Detecting homologous recombination deficiency in breast and ovarian cancers by genome mapping

Abstract

Some embodiments of the methods and compositions provided herein relate to detecting a homologous recombination deficiency (HRD) in a tumor sample from a subject, such as a breast or ovarian cancer tumor. Some embodiments include obtaining genome mapping (GM) data from the tumor sample, identifying structural variants (SVs) from the GM data, and determining the presence of absence of an HRD in the sample based on the SVs. Some embodiments also include determining a BRCA1 / BRCA2-like status for a sample comprising an HDR.

Description

68LH-885002-WO PATENTDETECTING HOMOLOGOUS RECOMBINATION DEFICIENCY IN BREAST AND OVARIAN CANCERS BY GENOME MAPPINGCROSS-REFERENCE TO RELATED APPLICATIONS

[0001] This application claims priority to U.S. Prov. App. No. 63 / 789974, filed April 16, 2025, the content of which is incorporated by reference herein in its entirety.FIELD

[0002] Some embodiments of the methods and compositions provided herein relate to detecting a homologous recombination deficiency (HRD) in a tumor sample from a subject, such as a breast or ovarian cancer tumor. Some embodiments include obtaining genome mapping (GM) data from the tumor sample, identifying structural variants (SVs) from the GM data, and determining the presence of absence of an HRD in the sample based on the SVs. Some embodiments also include determining a BRCA 1 / BRCA2-Yike status for a sample comprising an HDR.BACKGROUND

[0003] Homologous recombination deficiency (HRD) is an important feature to be recognized in cancers because it has direct implication in patient management and treatment options. HRD results from impairment of the homologous recombination repair (HRR) pathway, which ultimate role is maintaining genomic stability during DNA replication. HRD leads to genomic instability due to accumulation of improperly repaired replication associated double-strand breaks and is associated with cancer. HRD observed in cancers is includes inactivation of genes associated with HRR, including BRCA1 and BRCA2, and less frequently PALB2, RAD51 paralogs, such as RAD51C, RAD51B and RAD51D.

[0004] Germline HRR gene mutations can lead to a major predisposition to breast and ovarian carcinoma, including hereditary breast and ovarian cancer syndrome. HRD is found in approximately half of triple negative breast carcinomas (TNBC) which are hormone receptor negative with no over-expression of HER2; and high-grade serous ovarian carcinomas (HGOC), as well as at lower frequencies in other solid tumors.

[0005] The presence of an HRD in a tumor may affect a response to certain therapies. For example, patients with HRD HGOC have a better response to certain therapies including platinum-based chemotherapy, and PARP inhibitors in maintenance therapy. Prior strategies to identify HDR in tumors from patients with TNBC or HGOC have included whole genome sequencing (WGS), However, techniques such as WGS remain costly in terms of sequencing and data storage and demanding in terms of bioinformatics analysis. Therefore, there is a need for alternative methods and systems to identify tumors having an HRD.SUMMARY

[0006] Disclosed herein include embodiments of a method for determining the presence or absence of a homologous recombination deficiency (HRD) in a sample. In some embodiments, the method for determining the presence or absence of an HRD in a sample is under control of a processor (e g., a hardware processor or a virtual processor). The method can comprise receiving genome mapping (GM) data generated from a sample from a subject. The method can comprise determining a plurality of structural variants (SVs) in the GM data by comparing the GM data with a reference genome. The method can comprise determining the presence or absence of an HRD in the sample based on the plurality of SVs.

[0007] In some embodiments, the determining a plurality of SVs in the GM data comprises generating from the GM data a list comprising the plurality of SVs, and / or a map comprising the plurality of SVs. The determining a plurality of SVs in the GM data can comprise identifying SV types in the plurality of SVs. The SV types can be selected from deletions, duplications, inversions, and / or translocations. The determining a plurality of SVs in the GM data can comprise removing common SVs from the plurality of SVs by removing from the plurality of SVs an SV having a variant allele frequency (VAF) > 001 in a plurality of reference genomes. The plurality of references genomes can be from subjects lacking a reported disease phenotype.

[0008] In some embodiments, the determining a plurality of SVs in the GM data comprises determining the number of SVs in one or more SV types in the plurality of SVs. The one or more SV types can be selected from a deletion, a duplication, an inversion, and / or a translocation. The one or more SV types can be selected from an isolated translocation, a balanced translocation, an inverted duplication with a size in a range from 30 kb to 300 kb, an inversion, a deletion < 25 kb, and / or a duplication < 100 kb. The one or more SV types can be selected from an isolated translocation, a balanced translocation, an inverted duplication with a size in a range from 30 kb to 300 kb, and / or an inversion. Each end of an isolated translocation can be located at a distance from another translocation in an assembled map of the genome mapping data of at least 1.5 Mb, 2.0 Mb, 2.5 Mb, or 3.0 Mb. The distance can be at least 2.0 Mb.

[0009] In some embodiments, the determining the presence or absence of an HRD comprises applying to an HRD classifier the number of SVs in each of the one of more SV types to provide a probability that a sample comprises an HRD. The number of SVs in each of the one of more SV types can comprise the number of: (i) isolated translocations, (ii) balanced translocations, and (iii) inverted duplications with a size in a range from 30 kb to 300 kb together with the number of inversions. The method can comprise obtaining the HRD classifier. Obtaining the HRD classifier can comprise performing a least absolute shrinkage and selection operator (LASSO) logistic regression on the number of SVs in each SV type in a plurality of different SV types obtained from GM data from a plurality of control samples comprising an HRD and a plurality of control samples lacking an HRD. The SVs obtained from GM data from the plurality of control samples can lack an SV having a VAF > 001 in a plurality of reference genomes; optionally. The plurality of references genomes can be from subjects lacking a reported disease phenotype. The different SV types an be selected from duplications, inverted duplications, inversions, translocations, and / or deletions; optionally isolated translocations, balanced translocations, inverted duplications with a size in a range from 30 kb to 300 kb, deletions < 25 kb, and / or duplications < 100 kb.

[0010] In some embodiments, the determining the presence or absence of an HRD comprises comparing the number of SVs in one or more SV types in the plurality of SVs with the number of SVs in the one or more SV types in a control obtained from GM data from a sample lacking an HRD. An increase in the number of the SVs in the one or more SV types in the plurality of SVs compared to the number of SVs in the one or more SV types in the control can be indicative of the sample having an HRD. The one or more SV types can be selected from an isolated translocation, a balanced translocation, and / or a duplication < 100 kb. A decrease in the number of the SVs in the one or more SV types in the plurality of SVs compared to the number of SVs in the one or more SV types in the control can be indicative of the sample having an HRD. The one or more SV types can be selected from an inverted duplication with a size in a range from 30 kb to 300 kb, an inversion, and / or a deletion < 25 kb.

[0011] In some embodiments, the method further comprises determining the presence or absence of an (i) inactive BRCA1 gene or (ii) inactive BRCA2 or RAD51C gene in the sample. The sample can comprise an HRD. The determining the presence or absence of an (i) inactive BRCA1 gene or (ii) inactive BRCA2 or RAD51C gene in the sample can comprise applying to a BRCA1 / BRCA2-[ike classifier the number of SVs in each of the one of more SV types to provide a probability that a sample comprises an (i) inactive BRCA1 gene or (ii) inactive BRCA2 or RAD51C gene. The one of more SV types can comprise a deletion < 25 kb, and / or a duplication < 100 kb. Insome embodiments, the method comprises comprising comparing the number of SVs in one or more SV types in the plurality of SVs with the number of SVs in the one or more SV types in a control obtained from GM data from a sample lacking inactive BRCAl, BRCA2 and RAD51C genes. An increase in the number of the SVs in the one or more SV types in the plurality of SVs compared to the number of SVs in the one or more SV types in the control can be indicative of the sample having an HRD. The one or more SV types can comprise a duplication < 100 kb. A decrease in the number of the SVs in the one or more SV types in the plurality of SVs compared to the number of SVs in the one or more SV types in the control is indicative of the sample having an HRD. The one or more SV types can comprise a deletion < 25 kb.

[0012] Disclosed herein include embodiments of a method for determining the presence or absence of an (i) inactive BRCAl gene or (ii) inactive BRCA2 or RAD51C gene in a sample comprising a homologous recombination deficiency (HRD). In some embodiments, a method for determining the presence or absence of an (i) inactive BRCAl gene or (ii) inactive BRCA2 o RAD51C gene in a sample comprising an HRD is under control of a processor (e.g., a hardware processor or a virtual processor). The method can comprise receiving genome mapping (GM) data generated from the sample from a subject. The method can comprise determining a plurality of structural variants (SVs) in the GM data by comparing the GM data with a reference genome. The method can comprise determining the presence or absence of an (i) inactive BRCAl gene or (ii) inactive BRCA2 or RAD51C gene in the sample based on the plurality of SVs.

[0013] In some embodiments, the determining a plurality of SVs in the GM data comprises: (a) generating from the GM data a list comprising the plurality of SVs, and / or a map comprising the plurality of SVs. Alternatively or additionally, determining a plurality of SVs in the GM data can comprise (b) identifying SV types in the plurality of SVs. The SV types can be selected from deletions, duplications, inversions, and / or translocations. Alternatively or additionally, determining a plurality of SVs in the GM data can comprise (c) removing from the plurality of SVs an SV having a variant allele frequency (VAF)> 001 in a plurality of reference genomes. The plurality of references genomes can be from subjects lacking a reported disease phenotype. Alternatively or additionally, determining a plurality of SVs in the GM data can comprise (d) determining the number of SVs in one or more SV types in the plurality of SVs. In some embodiments, the one or more SV types is selected from a deletion, a duplication. The one or more SV types can be selected from a deletion < 25 kb, and / or a duplication < 100 kb. In some embodiments, the determining the presence or absence comprises applying to a BRCA 1 / BRCA2 ike classifier the number of SVs in each of the one of more SV types to provide a probability that a sample comprises an (i) inactive BRCAl gene or(ii) inactive BRCA2 or RAD51C gene. The one of more SV types can comprise a deletion < 25 kb, and / or a duplication < 100 kb. In some embodiments, the determining the presence or absence comprises comparing the number of SVs in one or more SV types in the plurality of SVs with the number of SVs in the one or more SV types in a control obtained from GM data from a sample lacking inactive BRCA1, BRCA2 and RAD51C genes, (i) An increase in the number of the SVs in the one or more SV types in the plurality of SVs compared to the number of SVs in the one or more SV types in the control can be indicative of the sample having an HRD. The one or more SV types can comprise a duplication < 100 kb. (ii) A decrease in the number of the SVs in the one or more SV types in the plurality of SVs compared to the number of SVs in the one or more SV types in the control can be indicative of the sample having an HRD. The one or more SV types can comprise a deletion < 25 kb.

[0014] In some embodiments, the subject is administered or prescribed a drug, such as a PARP inhibitor. In some embodiments, the method comprises administering or prescribing (or generating a prescription of) a drug, such as a PARP inhibitor, to the subject.

[0015] In some embodiments, the sample comprises cells, cell-free DNA, cell-free fetal DNA, amniotic fluid, a blood sample, a bone marrow sample, a biopsy sample, or a combination thereof. The sample can comprise one or more cancer cells (or a cancer sample). The sample can comprise one or more ovarian cancer cells (or an ovarian cancer sample) or one or more breast cancer cells (or a breast cancer sample). The sample can comprise one or more triple negative breast cancer (TNBC) cells (or a TNBC sample) or one or more high grade ovarian cancer (HGOC) cells (or a HGOC sample). The sample can comprise a formalin-fixed paraffin-embedded (FFPE) sample.

[0016] In some embodiments, the GM data comprises a consensus map for the sample from the subject. The GM data can be obtained by labelling genomic DNA from the sample from the subject with a detectable tag comprising a fluorophore, a quantum dot, a dendrimer, a nanowire, a bead, a hapten, streptavidin, avidin, neutravidin, biotin, a stabilized reactive group a radiolabel, an electromagnetic label, an azide, Dibenzocyclooctyne (DBCO), or an alkyne. The GM data can comprise optical genome mapping (OGM) data. The GM data can comprise electronic genome mapping (EGM) data.

[0017] In some embodiments, the method comprises: creating a file or a report and / or generating a ser interface (UI) comprising a UI element representing or comprising (i) the SV types in the GM data from the sample; (ii) the number of SVs in each SV type in the GM data; (iii) a probability that the sample comprises an HRD; and / or (iv) a probability that an HRD sample comprises an (a) inactive BRCA1 gene or (b) inactive BRCA2 or RAD51C gene.

[0018] Disclosed herein include embodiments of a system. In some embodiments, the system comprises non-transitory memory configured to store: executable instructions. The system can comprise a processor (e.g., a hardware processor or a virtual processor) in communication with the non-transitory memory. The processor can be programmed by the executable instructions to perform: any method (or one or more actions or steps thereof) disclosed herein.

[0019] Disclosed herein include embodiments of a system for determining the presence or absence of a homologous recombination deficiency (HRD) in a sample. In some embodiments, a system for determining the presence or absence of an HRD in a sample comprises: non-transitory memory. The non-transitory memory can be configured to store: executable instructions. The non-transitory memory can be configured to store: genome mapping (GM) data generated from a sample from a subject. The system can comprise a processor (e.g., a hardware processor or a virtual processor) in communication with the non-transitory memory. The processor can be programmed by the executable instructions to perform: determining a plurality of structural variants (SVs) in the GM data by comparing the GM data with a reference genome. The processor can be programmed by the executable instructions to perform: determining the presence or absence of an HRD in the sample based on the plurality of SVs.

[0020] In some embodiments, the processor is programmed by the executable instructions to perform: generating from the GM data a list comprising the plurality of SVs, and / or a map comprising the plurality of SVs . The non-transitory memory can be configured to store: the list and / or the map. The processor can be programmed by the executable instructions to perform: identifying SV types in the plurality of SVs; optionally. The SV types can be selected from deletions, duplications, inversions, and / or translocations. The processor can be programmed by the executable instructions to perform: removing common SVs from the plurality of SVs by removing from the plurality of SVs an SV having a variant allele frequency (VAF) > 001 in a plurality of reference genomes. The plurality of references genomes can be from subjects lacking a reported disease phenotype.

