an overall filtering rate of over 99.7% was achieved, leaving less than 1.3 germline variants per sample on average. The addition of neighbor filtering significantly reduced the number of false positives while having only minimal impact on somatic mutations. Figure 7 Shown are the distributions of remaining germline variant counts after database-only filtering (peak at approximately 3 germline residues / Mb) and after the hybrid strategy (peak at approximately 0 germline residues / Mb).

[0127] TMB reproducibility was evaluated in 8 different samples including 4 cell lines and 4 FFPE samples between 3 operators. The mean and standard deviation (SD) were calculated for each sample. Table 2 lists the TMB reproducibility evaluated in 4 cell lines and 4 FFPE samples, each repeated 12 times.

[0128] Table 2

[0129]

[0130]

[0131] In conclusion, TMB measurements generated from the tumor-only assay correlated highly with estimates generated from tumor / normal assay pair samples (R 2 =0.993, N=169, samples with TMB<200 only). Tumor-only TMB estimates also showed a high correlation with TMB values ​​obtained by whole-exome sequencing (R 2 = 0.931, N = 105, samples with WES TMB < 100 only). Fig. 8A TMB comparisons between tumor-only and tumor / normal assays are shown. Figure 8BTMB comparisons between tumor-only and WES tumor-normal assays are shown.

[0132] Finally, a positive percentage agreement (PPA) of 94.74% and a negative percentage agreement (NPA) of 96.08% were demonstrated at a TMB threshold of 10. The overall classification agreement was 95.37% in distinguishing between TMB-high and TMB-low samples. Table 3 lists the TMB classification performance.

[0133] Table 3

[0134] WES T / N TMB high WES T / N TMB Low TMB High 54 3 Low TMB 2 49

[0135] The above results demonstrate the ability of a tumor assay with only a database and proximity filter to robustly measure TMB in FFPE samples. Furthermore, TMB estimates showed a high level of correlation with WES-based measurements with high classification consistency.

[0136] As used herein, the term "comprising" is synonymous with "including," "containing," or "characterized by," and is inclusive or open-ended and does not exclude additional, unrecited elements or method steps.

[0137] The above description discloses several methods and materials of the present application. The present application is capable of accepting improvements to methods and materials and changes in manufacturing methods and equipment. In view of the invention disclosed in the present application or the practice of the invention, such modifications will be apparent to those skilled in the art. Therefore, it is not intended that the present application be limited to the disclosed specific embodiments, but rather all modifications and alternatives falling within the true scope and spirit of the present application are covered.

[0138] All references cited in this application, including but not limited to published and unpublished applications, patents and references, are hereby incorporated by reference into this application as a whole and thus constitute a part of this specification. In the event that the publications and patents or patent applications added by reference conflict with the disclosure contained in this specification, this specification is intended to replace and / or take precedence over any such conflicting materials.

Claims

1. A method for identifying a somatic variant among a plurality of variants for non-disease diagnostic purposes, comprising: (a) obtaining a plurality of variants comprising somatic variants and germline variants; (b) applying a database filter to the plurality of variants, comprising the steps of: determining a first germline variant among the plurality of variants, wherein each of the first germline variants has an allele count in a first variant reference set that is greater than or equal to a threshold allele count, wherein the threshold allele count is greater than or equal to 2; (c) applying a neighbor filter to the plurality of variants, comprising the steps of: (i) grouping variants of the plurality of variants into a plurality of bins, wherein variants located in the same region of the genome are grouped into the same bin, (ii) determining a database variant among the plurality of variants, wherein the database variant is present in a second reference set of variants, and (iii) determining a second germline variant in the plurality of variants, wherein each of the second germline variants has an allele frequency that is within an approximate range of the allele frequency of at least one database variant that is in the same bin as the second germline variant; and (d) determining a somatic variant in the plurality of variants by removing the first germline variant and the second germline variant from the plurality of variants, Wherein the first variant reference set or the second variant reference set comprises at least one database of variants of a plurality of individuals.

2. The method of claim 1, wherein (b) and (c) are performed continuously.

3. The method of claim 1, wherein (c) is performed before (b).

4. The method of any one of claims 1-3, wherein the threshold allele count is 5.

5. The method of any one of claims 1-3, wherein the threshold allele count is 10. The method of claim 1 , wherein the first variant reference set and the second variant reference set are the same reference set.

7. The method of claim 1, wherein the at least one database comprises at least one database selected from the group consisting of the genome aggregation database gnomAD and the 1000 Genomes database.