[0021] In some embodiments, the processor is programmed by the executable instructions to perform: determining the number of SVs in one or more SV types in the plurality of SVs. The one or more SV types can be selected from a deletion, a duplication, an inversion, and / or a translocation. The one or more SV types can be selected from an isolated translocation, a balanced translocation, an inverted duplication with a size in a range from 30 kb to 300 kb, an inversion, a deletion < 25 kb, and / or a duplication < 100 kb. The one or more SV types can be selected from an isolated translocation, a balanced translocation, an inverted duplication with a size in a range from 30 kb to 300 kb, and / or an inversion. Each end of an isolated translocation can be located at a distance fromanother translocation in an assembled map of the genome mapping data of at least 1.5 Mb, 2.0 Mb, 2.5 Mb, or 3.0 Mb. The distance can be at least 2.0 Mb.

[0022] In some embodiments, the non-transitory memory configured to store an HRD classifier. The processor can be programmed by the executable instructions to perform: applying to an HRD classifier the number of SVs in each of the one of more SV types to provide a probability that a sample comprises an HRD. The number of SVs in each of the one of more SV types can comprise the number of: (i) isolated translocations, (ii) balanced translocations, and (iii) inverted duplications with a size in a range from 30 kb to 300 kb together with the number of inversions. The HRD classifier can be obtained by performing a least absolute shrinkage and selection operator (LASSO) logistic regression on the number of SVs in each SV type in a plurality of different SV types obtained from GM data from a plurality of control samples comprising an HRD and a plurality of control samples lacking an HRD. The SVs obtained from GM data from the plurality of control samples can lack an SV having a VAF > 001 in a plurality of reference genomes. The plurality of references genomes can be from subjects lacking a reported disease phenotype. The different SV types can be selected from duplications, inverted duplications, inversions, translocations, and / or deletions; optionally isolated translocations, balanced translocations, inverted duplications with a size in a range from 30 kb to 300 kb, deletions < 25 kb, and / or duplications < 100 kb.

[0023] In some embodiments, the processor is programmed by the executable instructions to perform: comparing the number of SVs in one or more SV types in the plurality of SVs with the number of SVs in the one or more SV types in a control obtained from GM data from a sample lacking an HRD. An increase in the number of the SVs in the one or more SV types in the plurality of SVs compared to the number of SVs in the one or more SV types in the control can be indicative of the sample having an HRD. The one or more SV types can be selected from an isolated translocation, a balanced translocation, and / or a duplication < 100 kb. A decrease in the number of the SVs in the one or more SV types in the plurality of SVs compared to the number of SVs in the one or more SV types in the control can be indicative of the sample having an HRD. The one or more SV types can be selected from an inverted duplication with a size in a range from 30 kb to 300 kb, an inversion, and / or a deletion < 25 kb.

[0024] In some embodiments, the processor is programmed by the executable instructions to perform: determining the presence or absence of an (i) inactive BRCA1 gene or (ii) inactive BRCA2 or RAD51C gene in the sample. The sample can comprise an HRD. In some embodiments, the non-transitory memory configured to store: a BRCAl / BRCA2-like classifier. The processor can be programmed by the executable instructions to perform: applying to a BRCA1 / BRCA2Aike classifierthe number of SVs in each of the one of more SV types to provide a probability that a sample comprises an (i) inactive BRCA 1 gene or (ii) inactive BRCA2 or RAD51C gene. The one of more SV types can comprise a deletion < 25 kb, and / or a duplication < 100 kb. In some embodiments, the processor is programmed by the executable instructions to perform: comparing the number of SVs in one or more SV types in the plurality of SVs with the number of SVs in the one or more SV types in a control obtained from GM data from a sample lacking inactive BRCA1, BRCA2 and RAD51C genes. An increase in the number of the SVs in the one or more SV types in the plurality of SVs compared to the number of SVs in the one or more SV types in the control can be indicative of the sample having an HRD. The one or more SV types can comprise a duplication < 100 kb. A decrease in the number of the SVs in the one or more SV types in the plurality of SVs compared to the number of SVs in the one or more SV types in the control can be indicative of the sample having an HRD. The one or more SV types can comprise a deletion < 25 kb.

[0025] Disclosed herein include embodiments of a system for determining the presence or absence of an (i) inactive BRCA1 gene or (ii) inactive BRCA2 or RAD51C gene in a sample comprising a homologous recombination deficiency (HRD). In some embodiments, a system for determining the presence or absence of an (i) inactive BRCA1 gene or (ii) inactive BRCA2 ovRAD51C gene in a sample comprising an HRD comprises: non-transitory memory configured to store: executable instructions. The non-transitory memory can be configured to store: genome mapping (GM) data generated from a sample from a subject. The system can comprise a processor (e.g., a hardware processor or a virtual processor) in communication with the non-transitory memory. The processor can be programmed by the executable instructions to perform: determining a plurality of structural variants (SVs) in the GM data by comparing the GM data with a reference genome. The processor can be programmed by the executable instructions to perform: determining the presence or absence of an (i) inactive BRCA1 gene or (ii) inactive BRCA2 or RAD51C gene in the sample based on the plurality of SVs.

[0026] In some embodiments, the processor is programmed by the executable instructions to perform: (a) generating from the GM data a list comprising the plurality of SVs, and / or a map comprising the plurality of SVs. The non-transitory memory can be configured to store the list and / or the map. Alternatively or additionally, the processor is programmed by the executable instructions to perform: (b) identifying SV types in the plurality of SVs. The SV types can be selected from deletions, duplications, inversions, and / or translocations. Alternatively or additionally, the processor is programmed by the executable instructions to perform: (c) removing common SVs from the plurality of SVs by removing from the plurality of SVs an SV having a variant allele frequency (VAF) > 001in a plurality of reference genomes. The plurality of references genomes can be from subjects lacking a reported disease phenotype. Alternatively or additionally, the processor is programmed by the executable instructions to perform: (d) determining the number of SVs in one or more SV types in the plurality of SVs.

[0027] In some embodiments, the one or more SV types is selected from a deletion, a duplication. The one or more SV types can be selected from a deletion < 25 kb, and / or a duplication < 100 kb. In some embodiments, the non-transitory memory configured to store: a BRCA1 / BRCA2-like classifier. In some embodiments, the processor is programmed by the executable instructions to perform: applying to a BRCAHBRCA2-Y\ e classifier the number of SVs in each of the one of more SV types to provide a probability that a sample comprises an (i) inactive BRCA1 gene or (ii) inactive BRCA2 or RAD51C gene. The one of more SV types can comprise a deletion < 25 kb, and / or a duplication < 100 kb. In some embodiments, the processor is programmed by the executable instructions to perform: comparing the number of SVs in one or more SV types in the plurality of SVs with the number of SVs in the one or more SV types in a control obtained from GM data from a sample lacking inactive BRCA1, BRCA2 and RAD51C genes; optionally, (i) An increase in the number of the SVs in the one or more SV types in the plurality of SVs compared to the number of SVs in the one or more SV types in the control can be indicative of the sample having an HRD. The one or more SV types can comprise a duplication < 100 kb. (ii) A decrease in the number of the SVs in the one or more SV types in the plurality of SVs compared to the number of SVs in the one or more SV types in the control can be indicative of the sample having an HRD. The one or more SV types can comprise a deletion < 25 kb.

[0028] In some embodiments, the subject is administered or prescribed a drug, such as a PARP inhibitor. In some embodiments, the processor is programmed by the executable instructions to perform: generating a prescription of a drug, such as a PARP inhibitor, to the subject.

[0029] In some embodiments, the sample comprises cells, cell-free DNA, cell-free fetal DNA, amniotic fluid, a blood sample, a bone marrow sample, a biopsy sample, or a combination thereof. The sample can comprise one or more cancer cells (or a cancer sample). The sample can comprise one or more ovarian cancer cells (or an ovarian cancer sample) or one or more breast cancer cells (or a breast cancer sample). The sample can comprise one or more triple negative breast cancer (TNBC) cells (or a TNBC sample) or one or more high grade ovarian cancer (HGOC) cells (or a HGOC sample). The sample can comprise a formalin-fixed paraffin-embedded (FFPE) sample.

[0030] In some embodiments, the GM data comprises a consensus map for the sample from the subject. The GM data can be obtained by labelling genomic DNA from the sample from thesubject with a detectable tag comprising a fluorophore, a quantum dot, a dendrimer, a nanowire, a bead, a hapten, streptavidin, avidin, neutravidin, biotin, a stabilized reactive group a radiolabel, an electromagnetic label, an azide, dibenzocyclooctyne (DBCO), or an alkyne. The GM data can comprise optical genome mapping (OGM) data. The GM data can comprise electronic genome mapping (EGM) data.

[0031] In some embodiments, the processor is programmed by the executable instructions to perform: creating a fde or a report and / or generating a user interface (UI) comprising a UI element representing or comprising (i) the SV types in the GM data from the sample; (ii) the number of SVs in each SV type in the GM data; (iii) a probability that the sample comprises an HRD; and / or (iv) a probability that an HRD sample comprises an (a) inactive BRCA1 gene or (b) inactive BRCA2 or RAD51C gene.

[0032] Also disclosed herein include non-transitory computer-readable media. In some embodiments, non-transitory computer-readable medium stores executable instructions, when executed by a system (e.g., a computing system), causes the system to perform any method or one or more actions (or steps) of any method disclosed herein.

[0033] Details of one or more implementations of the subject matter described in this specification are set forth in the accompanying drawings and the description below. Other features, aspects, and advantages will become apparent from the description, the drawings, and the claims. Neither this summary nor the following detailed description purports to define or limit the scope of the inventive subject matter.BRIEF DESCRIPTION OF THE DRAWINGS

[0034] FIG. 1 depicts a schematic example of an OGM workflow.

[0035] FIG. 2 is a flow diagram showing an example method of determining the presence or absence of HDR in a sample from a subject using GM data.

[0036] FIG. 3 is a flow diagram showing an example method of determining the presence or absence of HDR in a sample from a subject using GM data.

[0037] FIG. 4 is a flow diagram showing an example method of determining the BRCALBRCA2- ike status of a sample from a subject using GM data.

[0038] FIG. 5 is a flow diagram showing an example method of determining the BRCAJ / BRCA2-Yike status of a sample from a subject using GM data.

[0039] FIG. 6 is a block diagram of an illustrative computing system configured to implement methods of determining the presence or absence of HDR and / or the BRCA1 BRCA2-\ke status in a sample from a subject using GM data, and / or .

[0040] FIG. 7 depicts flowcharts of a training series (left panel) and a validation series to detect homologous recombination deficient (HRD) triple negative breast cancer (TNBC) tumors or high-grade ovarian cancer (HGOC) tumors.

[0041] FIG. 8 depicts circus plots of optical genome mapping for four representative examples of tumors using a rare variant analysis (RVA) including two examples of HRD tumors (left panels), two examples of non-HRD (nonHRD) tumors (right panels)

[0042] FIG. 9A depicts a density plot showing the size distribution of duplication, duplication-split and insertion events called by the RVA in the training series.

[0043] FIG. 9B depicts a manual re-alignment showing a rearrangement called insertion by the RVA manually re-classified as a duplication. The upper bar with single arrow depicts a references genome. The lower bar with two arrows depicts results from a tumor sample in which the right-most arrow includes an inserted sequence

[0044] FIG. 9C depicts a representation of a duplication split. The lower panel depicts the duplication split. The upper panel depicts a copy number profile with gain indicated in the shaded portion and absence of gain in the non-shaded portion.

[0045] FIG. 10A depicts a graph for size distribution of duplications (DUP) events in HRD and nonHRD tumors from the training series, BRCA1 inactivated tumors, and BRCA2 and RAD51 C inactivated tumors (6AG42-like). The dashed vertical lines represent the thresholds for categorizing the events by size. In FIGs 6A-6D, HRD (‘*’); nonHRD (‘#’); BRCAl (dashed lined); and BRCA2-Wkc (dotted line).

[0046] FIG. 10B depicts a graph for size distribution of deletions (DEL) events in HRD and nonHRD tumors from the training series, BRCAl inactivated tumors, and B7?G42-like tumors. The dashed vertical lines represent the threshold for categorizing the events by size.

[0047] FIG. 10C depicts a graph for size distribution of deletions (INV) events in HRD and nonHRD tumors from the training series, BRCAl inactivated tumors, and BRCA2A e, tumors.

[0048] FIG. 10D depicts a graph for size distribution of duplicati on inverted (DUP INV) events in HRD and nonHRD tumors from the training series, BRCAl inactivated tumors, and BRCA2-like tumors.

[0049] FIG. 10E depicts a graph for a correlation between the number of INV and DU P _INV called by the RVA in the training series.

[0050] FIG. 10F depicts boxplots representing the number of DUP and DEL in HRD and nonHRD tumors.

[0051] FIG. 10G depicts boxplots representing the number of DUP, DEL and inversion and duplication inverted (DUP INVplus) in BRCA J and BRCA2-like tumors.

[0052] FIG. 10H depicts a graph showing a correlation between the number of isolated translocations (Isolated TRA) and the number of LGA detected by shallowHRD for the training series.

[0053] FIG. 101 depicts boxplots representing the number of Isolated- TRA in BRCAI-like and BACA2-like, depending on tumor type (TNBC and HGOC).

[0054] FIG. 11A depicts boxplots for the total number of structural variants (SV) in identified classes for: duplication (DUP), deletion (DEL), inversion and duplicati on inverted (DUP INVplus), and translocation (TRA), for HRD tumors and nonHRD tumors. In FIGs 3 A, 3B and 3E, a Wilcoxon Mann- Whitney nonparametric test P is indicated as **, ***, **** for P <10'2, P <10'3and P < 1 O'4, respectively.

[0055] FIG. 1 IB depicts boxplots for the total number of SVs in identified DUP and DEL classes for HRD BRCA1 inactivated (BRCA!) tumors, and HRD BRCA2 and RAD5IC inactivated (B7?C42-like) tumors.