8. The method of claim 1, wherein the same region of the genome is located within the same chromosome.

9. The method of claim 1, wherein the same region of the genome is located within the same chromosome arm.

10. The method of claim 1, wherein the same region of the genome is located within the same chromosomal cell band.

11. The method of claim 1, wherein the identical region of the genome is located within a 10 Mb region.

12. The method of claim 1, wherein applying a neighbor filter further comprises identifying a second germline variant having an allele frequency greater than or equal to 0.

9.

13. The method of claim 1, wherein the applying a neighbor filter further comprises identifying a second germline variant in the plurality of variants, wherein the second germline variant is a database variant present in the second variant reference set.

14. The method of claim 1, wherein the approximate range is a range having a maximum and a minimum value that differ from the allele frequency of the second germline variant by 0.

05.

15. The method of claim 1, wherein the approximate range is a range having a maximum and a minimum value that are two standard deviations from a binomial distribution of the allele frequency of the second germline variant and centered around the allele frequency of the second germline variant.

16. The method of claim 1, wherein the second germline variant has an allele frequency that is within a threshold close to the allele frequencies of at least five database variants that are in the same bin as the second germline variant.

17. The method of claim 1, wherein the second germline variant has an allele frequency that is within a threshold close to the allele frequencies of at least ten database variants that are in the same bin as the second germline variant.

18. The method of claim 1, wherein (a) comprises: Sequence data is obtained from a biological sample comprising tumor cells.

19. The method of claim 18, further comprising: The sequence data is aligned to a reference sequence, and variants in the sequence data are identified.

20. The method of claim 18 or 19, wherein the biological sample containing tumor cells is selected from a serum sample, a stool sample, a blood sample, a tumor sample.

21. The method of claim 20, wherein the tumor sample is a fixed sample.

22. A method for determining the tumor mutation burden of a tumor for non-disease diagnostic purposes, comprising: obtaining sequence data from a biological sample comprising tumor cells; determining a plurality of variants from said sequence data; as well as The method of any one of claims 1-21 determines the number of somatic variants in a plurality of variants, wherein the number of somatic variants is the tumor mutation burden of the tumor.

23. Electronic systems for analyzing genetic variation data, including: an informatics module that runs on the processor and is configured to identify a plurality of variants from sequence data from a biological sample comprising tumor cells, wherein the plurality of variants includes somatic variants and germline variants; a database filter module configured to remove first germline variants from the plurality of variants, wherein each of the first germline variants has an allele count in a first variant reference set that is greater than or equal to a threshold allele count; a neighbor filter module configured to remove the second germline variant from the plurality of variants, the neighbor filter module comprising: a binning module configured to return multiple bins, each bin containing variants of multiple variants located in the same region of the genome, a submodule configured to return an identification submodule of a database variant among the plurality of variants, wherein the database variant is present in a second reference set of variants, and a removal submodule configured to remove second germline variants from the plurality of variants, wherein each of the second germline variants has an allele frequency that is within an approximate range of an allele frequency of at least one database variant that is in the same bin as the second germline variant; as well as A display module is configured to return variants that have not been removed from the plurality of variants.

24. The system of claim 23, wherein the informatics module comprises a variant annotation tool.

25. The system of claim 23 or 24, wherein the threshold allele count is 5.

26. The system of claim 23 or 24, wherein the threshold allele count is 10.

27. The system of claim 23, wherein the first variant reference set and the second variant reference set are the same reference set.

28. The system of claim 23, wherein the first variant reference set or the second variant reference set comprises a database of variants of a plurality of individuals.

29. The system of claim 23, wherein the first variant reference set or the second variant reference set comprises at least one database selected from the group consisting of the genome aggregation database gnomAD and the 1000 Genomes database.

30. The system of claim 23, wherein the same region of the genome is located within the same chromosome.

31. The system of claim 23, wherein the same region of the genome is located within the same chromosome arm.

32. The system of claim 23, wherein the same region of the genome is located within the same chromosomal cell band.

33. The system of claim 23, wherein the identical region of the genome is located within a 10 Mb region.

34. The system of claim 23, wherein the removal submodule is configured to remove variants having an allele frequency greater than or equal to 0.9 from the plurality of variants.

35. The system of claim 23, wherein the removing submodule is configured to remove database variants present in the second variant reference set from the plurality of variants.