[0056] FIG. 11C depicts a schematic representation of TRA on a chromosome including a non-isolated TRA (left panel, and indicated by ‘ *’), and isolated TRA (right panel, and indicated by ‘ *’). Double arrows represent the distance between TRA breakpoints.

[0057] FIG. 1 ID depicts a graph of number of isolated TRA in HRD and nonHRD tumors with regard to the neighborhood of isolation X, from O to 5 Megabases (Mb). The arrow represents the X used to define isolated TRA.

[0058] FIG. 1 IE depicts boxplots representing the number of isolated TRA and balanced TRA, in HRD and nonHRD tumors.

[0059] FIG. 12A depicts a classification model with coefficients from a LASSO logistic regression model for an HRD classifier. In FIGs 4A and 4B, the dashed lined represents coefficients equal to zero. Classes included duplication (DUP); deletion (DEL); inversion and duplication inverted (DUP INVplus); isolated translocations (isolated- TRA); and balanced translocations (balanced- TRA).

[0060] FIG. 12B depicts a classification model with coefficients from a LASSO logistic regression model for a BRCA 1 classifier.

[0061] FIG. 12C depicts a prediction of the HRD classifier in which the upper and lower pabels represent tumors classified as HRD and nonHRD by shallowHRD, respectively; the x-axis represents the probability of being HRD. The dashed lines represent thresholds of the HRD classification in which: left of the left-most dashed line: nonHRD; between dashed lines: borderline; and right of the right-most dashed line: HRD. The arrow represents the sample classified as nonHRD by shallowHRD and HRD by OGM, with methylation of RAD51C promoter (meRAD51C).

[0062] FIG. 12D depicts a prediction of the BRCA1 classifier, the upper and lower blocks represent HRD tumors with a BRCAI inactivation and HRD tumors with a BRCA2 or RAD51C inactivation (BAG42-like), respectively. The x-axis represents the probability of being BRCAI inactivated. Red dashed lines represent thresholds of the BRCAI classification: left to the left line: BRCA2-\\ e.ibetween lines: borderline, and right to the right line: BRCAI.

[0063] FIG. 13A depicts boxplots representing the number of inversion (INV) and duplication inverted (DUP INV) variants in HRD and nonHRD tumors.

[0064] FIG. 13B depicts boxplots representing the number of INV and DUP INV between HGOC and TNBC. Discordant cases between OGM and shallowHRD were removed from these analyses. Wilcoxon MannWhitney nonparametric test T3is indicated as ** and *** for <10‘2and <10‘3, respectively.

[0065] Throughout the drawings, reference numbers may be re-used to indicate correspondence between referenced elements. The drawings are provided to illustrate example embodiments described herein and are not intended to limit the scope of the disclosure.DETAILED DESCRIPTION

[0066] In the following detailed description, reference is made to the accompanying drawings, which form a part hereof. In the drawings, similar symbols typically identify similar components, unless context dictates otherwise. The illustrative embodiments described in the detailed description, drawings, and claims are not meant to be limiting. Other embodiments can be utilized, and other changes can be made, without departing from the spirit or scope of the subject matter presented herein. It will be readily understood that the aspects of the present disclosure, as generally described herein, and illustrated in the Figures, can be arranged, substituted, combined, separated, and designed in a wide variety of different configurations, all of which are explicitly contemplated herein and made part of the disclosure herein.

[0067] All patents, published patent applications, other publications, and sequences from GenBank, and other databases referred to herein are incorporated by reference in their entirety with respect to the related technology.

[0068] Some embodiments of the methods and compositions provided herein relate to detecting a homologous recombination deficiency (HRD) in a tumor sample from a subject, such as a breast or ovarian cancer tumor. Some embodiments include obtaining genome mapping (GM) data from the tumor sample, identifying structural variants (SVs) from the GM data, and determining the presence of absence of an HRD in the sample based on the SVs. Some embodiments also include determining a BRCA J / BRCA2 -like status for a sample comprising an HDR.

[0069] In some embodiments, GM data is generated from a sample. The GM data can include a consensus map representing the genome of the sample. In some embodiments, SVs are identified in the GM data by comparing the GM data with a reference genome. A list of the SVs or map of the SVs can be generated based on the identified SVs. In some embodiments, rare SVs are identified in the GM data, such as by removing common SVs, such as an SV having a variant allele frequency (VAF) > 001 in a plurality of reference genomes. In some embodiments, different types of SVs, such as SVs in the list or the map, are identified and counted. The total number of certain SV types can be applied to an HRD classifier to generate a probability that the sample comprises an HRD. In some embodiments, the SV types can include the number of (i) isolated translocations, (ii) balanced translocations; and (iii) inverted duplications with a size in a range from 30 kb to 300 kb, together with the number of inversions. An isolated translocation can include an SV type in which a translocation is identified, and each end of an isolated translocation is located at a distance from another translocation in an assembled map of the genome mapping data of at least 1.5 Mb, 2.0 Mb, 2.5 Mb, or 3.0 Mb. In some embodiments, a BRCAJ / BRCA2-iike status of a tumor comprising an HRD can be determined by applying the total number of certain SV types to a BRCA1 / BRCA2 classifier. In some such embodiments the SV types can include (i) deletions < 25 kb; and duplications < 100 kb.

[0070] HRD leads to genomic instability resulting in specific genomic alterations, such as genomic scar. One strong component of this scar include SVs, which can be called by high coverage whole genome sequencing (WGS). Some embodiments provided herein include GM, such as optical genome mapping (OGM) to call SVs to provide a simple cost-effective and computationally efficient model for HRD testing, as compared to techniques such as WGS. Based on the training set of 41 samples of TNBC (n=26) and HGOC (n=l 5) the features associated with known HRD status of tumors were selected from the OGM readout. Isolated and balanced translocations were foundstrongly associated with HRD, while tandem inverted duplications and inversions were enriched in non-HRD (nonHRD) cases. A linear HRD classifier using these features had an accuracy of at least 94% in a validation series of 31 TNBC and HGOC. This HRD classifier corrected at least one case of the validation series to a subsequently confirmed HRD status. Consistent with previous studies, duplications and deletions of less 25 kb in size were found enriched in BRCA1- and BRCA2-inactivated tumors, respectively. A preliminary sub-classifier of BAC47-inactivated tumors based on these two features had an 80% accuracy in the validation series. Embodiments provided herein include GM (such as OGM) as a fast, cost-effective, and accurate technique to determine HRD status in TNBC and HGOC, tolerating low tumor content due to its ultra-high coverage. In some embodiments, a subject whose HRD status in TNBC or HGOC is determined can be administered a drug, such as a PARP inhibitor.

[0071] HRD genomic instability scar can include a pattern of somatic alterations accumulated during tumor genome evolution due to impaired HRR pathways. This HRD pattern can be specific and can be recognized based on different readouts from measured genomic alterations. Historically the first genomic signatures of HRD were based on allele specific copy number alterations (CNA) readout, which were compiled in the MyChoice CDx® Plus HRD assay by Myriad Genetics, which is an FDA-approved test used now in clinics. The test sums up the number of largescale state transitions (LSTs), such as CNA breaks between chromosome fragments of more than 10 Mb; the number of large chromosome fragments with loss of heterozygosity (LOH); and the number of chromosome arms with telomeric allelic imbalance (TAI) into an HRD score, where HRD is called if the score is at least 42. Within the same framework of mining CNA readout, a new generation of HRD tests emerged based on low-pass shallow whole genome sequencing (sWGS). CNA profiles obtained from sWGS can be used to generate large genomic alterations (LGA) score, genomic instability score (GIS) or various direct binned profile classification methods for HRD calling.

[0072] Whatever the method or technique used for HRD classification based on CNA readout, the performance would be around 85-90% because of the highly specific genomic scar of HRD. The remaining 10-15% cases could not be efficiently resolved based on CNA readout because of borderline scores and absence of a definitive biological classifier (ground truth for HRD status and CNA profile).

[0073] Full-coverage WGS may resolve complicated cases. WGS of a large set of breast cancers detailed the genomic scar of HRD from single-base substitutions (SBS) and small indels to the wide range of SVs, including inter-chromosomal translocations, deletions, tandem duplications(TD) and inversions. Among the genomic alterations most significantly associated with HRD include a signature SBS, small deletions (~10 bp) with microhomology at the ends, deletions smaller than 100 kb, and TD smaller than 100 kb associated specifically with BRCA / -mutated tumors. Clustered SVs, TD of more than 100 kb were the main features found anticorrelated with HRD. Using the full spectrum of genomic rearrangements extracted from WGS, the associated diagnostic pipeline HRDetect is 100% precise in HRD calling. However, WGS remains costly in terms of sequencing and data storage and demanding in terms of bioinformatics analysis. SV calling from WGS is still a challenging bioinformatics task requiring efficient false positive / germline SV filtering, complex algorithms for decomposing clustered events and is very sensitive to the tumor content, which is often compromised in the clinical settings. However, SV readout is highly specific for HRD and its deployment in resolving complicated cases not only improves clinical testing but also brings new insights into HRD cancer biology. Thus there is a need for alternative methods and systems for identifying HRD.

[0074] Embodiments of the methods and compositions provided herein include GM, such as OGM which detects HRD based on structural variants. Such embodiments are cost-effective alternatives to WGS and other prior methods. OGM is a non-sequencing-based genome-wide visualization technique with high coverage, able to detect structural rearrangements. OGM has already been successfully used to detect germline and somatic structural rearrangements, especially in hematological malignancies. Embodiments provided herein include an HRD classifier based on direct SV output from OGM which has been validated this classifier in a series of TNBC and HGOC.Optical genome mapping

[0075] FIG. 1 illustrates a non-limiting exemplary workflow of OGM. The data generated from OGM can be used to determine BAF values of SNPs for a sample. The OGM workflow can start with mega-base size DNA isolation, e.g., 150 kbp or longer. A single enzymatic reaction can label the genome at a specific sequence motif occurring, e.g., approximately 15 times per 100 kbp in the human genome. The long, labeled DNA molecules can be linearized in nanochannel arrays (e.g., provided by a cartridge or chip, such as a Saphyr Chip®, Bionano Genomics, Inc. (San Diego, CA)) and imaged in an automated manner by an OGM instrument (e.g., Saphyr® System, Bionano Genomics, Inc. (San Diego, CA)). The molecules can be assembled into local maps or whole genome maps. Changes in patterning or spacing of the labels can be detected, genome-wide, to call structural variants.

[0076] OGM is an imaging technology which evaluates the fluorescent labeling pattern of individual DNA molecules to perform an unbiased assessment of genome-wide structural variants down to, e.g., 500 base pairs (bp) in size, a resolution that far exceeds conventional cytogenetic approaches. OGM can rely on a specifically designed extraction protocol facilitating the isolation of high molecular weight (HMW) or ultra-high molecular weight (UHMW) DNA. This protocol can, in some embodiments, utilize a paramagnetic disk purposed with trapping DNA for wash steps thereby reducing sheering forces present in standard column-based extraction methods. The result can be DNA fragments (or molecules) of about 150 kilobases (kbp) to megabases (Mbp) in size, about 5-10X longer than the average fragment size from conventional DNA isolations techniques. Referring to FIG. 1, DNA can be fluorescently labeled via covalent modification at a motif (which can be 4, 5, 6, 7, 8, 9, 10, or more nucleotides in length), such as a hexamer motif (e.g., the CTTAAG hexamer motif), generating genome-wide density of a number of labels per 100 kb in sequence specific patterns (e.g., approximately 14-17 labels per 100 kb, or 10, 11, 12, 13, 14, 15, 16, 17, 18, 19, 20, or more or fewer labels per 100 kb). Labeled DNA can be loaded on chips (e.g., silicon chips) composed of hundreds of thousands of parallel nanochannels where individual DNA molecules are linearized, imaged, and digitized. The specific labeling profile of individual DNA molecules, including spacing and pattern of hexamers labels, can be subsequently grouped based on similarity, producing about 500 kbp (or longer or shorter, such as 300 kbp, 400 kbp, 500 kbp, 600 kbp, 700 kbp, 800 kbp, 900 kbp, 1000 kbp) to megabase-sized consensus maps, which can be compared in silico to the expected labeling pattern of a reference genome (FIG. 1). This imaging technology converts DNA into a “barcode” whose labeling profile and characteristics can sensitively and specifically resolve copy number and structural variation without the need for sequence level data (FIG. 1). The quality of the DNA, including both size and labeling characteristics, as well as the number of images captured can influence genome-wide coverage. For example, each flow cell, which can accommodate a single specimen, can generate, for example, up to 5000 Gigabase pairs (Gbp) of raw data (or 3000 Gbp, 4000 Gbp, 5000 Gbp, 6000 Gbp, 7000 Gbp, 8000 Gbp, 9000 Gbp, 10000 Gbp, or more or less, of raw data), achieving a maximum theoretical genome-wide coverage of about 1250x (or 500x, 750x, lOOOx, 1250x, 1500x, 1750x, 2000x, or more or less). Bioinformatics analyses can be performed. Example bioinformatics analysis can include: de novo structural variant analysis for typical germline assessments (e.g., greater than about 80x-coverage; requiring greater than about 400 Gbp data collection) or ‘Rare Variant Analysis (RVP)’ for somatic assessment down to a ~5% variant allele fraction (e.g., greater than about 340x coverage; requiring greater than about 1500 Gbp data). Bothalgorithms facilitate the detection of a wide array of structural variants; from copy number gains / losses to balanced / unbalanced translocations and insertions to inversions.

[0077] OGM can be used to analyze large eukaryotic genomes and their structural features at a high resolution. OGM uses linearized strands of HMW or UHMW DNA that are far longer than the DNA sequences analyzed in current second- and third-generation sequencing methods, achieving average read lengths in excess of 200 kbp. The usage of long molecules in OGM can allow repetitive regions and other regions that are complicated to map to be spanned more easily than with short molecules. This leads to the creation of maps that may cover the whole arm of a chromosome and yet allow the detection of insertions and deletions as small as 500 bp (or longer or shorter, such as 300 kbp, 400 kbp, 500 kbp, 600 kbp, 700 kbp, 800 kbp, 900 kbp, 1000 kbp) other SVs may need to be 30 kbp (or 10 kbp, 20k kbp, 30 kbp, 40 kbp, or 50 kbp)) or larger to be detectable. OGM can be used to, for example, detect the breakpoints of chromosomal translocations, for the diagnosis of facioscapulohumeral muscular dystrophy (FSHD). OGM may be used as a cytogenomic tool for prenatal diagnostics

[0078] Extraction / Isolation. UHMW DNA can be extracted for OGM, for example. UHMW DNA extraction can be done using isolation kits, such as kits from Bionano Genomics, Inc. (San Diego, CA). In some embodiments, DNA from approximately 1.5 x 106cells (or 1 * 105, 1.5 * 105, 2.5 x 105, 5 x 105, 7.5 x 105, 1 x 106, 1.5 x 106, 2.5 x 106, 5 x 106, 7.5 x 106, 1 x 107or more or fewer cells) can be extracted. The extraction can include immobilizing cells in agarose plugs and lysing the immunized cells by proteinase K; thereafter. The extraction can include washing, recovering, and quantifying the genomic DNA. Alternatively or additionally, the genomic DNA can be bound to a magnetic disk. Subsequently, the DNA can be washed, recovered, and quantified.

[0079] Labeling and Processing. A sufficient quantity of UHMW DNA (e.g., 250 ng, 500 ng, 750 ng, 1000 ng, 1250 ng, 1500 ng, 1750 ng, 2000 ng, or more UHMW DNA) can be labeled with a fluorophore. Such labeling can be done using a methyltransferase, such as the methyltransferase direct labeling enzyme (DLE-1) at the recognition motif of the methyltransferase, such as CTTAAG. This can generate a number of labels per 100 kbp (e.g., approximately 14-15 labels per 100 kbp, or 10, 11, 12, 13, 14, 15, 16, 17, 18, 19, 20, or more or less labels per kbp) when labeling human genomic DNA. In some embodiments, such labeling can be done using another enzyme (e.g., an endonuclease) at the recognition motif of the enzyme (e.g., GCTCTTCN of endonuclease Nt.BspQI).

[0080] Thereafter, the DNA can be dialyzed, its backbone stained, and finally the prepared DNA can be applied to flow cells (e.g., G1.2 flow cells from Bionano Genomics, Inc.) The flow cellcan then be inserted into an OGM instrument, such as the Saphyr® instrument from Bionano Genomics, Inc. In the instrument, the DNA can be fed by electrophoresis into the nanochannels of the flow cell for linearization. DNA-filled nanochannels can be scanned using, for example, a fluorescence microscope. The captured images can be converted to electronic representations of the DNA molecules. The virtual DNA strands can then filtered and de novo assembled into maps (FIG.1).

[0081] OGM Data Assembly. The data acquired with the OGM instrument can be processed. For example, the raw data can be filtered for a minimum length of 150 kbp (or 100 kbp, 125 kbp, 150 kbp, 175 kbp, 200 kbp, or more) and minimum of nine labels (or 5, 6, 7, 8, 9, 10, 11, 12, 13, 14, 15, or more labels) per molecule (or fragment). The filtered molecules can be assembled, e.g., with de novo assembly. The consensus maps of the molecules can be aligned to a reference genome sequence, such as the human reference genome GRCh38. Variants can be detected. Variants detection can be performed using, for example, an SV pipeline, comparing the maps to the aligned reference genome. There, patterns of markers from the maps deviating from the reference become apparent. Variants detections can be performed using, for example, a CNV pipeline,” which quantifies the mapped molecules and hence is able to detect gains and losses of several hundred kbp in size.

[0082] The results of the SV pipeline can then be augmented by, for example, a variant annotation pipeline, which adds quality metrics for the called variants and supplies their estimated frequency in the human population based on an internal database. The optional step of filtering based on the frequency of the SVs in the internal database may (or may not) be used in some implementations. The SVs can be detected or called. Automatic calling can be based on the confidence scores and sizes of the SVs (insertions and deletions: confidence > 0, size > 500 bp; inversions: confidence > 0.7, size > 30 kbp; duplications: confidence = -1, size > 30 kbp; intrachromosomal translocations: confidence > 0.3; interchromosomal translocations: confidence > 0.65; CNV confidence > 0.99, size > 500 kbp). Additionally, each called SV can be required to be spanned by > 5 strands of DNA.

[0083] The total amount of unfiltered DNA scanned by the OGM system can be, or be about, 750 Gbp, 800 Gbp, 850 Gbp, 900 Gbp, 916 Gbp, 925 Gbp, 950 Gbp, 1000 Gbp, 1250 Gbp, or more, per sample on average. An effective coverage of the reference can be, or can be greater than, 40*, 50*, 60*, 70X580*, 90xpor more, per sample. The effective coverage of the reference can be defined as the total length of filtered (>150 kbp) and aligned molecules divided by the length of the reference genome after de novo assembly

[0084] Further details regarding various aspects of OGM can be found in U.S. Pat. Nos.11,359,244; 11,292,713; 11,291,999; 10,995,364; 10,844,424; 10,676,352; 10,669,586; 10,654,715; 10,435,739; 10,247,700; 10,000,804; 10,000,803; 9,845,238; 9,809,855; 9,804,122; 9,725,315; 9,536,041; 9,533,879; 9,310,376; 9,181,578; 9,061,901; 8,722,327; and 8,628,919; U.S. Pat.App. Pub. No. 20250104808; and PCT App. Pub. Nos. WO 2020 / 005846; WO 2016 / 036647; WO 2015 / 134785; WO 2015 / 130696; WO 2015 / 126840; WO 2015 / 017801; WO 2014 / 200926; WO 2014 / 130589; WO 2014 / 123822; WO 2013 / 036860; WO 2012 / 054735; WO 2011 / 050147; WO 2011 / 038327 and WO 2010 / 13532; the content of each of which is incorporated herein by reference in its entirety.Electronic genome mapping

[0085] For electronic genome mapping (EGM), HMW or UHMW DNA molecules (e. ., 50 kbp to 500 kbp) can be isolated from a sample (e.g., a cell sample, a blood sample). The isolated DNA molecules can be labelled at known recognition sites. Labeling can include DNA nicking translocation and label (or tag) insertion. The recognition sites can be, for example, 4 kbp apart on average. A DNA binding protein e.g., RecA) can be used to stiffen the DNA molecules. An EGM chip (also referred to as an EGM detector) can comprise solid-state nanochannels (e.g., 256 parallel nanochannels), each with its own electronic sensor. The EGM chip can be in an EGM instrument. The labeled DNA molecules can be injected into the EGM chip. Single DNA molecules can be electrophoretically moved through a nanochannel. Single DNA molecules can be electrophoretically moved through nanochannels at the same time. In a nanochannel, the labels (or tags) of a DNA molecule can be electronically detected by changes in resistance, which can be inferred from changes in voltage. When a DNA molecule enters the nanochannel, it blocks the current that can go through the channel and can be measured as a voltage change. When a label (or tag) is also present on the DNA molecule, the current is further reduced resulting in a sharp signal. The voltage can be measured as a function of time so the time that a nanochannel is empty, the time it is occupied by untagged DNA, and the time each label (or tag) goes through the nanochannel can be determined. The times between voltage peaks correspond to distances between labels on a DNA molecule. These times can be converted to distances for each DNA molecule. The results can include single molecule maps with the location of each label (or tag). The single molecule maps for a single sample can be assembled into local maps or a whole genome map of all labeled (or tagged) locations for the sample which can be aligned against a reference genome. Analysis such as structural variant (SV) analysis can be performed.Certain methods for determining an HRD status of a sample

[0086] Some embodiments provided herein include methods for determining the presence or absence of an HRD in a sample. Some such embodiments include computer-implemented methods, such as methods comprising, under control of a hardware processor: receiving GM data generated from a sample from a subject; determining a plurality of SVs in the GM data by comparing the GM data with a reference genome; and determining the presence or absence of an HRD in the sample based on the plurality of SVs.

[0087] In some embodiments, the determining a plurality of SVs in the GM data comprises generating from the GM data a list of the plurality of SVs, and / or a map comprising the plurality of SVs. In some embodiments, the determining a plurality of SVs in the GM data comprises identifying SV types in the plurality of SVs. In some embodiments, the SV types are selected from deletions, duplications, inversions, and / or translocations. In some embodiments, the determining a plurality of SVs in the GM data comprises removing common SVs from the plurality of SVs. Some such embodiments include removing from the plurality of SVs, such as from the list and / or from the map, an SV having a VAT > 001 in a plurality of reference genomes. In some embodiments, the plurality of references genomes is from subjects lacking a reported disease phenotype.

[0088] In some embodiments, the determining a plurality of SVs in the GM data comprises determining the number of SVs in one or more SV types in the plurality of SVs, , such as in the list and / or in the map. In some embodiments, the one or more SV types is selected from a deletion, a duplication, an inversion, and / or a translocation. In some embodiments, the one or more SV types is selected from an isolated translocation, a balanced translocation, an inverted duplication with a size in a range from 30 kb to 300 kb, an inversion, a deletion < 25 kb, and / or a duplication < 100 kb. In some embodiments, the one or more SV types is selected from an isolated translocation, a balanced translocation, an inverted duplication with a size in a range from 30 kb to 300 kb, and / or an inversion. In some embodiments, each end of an isolated translocation is located at a distance from another translocation in an assembled map of the genome mapping data of at least 1.5 Mb, 2.0 Mb, 2.5 Mb, or 3.0 Mb. In some embodiments, the distance is at least 2.0 Mb.

[0089] In some embodiments, the determining the presence or absence of an HRD comprises applying to an HRD classifier the number of SVs in each of the one of more SV types to provide a probability that a sample comprises an HRD. In some embodiments, the number of SVs in each of the one of more SV types comprises the number of: (i) isolated translocations, (ii) balancedtranslocations, and (iii) inverted duplications with a size in a range from 30 kb to 300 kb together with the number of inversions.

[0090] In some embodiments, the HRD classifier is obtained by performing a least absolute shrinkage and selection operator (LASSO) logistic regression on the number of SVs in each SV type in a plurality of different SV types obtained from GM data from a plurality of control samples comprising an HRD and a plurality of control samples lacking an HRD. In some embodiments, the SVs obtained from GM data from the plurality of control samples lack an SV having a VAF > 001 in a plurality of reference genomes. In some embodiments, the plurality of references genomes is from subjects lacking a reported disease phenotype. In some embodiments, the different SV types are selected from duplications, inverted duplications, inversions, translocations, and / or deletions. In some embodiments, the different SV types include isolated translocations, balanced translocations, inverted duplications with a size in a range from 30 kb to 300 kb, deletions < 25 kb, and / or duplications < 100 kb.

[0091] In some embodiments, the determining the presence or absence of an HRD comprises comparing the number of SVs in one or more SV types in the plurality of SVs with the number of SVs in the one or more SV types in a control obtained from GM data from a sample lacking an HRD. In some embodiments, an increase in the number of the SVs in the one or more SV types in the plurality of SVs compared to the number of SVs in the one or more SV types in the control is indicative of the sample having an HRD, wherein the one or more SV types is selected from an isolated translocation, a balanced translocation, and / or a duplication < 100 kb. In some embodiments, a decrease in the number of the SVs in the one or more SV types in the plurality of SVs compared to the number of SVs in the one or more SV types in the control is indicative of the sample having an HRD, wherein the one or more SV types is selected from an inverted duplication with a size in a range from 30 kb to 300 kb, an inversion, and / or a deletion < 25 kb.

[0092] Some embodiments also include determining the presence or absence of an (i) inactive BRCA1 gene or (ii) inactive BRCA2 or RAD51C gene in the sample. In some such embodiments, the sample comprises an HRD. In some embodiments, the determining the presence or absence of an (i) inactive BRCA1 gene or (ii) inactive BRCA2 or RAD51C gene in the sample comprises applying to a / A / / / A2-like classifier the number of SVs in each of the one of more SV types to provide a probability that a sample comprises an (i) inactive BRCA1 gene or (ii) inactive BRCA2 or RAD51C gene. In some embodiments, the one of more SV types comprises a deletion < 25 kb, and / or a duplication < 100 kb.

[0093] In some embodiments, the determining the presence or absence of an (i) inactive BRCA1 gene or (ii) inactive BRCA2 o RAD51C gene in the sample comprises comparing the number of SVs in one or more SV types in the plurality of SVs with the number of SVs in the one or more SV types in a control obtained from GM data from a sample lacking inactive BRCA1, BRCA2 and RAD51C genes. In some embodiments, an increase in the number of the SVs in the one or more SV types in the plurality of SVs compared to the number of SVs in the one or more SV types in the control is indicative of the sample having an HRD, wherein the one or more SV types comprise a duplication < 100 kb. In some embodiments, a decrease in the number of the SVs in the one or more SV types in the plurality of SVs compared to the number of SVs in the one or more SV types in the control is indicative of the sample having an HRD, wherein the one or more SV types comprise a deletion < 25 kb.

[0094] In some embodiments, the sample comprises cells, cell-free DNA, cell-free fetal DNA, amniotic fluid, a blood sample, a bone marrow sample, a biopsy sample, or a combination thereof. In some embodiments, the sample comprises a cancer. In some embodiments, the sample comprises an ovarian cancer or a breast cancer. In some embodiments, the sample comprises a TNBC or a HGOC. In some embodiments, the sample comprises a formalin-fixed paraffin-embedded (FFPE) sample.

[0095] In some embodiments, the GM data comprises a consensus map for the sample from the subject. In some embodiments, the GM data is obtained by labelling genomic DNA from the sample from the subject with a detectable tag comprising a fluorophore, a quantum dot, a dendrimer, a nanowire, a bead, a hapten, streptavidin, avidin, neutravidin, biotin, a stabilized reactive group a radiolabel, an electromagnetic label, an azide, dibenzocyclooctyne (DBCO), or an alkyne. In some embodiments, the GM data comprises OGM data. In some embodiments, the GM data comprises electronic genome mapping (EGM) data.