36. The system of claim 23, wherein the approximate range is a range having a maximum and a minimum value that differ from the allele frequency of the second germline variant by 0.

05.

37. The system of claim 23, wherein the approximate range is a range having a maximum and a minimum value that are two standard deviations from a binomial distribution of the allele frequency of the second germline variant and centered around the allele frequency of the second germline variant.

38. The system of claim 23, wherein the second germline variant has an allele frequency that is within a threshold close to the allele frequencies of at least five database variants that are in the same bin as the second germline variant.

39. The system of claim 23, wherein the second germline variant has an allele frequency that is within a threshold close to the allele frequencies of at least ten database variants that are in the same bin as the second germline variant.

40. The system of claim 23, wherein the biological sample comprising tumor cells is selected from the group consisting of a serum sample, a stool sample, a blood sample, a tumor sample.

41. The system of claim 40, wherein the tumor sample is a fixed sample.

42. A computer-implemented method for identifying a somatic variant among a plurality of variants for non-disease diagnostic purposes, comprising: Execute the method of any one of claims 1 to 21.

43. A computer-implemented method for identifying a somatic variant among a plurality of variants for non-disease diagnostic purposes, comprising: (a) receiving a plurality of variants from sequence data from a biological sample comprising a tumor cell, the plurality of variants comprising somatic variants and germline variants; (b) applying a database filter to the plurality of variants, comprising the steps of: creating a file index for the plurality of variants, searching a first variant reference set with the index to identify first germline variants in the index, wherein each of the first germline variants has an allele count in the first variant reference set that is greater than or equal to a threshold allele count, and removing the identified first germline variant from the index to create an index of first filtered variants; (c) applying a neighbor filter to the index of the first filtered variant, comprising the steps of: (i) creating multiple bins for different regions of the genome, (ii) binning the variants of the index of the first filtered variants, wherein variants located in the same region of the genome are grouped into the same bin, (iii) searching a second reference set of variants using the index of the first filtered variants to identify database variants in the index of the first filtered variants, (iii) generating an index of second germline variants from the index of first filtered variants by identifying second germline variants, wherein each of the second germline variants has an allele frequency that is within an approximate range of the allele frequency of at least one database variant that is in the same bin as the second germline variant, and (iv) removing the identified second germline variant from the index of first filtered variants to generate an index of somatic variants, thereby identifying the somatic variant in the plurality of variants.

44. The method of claim 43, wherein the threshold allele count is 5.

45. The method of claim 43, wherein the threshold allele count is 10.

46. ​​The method of claim 43, wherein the first variant reference set and the second variant reference set are the same reference set.

47. The method of claim 43, wherein the first variant reference set or the second variant reference set comprises a database of variants of a plurality of individuals.

48. The method of claim 43, wherein the first variant reference set or the second variant reference set comprises at least one database selected from the group consisting of the genome aggregation database gnomAD and the 1000 Genomes database.

49. The method of claim 43, wherein the same region of the genome is located within the same chromosome.

50. The method of claim 43, wherein the same region of the genome is located within the same chromosome arm.

51. The method of claim 43, wherein the same region of the genome is located within the same chromosomal cell band.

52. The method of claim 43, wherein the identical region of the genome is located within a 10 Mb region.

53. The method of claim 43, wherein generating an index of second filtered variants further comprises identifying a second germline variant having an allele frequency greater than or equal to 0.

9.

54. The method of claim 43, wherein generating an index of second filtered variants further comprises identifying a second germline variant in the plurality of variants, wherein the second germline variant is a database variant present in the second reference set of variants.

55. The method of claim 43, wherein the approximate range is a range having a maximum and a minimum value that differ from the allele frequency of the second germline variant by 0.

05.

56. The method of claim 43, wherein the approximate range is a range having a maximum and a minimum value that are two standard deviations from a binomial distribution of the allele frequency of the second germline variant and centered around the allele frequency of the second germline variant.

57. The method of claim 43, wherein the second germline variant has an allele frequency within a threshold that is close to the allele frequencies of at least five database variants that are in the same bin as the second germline variant.

58. The method of claim 43, wherein the second germline variant has an allele frequency that is within a threshold close to the allele frequencies of at least ten database variants that are in the same bin as the second germline variant.

59. The method of claim 43, wherein the biological sample comprising tumor cells is selected from the group consisting of a serum sample, a stool sample, a blood sample, a tumor sample.

60. The method of claim 59, wherein the tumor sample is a fixed sample.

Images