[0096] Some embodiments also include creating a file or a report and / or generating a user interface (UI) comprising a UI element representing or comprising (i) the SV types in the GM data from the sample; (ii) the number of SVs in each SV type in the GM data; (iii) a probability that the sample comprises an HRD; and / or (iv) a probability that an HRD sample comprises an (a) inactive BRCA1 gene or (b) inactive BRCA2 or RAD51C gene.

[0097] In some embodiments, the subject is administered or prescribed a drug, such as a PARP inhibitor. In some embodiments, the method comprises administering or prescribing (or generating a prescription of) a drug, such as a PARP inhibitor to the subject. Non-limiting examples of PARP inhibitors include Olaparib, Rucaparib, Niraparib, and Talazoparib.Certain methods for determining the BRCA 1 BRCA2-\\\<e status of a sample

[0098] Some embodiments provided herein include methods for determining the presence or absence of an (i) inactive BRCA1 gene or (ii) inactive BRCA2 or RAD51C gene in a sample comprising an HRD. Some such methods include computer-implemented methods, comprising, under control of a hardware processor: receiving GM data generated from the sample from a subject; determining a plurality of SVs in the GM data by comparing the GM data with a reference genome; and determining the presence or absence of an (i) inactive BRCA1 gene or (ii) inactive BRCA2 or RAD51C gene in the sample based on the plurality of SVs.

[0099] In some embodiments, the determining a plurality of SVs in the GM data comprises generating from the GM data a list of the plurality of SVs, and / or a map comprising the plurality of SVs. In some embodiments, the determining a plurality of SVs in the GM data comprises identifying SV types in the plurality of SVs. In some embodiments, the SV types are selected from deletions, duplications, inversions, and / or translocations. In some embodiments, the determining a plurality of SVs in the GM data comprises removing from the plurality of SVs, such as from the list and / or from the map, an SV having a VAF > 001 in a plurality of reference genomes. In some embodiments, the plurality of references genomes is from subjects lacking a reported disease phenotype; and / or

[0100] In some embodiments, the determining a plurality of SVs in the GM data comprises determining the number of SVs in one or more SV types in the plurality of SVs, such as in the list and / or in the map. \In some embodiments, the one or more SV types is selected from a deletion, a duplication; optionally, wherein the one or more SV types is selected from a deletion < 25 kb, and / or a duplication < 100 kb.

[0101] In some embodiments, the determining the presence or absence comprises applying to a BRCAHBRCA2-\ e classifier the number of SVs in each of the one of more SV types to provide a probability that a sample comprises an (i) inactive BRCA1 gene or (ii) inactive BRCA2 or RAD51C gene. In some embodiments, the one of more SV types comprises a deletion < 25 kb, and / or a duplication < 100 kb.

[0102] In some embodiments, the determining the presence or absence comprises comparing the number of SVs in one or more SV types in the plurality of SVs with the number of SVs in the one or more SV types in a control obtained from GM data from a sample lacking inactive BRCA1, BRCA2 and RAD51C genes. In some such embodiments, an increase in the number of the SVs in the one or more SV types in the plurality of SVs compared to the number of SVs in the one ormore SV types in the control is indicative of the sample having an HRD, wherein the one or more SV types comprise a duplication < 100 kb. In some such embodiments, a decrease in the number of the SVs in the one or more SV types in the plurality of SVs compared to the number of SVs in the one or more SV types in the control is indicative of the sample having an HRD, wherein the one or more SV types comprise a deletion < 25 kb.

[0103] In some embodiments, the sample comprises cells, cell-free DNA, cell-free fetal DNA, amniotic fluid, a blood sample, a bone marrow sample, a biopsy sample, or a combination thereof. In some embodiments, the sample comprises a cancer. In some embodiments, the sample comprises an ovarian cancer or a breast cancer. In some embodiments, the sample comprises a TNBC or an HGOC. In some embodiments, the sample comprises a FFPE sample.

[0104] In some embodiments, the GM data comprises a consensus map for the sample from the subject. In some embodiments, the GM data is obtained by labelling genomic DNA from the sample from the subject with a detectable tag comprising a fluorophore, a quantum dot, a dendrimer, a nanowire, a bead, a hapten, streptavidin, avidin, neutravidin, biotin, a stabilized reactive group a radiolabel, an electromagnetic label, an azide, DBCO, or an alkyne. In some embodiments, the GM data comprises OGM data. In some embodiments, the GM data comprises EGM data.

[0105] Some embodiments also include creating a file or a report and / or generating a UI comprising a UI element representing or comprising (i) the SV types in the GM data from the sample; (ii) the number of SVs in each SV type in the GM data; and / or (iii) a probability that an HRD sample comprises an (a) inactive BRCA1 gene or (b) inactive BRCA2 or RAD51C gene.

[0106] In some embodiments, the subject is administered or prescribed a drug, such as a PARP inhibitor. In some embodiments, the method comprises administering or prescribing (or generating a prescription of) a drug, such as a PARP inhibitor to the subject. Non-limiting examples of PARP inhibitors include Olaparib, Rucaparib, Niraparib, and Talazoparib.Certain systems for determining an HRD status of a sample

[0107] Some embodiments provided herein include systems for determining the presence or absence of an HRD in a sample. Some such embodiments include: non-transitory memory configured to store: executable instructions, GM data generated from a sample from a subject; a hardware processor in communication with the non-transitory memory, wherein the hardware processor is programmed by the executable instructions to perform: determining a plurality of SVs in the GM data by comparing the GM data with a reference genome, and determining the presence or absence of an HRD in the sample based on the plurality of SVs.

[0108] In some embodiments, the hardware processor is programmed by the executable instructions to perform: generating from the GM data a list of the plurality of SVs, and / or a map comprising the plurality of SVs . In some embodiments, the non-transitory memory configured to store: the list and / or the map. In some embodiments, the hardware processor is programmed by the executable instructions to perform: identifying SV types in the plurality of SVs. In some embodiments, the SV types are selected from deletions, duplications, inversions, and / or translocations. In some embodiments, the hardware processor is programmed by the executable instructions to perform: removing common SVs from the plurality of SVs, such as from the list and / or from the map, by removing from the plurality of SVs, such as from the list and / or the map, an SV having a VAF > 001 in a plurality of reference genomes. In some embodiments, the plurality of references genomes is from subjects lacking a reported disease phenotype.

[0109] In some embodiments, the hardware processor is programmed by the executable instructions to perform: determining the number of SVs in one or more SV types in the plurality of SVs, such as in the list and / or in the map. In some embodiments, the one or more SV types is selected from a deletion, a duplication, an inversion, and / or a translocation. In some embodiments, the one or more SV types is selected from an isolated translocation, a balanced translocation, an inverted duplication with a size in a range from 30 kb to 300 kb, an inversion, a deletion < 25 kb, and / or a duplication < 100 kb. In some embodiments, the one or more SV types is selected from an isolated translocation, a balanced translocation, an inverted duplication with a size in a range from 30 kb to 300 kb, and / or an inversion. In some embodiments, each end of an isolated translocation is located at a distance from another translocation in an assembled map of the genome mapping data of at least 1.5 Mb, 2.0 Mb, 2.5 Mb, or 3.0 Mb. In some embodiments, the distance is at least 2.0 Mb.

[0110] In some embodiments, the non-transitory memory configured to store an HRD classifier. In some embodiments, the hardware processor is programmed by the executable instructions to perform: applying to an HRD classifier the number of SVs in each of the one of more SV types to provide a probability that a sample comprises an HRD. In some embodiments, the number of SVs in each of the one of more SV types comprises the number of: (i) isolated translocations, (ii) balanced translocations, and (iii) inverted duplications with a size in a range from 30 kb to 300 kb together with the number of inversions.[OHl] In some embodiments, the HRD classifier is obtained by performing a LASSO logistic regression on the number of SVs in each SV type in a plurality of different SV types obtained from GM data from a plurality of control samples comprising an HRD and a plurality of control samples lacking an HRD. In some embodiments, the SVs obtained from GM data from the pluralityof control samples lack an SV having a VAF > 001 in a plurality of reference genomes. In some embodiments, the plurality of references genomes is from subjects lacking a reported disease phenotype. In some embodiments, the different SV types are selected from duplications, inverted duplications, inversions, translocations, and / or deletions. In some embodiments, the different SV types are selected from isolated translocations, balanced translocations, inverted duplications with a size in a range from 30 kb to 300 kb, deletions < 25 kb, and / or duplications < 100 kb.

[0112] In some embodiments, the hardware processor is programmed by the executable instructions to perform: comparing the number of SVs in one or more SV types in the plurality of SVs with the number of SVs in the one or more SV types in a control obtained from GM data from a sample lacking an HRD. In some embodiments, an increase in the number of the SVs in the one or more SV types in the plurality of SVs compared to the number of SVs in the one or more SV types in the control is indicative of the sample having an HRD, wherein the one or more SV types is selected from an isolated translocation, a balanced translocation, and / or a duplication < 100 kb. In some embodiments, a decrease in the number of the SVs in the one or more SV types in the plurality of SVs compared to the number of SVs in the one or more SV types in the control is indicative of the sample having an HRD, wherein the one or more SV types is selected from an inverted duplication with a size in a range from 30 kb to 300 kb, an inversion, and / or a deletion < 25 kb.

[0113] In some embodiments, the hardware processor is programmed by the executable instructions to perform: determining the presence or absence of an (i) inactive BRCA1 gene or (ii) inactive BRCA2 or RAD51C gene in the sample. In some such embodiments, the sample comprises an HRD. In some embodiments, the non-transitory memory configured to store: a BRCA1 / BRCA2-like classifier. In some embodiments, the hardware processor is programmed by the executable instructions to perform: applying to a BRCA1 IBRCA2- ke classifier the number of SVs in each of the one of more SV types to provide a probability that a sample comprises an (i) inactive BRCA1 gene or (ii) inactive BRCA2 or RAD51C gene. In some embodiments, the one of more SV types comprises a deletion < 25 kb, and / or a duplication < 100 kb.

[0114] In some embodiments, the hardware processor is programmed by the executable instructions to perform: comparing the number of SVs in one or more SV types in the plurality of SVs with the number of SVs in the one or more SV types in a control obtained from GM data from a sample lacking inactive BRCA1, BRCA2 and RAD51C genes. In some embodiments, an increase in the number of the SVs in the one or more SV types in the plurality of SVs compared to the number of SVs in the one or more SV types in the control is indicative of the sample having an HRD, wherein the one or more SV types comprise a duplication < 100 kb. In some embodiments, a decrease in thenumber of the SVs in the one or more SV types in the plurality of SVs compared to the number of SVs in the one or more SV types in the control is indicative of the sample having an HRD, wherein the one or more SV types comprise a deletion < 25 kb.

[0115] In some embodiments, the sample comprises cells, cell-free DNA, cell-free fetal DNA, amniotic fluid, a blood sample, a bone marrow sample, a biopsy sample, or a combination thereof. In some embodiments, the sample comprises a cancer. In some embodiments, the sample comprises an ovarian cancer or a breast cancer. In some embodiments, the sample comprises a TNBC or an HGOC. In some embodiments, the sample comprises an FFPE sample.

[0116] In some embodiments, the GM data comprises a consensus map for the sample from the subject. In some embodiments, the GM data is obtained by labelling genomic DNA from the sample from the subject with a detectable tag comprising a fluorophore, a quantum dot, a dendrimer, a nanowire, a bead, a hapten, streptavidin, avidin, neutravidin, biotin, a stabilized reactive group a radiolabel, an electromagnetic label, an azide, DBCO, or an alkyne. In some embodiments, the GM data comprises OGM data. In some embodiments, the GM data comprises EGM data.

[0117] In some embodiments, the hardware processor is programmed by the executable instructions to perform: creating a file or a report and / or generating a U1 comprising a U1 element representing or comprising (i) the SV types in the GM data from the sample; (ii) the number of SVs in each SV type in the GM data; (iii) a probability that the sample comprises an HRD; and / or (iv) a probability that an HRD sample comprises an (a) inactive BRCA1 gene or (b) inactive BRCA2 or RAD51C gene.

[0118] In some embodiments, the subject is administered or prescribed a drug, such as a PARP inhibitor. In some embodiments, the hardware processor is programmed by the executable instructions to perform: generating a prescription of a drug, such as a PARP inhibitor to the subject. Non-limiting examples of PARP inhibitors include Olaparib, Rucaparib, Niraparib, and Talazoparib.Certain systems for determining the BACA 7 / BRCA2-like status of a sample

[0119] Some embodiments provided herein include systems for determining the presence or absence of an (i) inactive BRCA1 gene or (ii) inactive BRCA2 or RAD51C gene in a sample comprising an HRD. Some such embodiments include: non-transitory memory configured to store: executable instructions, GM data generated from a sample from a subject; a hardware processor in communication with the non-transitory memory, wherein the hardware processor is programmed by the executable instructions to perform: determining a plurality of SVs in the GM data by comparingthe GM data with a reference genome, and determining the presence or absence of an (i) inactive BRCA1 gene or (ii) inactive BRCA2 or RAD51C gene in the sample based on the plurality of SVs.

[0120] In some embodiments, the hardware processor is programmed by the executable instructions to perform: generating from the GM data a list of the plurality of SVs, and / or a map comprising the plurality of SVs. In some embodiments, the non-transitory memory configured to store: the list and / or the map. In some embodiments, the hardware processor is programmed by the executable instructions to perform: identifying SV types in the plurality of SVs. In some embodiments, the SV types are selected from deletions, duplications, inversions, and / or translocations. In some embodiments, the hardware processor is programmed by the executable instructions to perform: removing common SVs from the plurality of SVs by removing from the plurality of SVs, such as from the list and / or from the map, an SV having a VAF > 001 in a plurality of reference genomes. In some embodiments, the plurality of references genomes is from subjects lacking a reported disease phenotype.

[0121] In some embodiments, the hardware processor is programmed by the executable instructions to perform: determining the number of SVs in one or more SV types in the plurality of SVs, such as in the list and / or in the map. In some embodiments, the one or more SV types is selected from a deletion, a duplication. In some embodiments, the one or more SV types is selected from a deletion < 25 kb, and / or a duplication < 100 kb.

[0122] In some embodiments, the non-transitory memory configured to store: a BRCAJ / BRCA2-[ike classifier. In some embodiments, the hardware processor is programmed by the executable instructions to perform: applying to a. BRC A 1 / BRC A 2 -kike classifier the number of SVs in each of the one of more SV types to provide a probability that a sample comprises an (i) inactive BRCA1 gene or (ii) inactive BRCA2 or RAD51C gene. In some embodiments, the one of more SV types comprises a deletion < 25 kb, and / or a duplication < 100 kb.

[0123] In some embodiments, the hardware processor is programmed by the executable instructions to perform: comparing the number of SVs in one or more SV types in the plurality of SVs with the number of SVs in the one or more SV types in a control obtained from GM data from a sample lacking inactive BRCA1, BRCA2 and RAD51C genes. In some embodiments, an increase in the number of the SVs in the one or more SV types in the plurality of SVs compared to the number of SVs in the one or more SV types in the control is indicative of the sample having an HRD, wherein the one or more SV types comprise a duplication < 100 kb. In some embodiments, a decrease in the number of the SVs in the one or more SV types in the plurality of SVs compared to the number ofSVs in the one or more SV types in the control is indicative of the sample having an HRD, wherein the one or more SV types comprise a deletion < 25 kb.

[0124] In some embodiments, the sample comprises cells, cell-free DNA, cell-free fetal DNA, amniotic fluid, a blood sample, a bone marrow sample, a biopsy sample, or a combination thereof. In some embodiments, the sample comprises a cancer. In some embodiments, the sample comprises an ovarian cancer or a breast cancer. In some embodiments, the sample comprises a TNBC or an HGOC. In some embodiments, the sample comprises an FFPE sample.

[0125] In some embodiments, the GM data comprises a consensus map for the sample from the subject. In some embodiments, the GM data is obtained by labelling genomic DNA from the sample from the subject with a detectable tag comprising a fluorophore, a quantum dot, a dendrimer, a nanowire, ahead, a hapten, streptavidin, avidin, neutravidin, biotin, a stabilized reactive group a radiolabel, an electromagnetic label, an azide, DBCO, or an alkyne. In some embodiments, the GM data comprises OGM data. In some embodiments, the GM data comprises EGM data.

[0126] In some embodiments, the hardware processor is programmed by the executable instructions to perform: creating a file or a report and / or generating a UI comprising a UI element representing or comprising (i) the SV types in the GM data from the sample; (ii) the number of SVs in each SV type in the GM data; and / or (iii) a probability that an HRD sample comprises an (a) inactive BRCA1 gene or (b) inactive BRCA2 or RAD51C gene.

[0127] In some embodiments, the subject is administered or prescribed a drug, such as a PARP inhibitor. In some embodiments, the hardware processor is programmed by the executable instructions to perform: generating a prescription of a drug, such as a PARP inhibitor to the subject. Non-limiting examples of PARP inhibitors include Olaparib, Rucaparib, Niraparib, and Talazoparib.Example methods

[0128] FIG. 2 is a flow diagram showing an exemplary method 200 of determining the presence or absence of a homologous recombination deficiency (HRD) in a sample from GM data. The method 200 (or one or more actions of steps thereof) may be embodied in a set of executable program instructions stored on a computer-readable medium, such as one or more disk drives, of a computing system. For example, the computing system 600 shown in FIG. 6 and described in greater detail below can execute a set of executable program instructions to implement the method 200. When the method 200 is initiated, the executable program instructions can be loaded into memory, such as RAM, and executed by one or more processors of the computing system 600. Although the method 200 is described with respect to the computing system 600 shown in FIG. 6, the descriptionis illustrative only and is not intended to be limiting. In some embodiments, the method 200 or portions thereof may be performed serially or in parallel by multiple computing systems. After the method 200 begins at block 204, the method 200 proceeds to block 208, where a computing system receives GM data generated from a sample from a subject. The method 200 proceeds from block 208 to block 212, where the computing system determines a plurality of SVs in the GM data by comparing the GM data with a reference genome. After determining a plurality of SVs in the GM data at block 212, the method 200 proceeds to block 216, where the computing system determines the presence or absence of an HRD in the sample based on the plurality of SVs. The method 200 ends at block 220.

[0129] FIG. 3 is a flow diagram showing an exemplary method 300 of determining the presence or absence of a homologous recombination deficiency (HRD) in a sample from GM data. The method 300 (or one or more actions of steps thereof) may be embodied in a set of executable program instructions stored on a computer-readable medium, such as one or more disk drives, of a computing system. For example, the computing system 600 shown in FIG. 6 and described in greater detail below can execute a set of executable program instructions to implement the method 300. When the method 300 is initiated, the executable program instructions can be loaded into memory, such as RAM, and executed by one or more processors of the computing system 600. Although the method 300 is described with respect to the computing system 600 shown in FIG. 6, the description is illustrative only and is not intended to be limiting. In some embodiments, the method 300 or portions thereof may be performed serially or in parallel by multiple computing systems. After the method 300 begins at block 304, the method 300 proceeds to block 308, where a computing system receives GM data generated from a sample from a subject. The method 300 proceeds from block 308 to block 312, where the computing system determines a plurality of SVs in the GM data by comparing the GM data with a reference genome. After the determining at block 312, the method 300 proceeds to block 316, where the computing system identifies SV types in the plurality of SVs. The method 300 proceeds from block 316 to an optional step shown in block 320, where the computing system optionally removes common SVs from the plurality of SVs. After the identifying of block 316 or the removing of block 320, the method 300 proceeds to block 324, where the computing system counts the number of SVs in one or more SV types in the plurality of SVs. The method 300 proceeds from block 324 to block 330, where the computing system applies to an HRD classifier, the number of SVs in each of the one of more SV types to provide a probability that a sample comprises an HRD. In some embodiments, the computing system applies to an BRCAl / BRCA2-like classifier, the number of SVs in each of the one of more SV types to provide a probability that a sample comprises an (i) inactive BRCA1 gene or (ii) inactive BRCA2 or RAD51C gene. After the application of the HRDclassifier to the number of SVs at block 330, the method 300 proceeds to block 334, where the computing system determines the presence or absence of an HRD in a sample based on the probability. The method 300 ends at block 338.

[0130] FIG. 4 is a flow diagram showing an exemplary method 400 of determining the BRCAl / BRCA2-like status of a sample from GM data, such as a sample having an HRD. The method 400 (or one or more actions of steps thereof) may be embodied in a set of executable program instructions stored on a computer-readable medium, such as one or more disk drives, of a computing system. For example, the computing system 600 shown in FIG. 6 and described in greater detail below can execute a set of executable program instructions to implement the method 400. When the method 400 is initiated, the executable program instructions can be loaded into memory, such as RAM, and executed by one or more processors of the computing system 600. Although the method 400 is described with respect to the computing system 600 shown in FIG. 6, the description is illustrative only and is not intended to be limiting. In some embodiments, the method 400 or portions thereof may be performed serially or in parallel by multiple computing systems. After the method 400 begins at block 404, the method 400 proceeds to block 408, where a computing system receives GM data generated from a sample from a subject. The method 400 proceeds from block 408 to block 412, where the computing system determines a plurality of SVs in the GM data by comparing the GM data with a reference genome. After determining a plurality of SVs in the GM data at block 412, the method 400 proceeds to block 416, where the computing system provides a determination that a sample comprises an (i) inactive BRCA1 gene or (ii) inactive BRCA2 or RAD51C gene. The method 400 ends at block 420.

[0131] FIG. 5 is a flow diagram showing an exemplary method 500 of determining the BRCAl / BRCA2-like status of a sample from GM data, such as a sample having an HRD. The method 500 (or one or more actions of steps thereof) may be embodied in a set of executable program instructions stored on a computer-readable medium, such as one or more disk drives, of a computing system. For example, the computing system 600 shown in FIG. 6 and described in greater detail below can execute a set of executable program instructions to implement the method 500. When the method 500 is initiated, the executable program instructions can be loaded into memory, such as RAM, and executed by one or more processors of the computing system 600. Although the method 500 is described with respect to the computing system 600 shown in FIG. 6, the description is illustrative only and is not intended to be limiting. In some embodiments, the method 500 or portions thereof may be performed serially or in parallel by multiple computing systems. After the method 500 begins at block 504, the method 500 proceeds to block 508, where a computing system receivesGM data generated from a sample from a subject. The method 500 proceeds from block 508 to block 512, where the computing system determines a plurality of SVs in the GM data by comparing the GM data with a reference genome. After the determining at block 512, the method 500 proceeds to block 516, where the computing system identifies SV types in the plurality of SVs. The method 500 proceeds from block 516 to an optional step shown in block 520, where the computing system optionally removes common SVs from the plurality of SVs. After the identifying of block 516 or the removing of block 520, the method 500 proceeds to block 524, where the computing system counts the number of SVs in one or more SV types in the plurality of SVs.

[0132] The method 500 proceeds from block 524 to block 530, where the computing system applies to a BRCAl / BRCA2-like classifier, the number of SVs in each of the one of more SV types to provide a probability that a sample comprises an (i) inactive BRCA1 gene or (ii) inactive BRCA2 or RAD51C gene. After the application of the HRD classifier to the number of SVs at block 530, the method 500 proceeds to block 534, where the computing system provides a determination that a sample comprises an (i) inactive BRCA1 gene or (ii) inactive BRCA2 or RAD51C gene based on the probability. The method 500 ends at block 538.

[0133] With respect to any of the methods 200, 300, 400, and 500, the subject can be administered or prescribed a drug, such as a PARP inhibitor. In some embodiments, the method comprises administering or prescribing a drug, such as a PARP inhibitor to the subject. Non-limiting examples of PARP inhibitors include Olaparib, Rucaparib, Niraparib, and Talazoparib.Execution environment

[0134] FIG. 6 depicts a general architecture of an example computing device 600 configured for determining the presence or absence of an HRD and / or the BRCALBRCA2-like status in a sample from GM data / configured to execute the processes and implement the features described herein / that can be used in some embodiments to execute the processes and implement the features described herein. The general architecture of the computing device 600 depicted in FIG. 6 includes an arrangement of computer hardware and software components. The computing device 600 may include many more (or fewer) elements than those shown in FIG. 6. It is not necessary, however, that all of these generally conventional elements be shown in order to provide an enabling disclosure. As illustrated, the computing device 600 includes a processing unit 610, a network interface 620, a computer readable medium drive 630, an input / output device interface 640, a display 650, and an input device 660, all of which may communicate with one another by way of a communication bus. The network interface 620 may provide connectivity to one or more networks or computing systems.The processing unit 610 may thus receive information and instructions from other computing systems or services via a network. The processing unit 610 may also communicate to and from memory 670 and further provide output information for an optional display 650 via the input / output device interface 640. The input / output device interface 640 may also accept input from the optional input device 660, such as a keyboard, mouse, digital pen, microphone, touch screen, gesture recognition system, voice recognition system, gamepad, accelerometer, gyroscope, or other input device.

[0135] The memory 670 may contain computer program instructions (grouped as modules or components in some embodiments) that the processing unit 610 executes in order to implement one or more embodiments. The memory 670 generally includes RAM, ROM and / or other persistent, auxiliary or non-transitory computer-readable media. The memory 670 may store an operating system 672 that provides computer program instructions for use by the processing unit 610 in the general administration and operation of the computing device 600. The memory 670 may further include computer program instructions and other information for implementing aspects of the present disclosure.

[0136] For example, in one embodiment, the memory 670 includes an HRD determination module 674 for determining the presence or absence of a homologous recombination deficiency (HRD) in a sample from GM data, such as the method 200 or 300 described with reference to FIG. 2 and FIG. 3, respectively. The memory 470 may additionally or alternatively include a BRCA1 / BRCA2-[ike determination module 476 for determining the BRCA1 BRCA2 -like status in a sample from GM data, such as the method 400 or 500 described with reference to FIG. 4 and FIG. 5, respectively. In addition, memory 470 may include or communicate with the data store 490 and / or one or more other data stores that store input, intermediate results, and results of any method described herein. For example, memory 470 may include or communicate with the data store 490 and / or one or more other data stores that store one or more of: GM data generated from a sample from a subject; GM data generated from a reference genome; a list of SV variants which each have a VAF > 001 in a plurality of reference genomes, such as a plurality of references genomes from subjects lacking a reported disease phenotype; an HRD classifier; a BRCAl / BRCA2Aike classifier; GM data from a plurality of control samples comprising an HRD and a plurality of control samples lacking an HRD; and / or GM data from a plurality of control samples lacking inactive BRCAL BRCA2 and RAD51C genes.EX AMPLE SExample 1 — Identification and validation of classifiers for HRD and BACA / / BACA2-like status from PGM data

[0137] Classifiers for HRD and BRCAl / BRCA2-like status were identified from PGM data in a training series of TNBC tumors and HGPC tumors. The classifiers were tested with PGM data from a validation series of tumor samples.Training series and quality control

[0138] Tumor samples for TNBC and HGPC were obtained for which HRD status had been assessed by shallowWGSv2, and sequencing of HRR genes to determine HRR mutation status. The training set included 52 cases: 34 TNBC and 18 HGPC. Eight HGPC cases had been annotated by the Myriad MyChoice® test (Myriad Genetics, Salt Lake City, UT) which were consistent with results from shallowHRDv2 or HRR mutation status. Six cases were excluded because of insufficient material for PGM (5 TNBC, 1 HGPC). Forty-three cases out of 46 passed post-PGM quality filters including 27 TNBC and 16 HGPC (FIG. 7). PGM was performed from frozen blocks and shallowHRDv2 from formalin-fixed paraffin-embedded (FFPE) slides corresponding to different parts of the same tumor.Tumor content and effective coverage

[0139] Certain characteristics of samples used in the training series and validation series are listed in TABLE 1. Effective coverage in the training series ranged from 160 X to 1500 X (median ± interquartile range [IQR] = 676 ± 388). All SVs with a variant allele frequency (VAF) > 0.01 were extracted from the RVA pipeline. The total number of SVs detected per tumor ranged from 59 to 574 (262 ± 177). The median number of molecules per detected SV ranged from 13 to 377 (79 ± 86). An outlier case had 31 detected SVs and a flat PGM profile and was discarded due to undetectable tumor content. Another outlier case had less than 10 median molecules per detected SV with an effective coverage (207 X) and was also discarded. Six cases were kept in the training set despite having less than 20 median number of molecules per detected SV, corresponding to low tumor content and / or low effective coverage. In total, 78% of initially selected cases were analyzable, including 26 TNBC (14 HRD cases including: 5 BRCAlmut, 2 BRCA2mut and 1 RAD51Cmut, and 12 nonHRD cases) and 15 HGPC (11 HRD cases including: 2 BRCAlmut and 5 BRCA2mut, and 4 nonHRD cases).

[0140] TABLE 1 lists certain characteristics for samples used in the training and validation series including: sample ID; tissue type (triple negative breast cancer (TNBC) or high gradeovarian cancer (HGOC)); sample classification under TNM classification and FIGO staging system; age of subject at diagnosis; Shallowv2 HRD status (H: HRD, nH = non-HRD); Myriad HRD status;i TsseuHRR mutant gene; T / M; Series (TR= training, V = validation); effective coverage of OGM; ‘N50 > 20 kb’ (i.e. median size of molecules > 20 kb); average map rate; total number of SVs; median number of molecules per SV / GO TNM FI; and reason for exclusion (QC issue, IM = insufficient material, UT =ifilt casscaoniundetectable tumor content).Aii dtge aagnoss TABLE 1Shll2aowvi MdrayHRRtt man geneuQ1610 TNBC T1N1M1 39 nH T / TM TR 429.18 118 82 185 521611 TNBC T1N0M1 37 nH T V 727.68 157 88 52 101.5Sriees1612 TNBC T4dNlMl 62 nH T V 1168.53 175 88 114 113.5 1613 TNBC T1N0M1 74 H RAD51C T TR 821.57 206 89 295 92ffi Etece coeragevv1614 TNBC TxNIMl 56 H BRCA2 M V 614.95 133 75 157 781839 TNBC T1N0M0 45 H BRCA1 T TR 984 1 N5020 kb >87 86 201 1021615 TNBCPT2N3aMl 56 nH M TR 320 97 56 QC 1840 TNBC pTisNOMl 24 H BRCA2 T TR 652.1 AGVt map rae 132 168.5 1841 TNBC ypTlcNlaMl 38 nH T TR 508.89 195 87 1f S TlbVt7oa nmer ou3 471800 TNBC T2N+M1 38 nH FANCD2 T TR 487.82 191 88 177 3if Mdb7ean nmer ou1616 TNBC T2N0M1 40 H BRCA1 T V 802.15 141 71 220 38.5 SllV moeces peru 1796 TNBC T3N+M0 56 H T TR IM i Elcsonxu 1617 TNBC yT4bNxMl 39 H T V 1061.08 174 86 272 1801801 TNBC ypTlcNOMO 65 H BRCA1 T TR 532.18 318 75 333 361802 TNBC ypTlcN2aM0 38 H BRCA1 T TR 411.31 170 86 272 351803 TNBC ypTisNIMO 54 nH T TR 162.5 118 58 59 231804 TNBC T2N2M1 59 H BRCA1 T TR 290.29 122 83 215 291805 TNBC ypT3N2aMl 68 nH T TR 406.86 192 89 152 19.51806 TNBC ypT2N0M0 47 H T TR 788.29 181 84 574 1021807 TNBC T2N0M0 49 H T TR 813.48 176 84 273 851618 TNBC ypT2N2aMl 49 nH M TR 740.59 142 84 170 1311808 TNBC ypT2NlMl 48 nH T TR 740.04 125 85 199 491797 TNBC T2N0 31 H BRCA1 T TR IM 1809 TNBC T2N0M0 41 H T TR 318.37 156 79 311 161826 TNBCPT2NxMl 78 nH T V 1056 202 92 577 961619 TNBC T2N1M0 40 H M TR 241.16 132 64 348 131810 TNBC ypTlbNOMO 25 H BRCA1 T TR 275.5 182 63 140 211811 TNBC ypTIbNIMO 37 H T TR 359.27 159 77 140 16 1588 HGOC FIGO IV 72 nH T V 687.35 193 90 16 96 UT 1812 TNBC YpTONIMl 53 nH T TR 799.65 218 84 108 28 1792 TNBC T3N1M1 42 ND T TR IM 1813 TNBC T2N0M0 57 nH T TR 567.92 219 79 295 44 1620 TNBC pTINIMl 51 nH T TR 1499.61 238 93 238 341.5 1621 TNBC ypT2N0Ml 65 H T TR 675.56 211 90 349 22 1622 TNBC T2N0M1 45 nH T TR 521.46 159 84 97 102 1793 TNBC T4dNlMl 26 nH T TR IM 1623 TNBC ypTlcNOMl 60 nH T V 429.18 179 85 19 109 UT 1624 TNBC pT2N2aMl 62 nH T TR 759.25 223 91 142 82 1798 TNBC pT3N2aMl 43 nH M V IM 1827 HGOC FIGO me 75 nH nH T V 1382.81 244 90 307 147 1814 HGOC FIGO me 66 nH nH T TR 1192.88 252 91 288 78 1828 HGOC FIGO me 70 nH T V 362.41 155 76 346 38 1815 HGOC FIGO IIIB 66 H H T TR 514.1 125 89 324 48.5 1829 TNBC T4dN3Ml 35 nH T V 292.23 78 74 QC 1830 HGOC FIGO IIIC 90 nH nH T V 1477.3 287 91 244 68.5 1589 HGOC FIGO IIIC 56 nH nH T TR 817.14 211 90 194 94 1590 HGOC FIGO IIIC 53 H H T TR 778.56 275 89 268 163 1591 HGOC FIGO IIIC 70 nH nH T TR 304.82 227 79 377 17 1592 HGOC FIGO IV 67 H BRCA2 T V 1109.76 193 86 347 105 1593 HGOC1 FIGO IIIC 72 nH T V 1295.64 210 92 95 25 1594 HGOC FIGO IVA 80 nH nH T V 1462.36 260 90 288 192 1816 HGOC FIGO lb 69 H BRCA2 T TR 1047.51 265 91 399 155 1625 TNBC ypT2N3aMl 48 nH T V 808.32 209 90 279 65 1595 HGOC FIGO IIIC 73 nH T TR 1101.14 241 89 31 260 UT 1596 HGOC FIGO IIIC 49 H H BRCA1 T TR 638.32 143 80 383 84 1597 HGOC2 FIGO IV 66 H H T V 1080.25 256 93 307 149 1598 HGOC FIGO IIIC 81 nH nH T V 1135.38 251 93 153 74 1599 HGOC FIGO IVB 86 nH T V 591.58 129 50 123 50 1600 HGOC FIGO IVB 74 nH nH T V 992.27 198 90 72 34.5 1601 HGOC FIGO IIIC 76 H H T TR 1457.98 245 90 436 377 1602 HGOC FIGO IIIA 73 H H T V 1006.66 208 86 707 100 1831 TNBC ypT4bNlaMl 45 nH T V 845.26 200 90 309 110 1603 HGOC FIGO IV 73 nH nH T V 802.97 217 72 207 88 1604 HGOC FIGO IIIC 65 nH nH T TR 1225.04 208 86 92 131.5 1605 HGOC FIGO IV 72 H H M V 538.46 238 80 44 147.5 UT 1606 HGOC FIGO IIIC 81 nH nH T V 328.38 261 86 101 11 1817 HGOC FIGO lb 69 H BRCA2 T TR 803.02 168 85 343 80 1607 HGOC FIGO IIIC 74 H BRCA2 T V 952.28 258 92 231 1751608 HGOC FIGO IV 57 H BRCA1 T V 568.93 117 74 287 371818 TNBC T4dN+Ml 69 nH T TR 27.4 47 43 QC 1794 HGOC FIGO me 80 nH T TR IM 1609 HGOC FIGO IB 57 H T TR 1291.65 235 92 255 2601832 HGOC FIGO me 75 H H T V 714.34 205 89 154 79.51820 TNBC T4bN+Ml 51 ND T TR 206.71 126 72 74 9 QC 1834 TNBC T3N+M1 35 nH T V 455.71 181 87 174 49.51835 TNBC ypT2N2aMl 37 nH T V 1029.94 194 91 136 1931821 HGOC FIGO IVB 71 H BRCA2 T TR 575.92 134 89 333 201833 TNBC pT2N0Ml 36 H T V 1236.47 219 85 403 2521822 TNBC TxN3Ml 52 H BRCA2 M TR 453.61 110 77 135 431819 HGOC FIGO IV 50 H BRCA2 T TR 15.12 72 6 QC 1836 TNBC T3NxMl 36 H T V 798.45 185 92 172 261626 TNBC ypT3NlMl 33 nH meRAD51C T V 1006.52 242 94 263 1831823 HGOC FIGO IIIA2 42 H BRCA1 T TR 1098.08 164 81 333 2351824 HGOC FIGO me 82 H BRCA2 T TR 835.77 157 79 296 961837 TNBC T1N+M1 49 nH M V 831.65 153 87 228 181825 HGOC FIGO me 54 H BRCA2 T TR 861.14 191 93 378 1211799 TNBC T4dN3Ml 50 nH T V IM 1838 TNBC ypTONOMl 68 nH T V 947.49 174 93 87 651795 TNBC ypTlbNlaMl 55 ND T TR IMStructural variant annotation

[0141] A rare variant analysis (RVA) pipeline classified 10 categories of SV including: deletion, insertion, duplication, duplication-split, inverted-duplication, inversion, inversion-paired, inversion-partial, inter-chromosomal translocation and intra-chromosomal translocation. Certain SVs are visualized in Circos plots depicted in FIG. 8.

[0142] RVA called head-to-tail tandem duplications as insertion, duplication or duplication-split depending on their sizes. Detection of duplications smaller than 30 kb was limited by the low number of labels that prevented robust mapping. Detection of large duplications was limited when exceeding the average size of DNA fibers, which was around 180 kb. Most insertions were < 30 kb; duplications were in a range of about 30-300 kb, and duplication-split were mostly > 200 kb (FIG. 9A). Consistently, 9 / 10 randomly chosen > 30 kb insertions were manually confirmed as duplications; and 10 / 10 randomly chosen duplication-split with sufficient coverage (visible CNAs) were manually confirmed as duplications in the genomic viewer (FIG. 9B and FIG. 9). Therefore, insertions, duplications and duplication-split were merged in a single “DUP” SV type.

[0143] Duplication inverted were called for tandem inverted duplication (tail-to-tail or head-to-head duplication) with size range 30 kb - 300 kb and denoted DUP_INV. Small DUP_INVcould also be annotated as insertions and the large ones as two different DUP INV. Because of rarity of such events and their marginal effects on SV classification, further adjustments were not made. Simple inverted DNA fragments were reported by RVA as inversion, inversion-paired or inversionpartial denoted here by INV. Actual inversions were reported as two coupled events, thus a couple of inversion-paired was counted as one inversion, and inversion-partial were discarded.

[0144] Inter-chromosomal and intra-chromosomal translocations (representing all intra-chromosomal rearrangements with the distance between start and end genomic position more than 5 Mb) were counted together as TRA, ending up with five SV types: DUP, DUP_INV, INV, TRA and deletions (DEL). TABLE 2 lists certain characteristics related to SVs of the training set including SV categories, types and subtypes, and their abundance (median value; interquartile range).

[0145] TABLE 2 lists certain OGM SV categories, SV types and subtypes and their relative abundance in certain groups of samples.TABLE 24)0 90ftj H05 5.OG SMVt caegory Median ± IQR (P) Median ± IQR (P)DEL<25kb 22 ±20 16 ±9.5 33 ±49 1.05 e-03 19 ±8 73.5 ±39 1.24 e-03 Deletion. DELDEL>25kb 22 ±17 21.5 ±16.5 24 ±17 9.36 e-01 21 ±12 37 ±31.75 1.05 e-01 DUP<100kb 51 ±82 17 ±23 75 ±90 4.24 e-06 136 ±74 47.5 ±25.5 1.45 e-03 Insertion. DUP>100kb 20 ±20 18.5 ±13.75 27 ±20 2.04 e-01 27 ±32 20.5 ±15.75 3.84 e-01 Duplication. DUP SV tpeyDuplication-split. DUP<25kb 30 ±57 9.5 ±14.5 60 ±94 7.08 e-06 114 ±62 24.5 ±10.75 1.24 e-03DUP>25kb 33 ±40 24.5 ±12.75 47 ±41 6.39 e-03 55 ±35.5 42.5 ±19.25 2.81 e-01 Duplication inverted.DUP INVInversion. DUP INVplus 15 ±12 22.5 ±8.5 12 ±8 3.92 e-04 12 ±7.5 15 ±7 1.64 e-01INVInversion paired. SVbt speuyTranslocation all TRA 71 ±61 61.5 ±52.5 81 ±51 9.48 e-02 55 ±33 105 ±40.5 7.21 e-02 Interchromosomal.TRA isolated TRA 21 ±18 7 ±8.25 27 ±7 9.49 e-08 21 ±5.5 29.5 ±6 4.89 e-02 Translocation3 ±10 0 ±0.25 9 ±5 1.37 e-06 8 ±8.5 9.5 ±5.25 2.21 e-01intrachromosomal . balanced- TRAA ()ll51n=HRD non(16)n=() HRD25n=HRD nons vHRDCA BR17 ()n=CA BR1s v CAlik BR2e-Selection of SVs significantly associated with HRD status

[0146] In order to build an HRD OGM classifier, HRD (24 cases) andnonHRD (17 cases) OGM profiles from the 41 qualified cases were compared. First, the size of the four SV types (DEL, DUP, DUP INV and INV) were examined to define the most discriminant intervals between HRD and nonHRD cases, and between BRCA1 and BRCA2 HRD tumors. Of note, cases inactivated for RAD51C were considered within the group of BRCA2 HRD (termed B7?G42-like) since they were believed to share genomic characteristics.

[0147] The distribution of sizes of DUP with regard to HRD and mutation status was consistent with previous studies based on WGS (FIG. 10A). As depicted in FIG 6A, HRD BRCAlmut (n=7) and HRD BRCAwt (n=10) cases showed a clear single peak around 10 kb and drop around 25 kb. BRCA2-Yike HRD tumors (n=8) had a density peak around 50 kb and drop around 100 kb, whereas nonHRD tumors (n=16) had almost flat density profiles across all sizes. Thus, the SV subtypes DUP<100kb and DUP>100kb were considered to characterize the HRD status, and DUP<25kb and DUP>25kb SV subtypes to discriminate BRCA1 and BRCA 2-like HRD cases.

[0148] As depicted in FIG. 10B, the distribution of sizes of DEL showed a single peak for all HRD tumors around 10 kb and drop around 25 kb, whereas nonHRD tumors had almost flat density profiles. Therefore, the DEL<25kb and DEL>25kb SV subtypes were considered to characterize the HRD status.

[0149] INV and DUP_INV were less numerous and displayed no substantial difference in size distribution depending on the HRD status (FIG. 10C, FIG. 10D). As depicted in FIG. 10E, INV and DUP INV had correlated counts (R2= 0.28; P = 4 x 10'4) and DUP INV outnumbered INV, therefore INV and DUP_INV were merged into a single SV subtype ‘DUP_INVplus’ (FIG. 10E).

[0150] Second, HRD and nonHRD tumors were compared using all selected features (TABLE 2; TABLE 3; FIG. 10F, FIG. 11 A,). Three SV subtypes displayed significant differences. DUP<100kb and DEL<25kb were more numerous in HRD than in nonHRD tumors (DUP<100kb: 75 ±90 versus (vs) 17±23, respectively; P = 4.24 x 10'6; DEL<25kb: 33±49 vs 16±9.5, respectively; P = 1.05 x 10'3). Conversely, DUP INVplus were more numerous in nonHRD than in HRD cases (22.5±8.5 vs 12±8, respectively; P = 3.92 x 10’4). TABLE 2 lists the abundance (median; interquartile range) of certain SV subtypes in the training series; and a probability using classifiers provided herein for a sample being HRD and BRCA1 inactivated.

[0151] TABLE 3 lists certain characteristics of samples used in the training series.TABLE 3i TsseuQ HDRtt sasuHRRtt manugene1610 TNBC nH 9 85 5 89 16 16 20 42 7 0 0.0005852421613 TNBC H RAD51C 4 DUP100kb<5 11 25 31 116 42 13 68 32 8 0.999995736 6.63066E-06 1839 TNBC H BRCA1 75 27 67 35 21 12 12 54 21 0 0.973374513 0.999584038DUP100kb>1840 TNBC H BRCA2 8 4 7 5 22 22 5 71 17 2 0.918796029 0.069048373 1841 TNBC nH 36 57 2 DUP25kb<4 69 13 13 21 34 7 0 0.0004906441800 TNBC nH FANCD2 18 11 12 1 DUP25kb7> 20 21 17 90 3 0 6.34433E-051801 TNBC H BRCA1 145 35 125 55 27 29 16 81 30 10 0.99997969 0.999999954 1802 TNBC H BRCA1 154 15 127 42 1 DEL25kb<9 17 12 55 19 10 0.980980727 0.999999989 1803 TNBC nH 5 6 3 8 11 9 DEL25kb> 4 24 1 0 0.0001587691804 TNBC H BRCA1 64 63 31 96 17 21 7 DUPINVlpsu 43 20 1 0.981385052 0.896571733 1805 TNBC nH 10 16 8 18 7 32 20 68 6 0 0.000294246A TR1806 TNBC H 273 41 206 108 39 34 15 173 39 13 0.9999999791807 TNBC H 55 19 42 32 55 32 19 95 3A IldTRtsoae0 10 0.9999655241618 TNBC nH 32 13 21 24 18 21 24 64 12 3A BldTRaance 0.0148762691808 TNBC nH 10 21 6 25 8 24 33 105 2 1 2.2522E-061809 TNBC H 178 57 150 85 16 13 9 40 24 2 0.998551533() PHRD1619 TNBC H 199 27 165 61 22 17 7 76 24 8 0.9996369481810 TNBC H BRCA1 76 4 61 19 13 11 1 35 16 2 0.920106873 0.999596246 1811 TNBC H 62 4 55 11 18 10 7 39 21 1 0.9905561241812 TNBC nH 16 32 10 38 12 17 20 13 1 0 9.4423 IE-06 (CA) PBR1 1813 TNBC nH 11 10 7 14 25 58 28 165 7 1 0.0001695661620 TNBC nH 58 8 48 18 41 33 28 70 10 3 0.0018803791621 TNBC H 127 58 69 116 16 20 9 119 31 13 0.9999982271622 TNBC nH 10 20 6 24 10 15 24 22 3 0 1.8456E-051624 TNBC nH 12 26 9 29 16 24 19 48 12 0 0.0213108031814 HGOC nH 42 18 30 30 29 95 55 59 12 0 3.80185E-051815 HGOC H 51 12 40 23 87 33 29 115 25 13 0.9962703341589 HGOC nH 18 19 10 27 29 22 29 82 8 0 0.0002380961590 HGOC H 100 33 60 73 33 28 11 63 24 6 0.9989636361591 HGOC nH 18 24 9 33 19 59 32 226 11 0 0.0011035711816 HGOC H BRCA2 60 34 50 44 98 84 15 109 35 11 0.99999954 0.005081269 1596 HGOC H BRCA1 188 19 144 63 34 44 4 94 27 8 0.999972834 0.999999996 1601 HGOC H 267 43 223 87 29 10 4 83 28 11 0.9999918491604 HGOC nH 36 4 22 18 15 15 9 13 13 0 0.2017971141817 HGOC H BRCA2 59 32 24 67 76 58 15 106 31 11 0.999992795 0.0010919471609 HGOC H 117 16 100 33 33 16 3 71 29 9 0.999995156 1821 HGOC H BRCA2 34 13 27 20 71 62 17 136 28 14 0.999951805 0.003517434 1822 TNBC H BRCA2 32 20 11 41 36 6 1 40 26 6 0.99995509 0.022656309 1823 HGOC H BRCA1 136 63 114 85 15 24 14 82 23 11 0.998519509 0.999999939 1824 HGOC H BRCA2 50 21 22 49 68 32 21 104 27 12 0.999726162 0.0022260461825 HGOC H BRCA2 59 27 39 47 102 29 46 116 40 6 0.999991734 0.000459643

[0152] The 7 BRCA1 were compared with the 8 BRCA2-V\ke HRD tumors (FIG. 1 IB, FIG.10G). A significant difference was observed for the DUP<25kb, more numerous in the BRCA1 cases (114±62 vs24.5±10.75;P = 1.24x 10'3). Conversely, DEL<25kb were more abundant in the BRCA2-like than in BRCA1 cases (73.5±39 vs 19±8, respectively; P = 1.24 x 10‘3). No significant difference was found for DUP INVplus.

[0153] Third, the translocations (TRA) were examined. There was no substantial difference in the number of TRA between HRD and nonHRD groups (TABLE 2; FIG. 11 A). However, in nonHRD tumors, TRA were clustered at discrete genomic locations, while in HRD tumors, TRA more uniformly distributed across the genome (FIG. 8). Following a strategy used to define large-scale state transitions (LST) and large-scale genomic alterations (LGA), an isolated TRA SV subtype was identified, which was defined as a TRA with no other TRA closer than at a given distance from the start or end position. The distance was fitted using the training series (FIG. 11C).A perfect discrimination between HRD and nonHRD cases was observed using distances from ~1.5 to 2.5 Mb, and the 2 Mb distance was selected to call the Isolated TRA SV subtype (FIG. 1 ID). To further validate the discriminative ability of this feature, numbers of Isolated TRA detected by OGM were compared with numbers of LGA identified by shallowHRDv2. Both variables correlated well (R2= 0.73; P = 9 x 1013, FIG. 10H), suggesting that these two independent methods measured a similar genomic hallmark. Importantly, Isolated TRA separated HRD from nonHRD cases with no overlap (range [16-40] vs [1-13], respectively; P = 9.49 x 10’8) (FIG. HE). Balanced_TRA, corresponding to reciprocal (or balanced) translocations where the two TRA occurred at an immediate vicinity with opposite orientations, were found significantly enriched in the HRD tumors as compared with nonHRD tumors (median±IQR= 9±5 vs 0±0.25 respectively; P = 1.37 x 10’6) (FIG. 1 IE).Classification models

[0154] Overall, five features discriminated HRD from nonHRD tumors: DEL<25kb, DUP<100kb, Isolated TRA and Balanced TRA enriched in HRD tumors and DUP INVplus enriched in nonHRD tumors. A classification model was developed which was adapted to thesemulticollinear features using LASSO logistic regression. This model was fitted on the 41 samples from the training set to predict the probability of being HRD. The final model used only three features, Isolated TRA being the most predictive, Balanced TRA and DUP INVplus (coefficients = 0.69, 0.17 and -0.18, respectively; FIG. 12A). The probability output classified samples as HRD when P > 0.55, as nonHRD when P < 0.45 (putative samples with P = 0.5 ±0.05 are declared borderline). The model was fully discriminant on the training cohort (Area Under Curve [AUC]=1, Mean-Squared Error [MSE]=0.0027 and R2=0.994).

[0155] Similarly, a model was developed to discriminate BRCA1 from BRCA2-\ke tumors using the two significant features: DUP<25kb (enriched in BRCA1 samples) and DEL<25kb (enriched in BRCA2-\ .e samples). Isolated TRA was also found discriminative for BRCA1 and BRCA2-\ e (TABLE 1), but the difference was best explained by the tumor origin (FIG. 101) and this feature was not included in the model. Model fitting was performed on the 15 samples with known major HRR gene inactivation from the training series (7 BRCAl 7 BRCA2 and 1 RAD51C). Both features were informative with DUP<25kb being the most predictive ( =0.17 for DUP<25kb; 0=-0.13 for DEL<25kb; FIG. 12B). Its probability output classified samples as BRCA1 when P>0.55, as BRCA2- .Q when P < 0.45 (putative samples with P = 0.5±0.05 are declared borderline). The model was fully discriminant on the training cohort (AUC=1, MSE=0.0021 and R2=0.996).Validation series

[0156] The classifiers were tested on an independent validation series of 19 TNBC and 18 HGOC with HRD diagnostics by shallowHRDv2 and HRR gene mutations status. TABLE 1. Two cases had insufficient material for OGM, three cases were discarded for undetectable tumor content and one case did not pass quality filters. In total, the validation series included 31 cases (31 / 37; 84%), including 11 HRD and 20 nonHRD cases. TABLE 4. The whole validation analysis was performed without manual supervision and with masked HRD status. Counts of all SV types and subtypes were calculated as previously described and the HRD classifier was applied. After unmasking the HRD status, 28 samples were correctly classified (FIG. 12C).

[0157] TABLE 4 lists certain characteristics of samples used in the validation series.TABLE 4i TsseuQ 5ShllV2aow §0- 1611 TNBC nH HRRtt manu 7 1 6 2 10 5 3 26 4 6 0.004169421gene1612 ['NBC nH 28 54 14 68 14 6 2 10 5 0 0.0035279651614 TNBC H BRCA2 14 3 6 11 45 29 8 58 23 8 0.999138781 0.003000414DUP100kb<1616 TNBC H BRCA1 49 27 38 38 24 51 4 65 17 4 0.950097808 0.919287095 1617 TNBC H 143 1 DUP100kb>1 121 33 10 17 8 83 23 10 0.9993893041826 TNBC nH 138 197 6 DUP25kb<0 275 16 58 44 123 17 1 0.0097036711827 HGOC nH 20 17 15 22 30 29 40 171 12 0 0.0005357DUP25kb>1828 HGOC nH 70 40 43 67 24 58 51 102 12 1 9.14342E-051830 HGOC nH 45 30 25 50 2 DEL25kb<7 38 39 65 14 0 0.0025244611592 HGOC H BRCA2 98 28 69 57 82 5 DEL25kb>6 10 72 24 10 0.999563113 0.52008918 1593 HGOC1 nH 16 37 8 45 10 3 DUPINYl 0pu 29 4 0 0.0025259721594 HGOC nH 40 16 34 22 32 28 38 1 TRA34 13 1 0.0017990431625 TNBC nH 21 24 13 32 25 28 36 145 13 1 0.002558174IldTRAtsoae1597 HGOC2 H 41 5 29 17 28 73 51 108 18 1 0.005638964BldTRaance1598 HGOC nH 26 6 18 14 17 30 29 45 18 1 0.2155821171599 HGOC nH 8 4 6 6 9 21 12 69 9 0 0.0094158541600 HGOC nH 19 11 8 22 11 11 5 15 5 1 0.0024() PHRD708971602 HGOC H 189 47 158 78 88 116 31 235 47 19 0.9999999991831 TNBC nH 43 61 23 81 24 22 38 120 16 1 0.0139944661603 HGOC nH 14 38 5 47 19 52 38 46 21 0 0.2714061571606 HGOC nH 12 24 9 27 20 4 11 30 6 0 0.0014378031607 HGOC H BRCA2 69 29 13 85 43 30 4 56 20 4 0.993374619 0.012794031 1608 HGOC H BRCA1 106 28 70 64 18 17 15 103 32 6 0.999991442 0.999832255 1832 HGOC H 12 10 9 13 46 17 9 60 27 7 0.9999220611834 TNBC nH 6 5 3 8 22 30 27 83 11 2 0.0037505321835 TNBC nH 8 29 6 31 10 19 29 41 11 0 0.0018719521833 TNBC H 241 34 211 64 28 40 6 54 30 6 0.9999930751836 TNBC H 20 10 16 14 37 64 18 23 10 0 0.0065196081626 TNBC nH meRAD51C 19 19 15 23 76 42 22 84 24 5 0.99149534 0.000234739 1837 TNBC nH 24 16 18 22 9 34 36 109 9 5 0.0003256051838 TNBC nH 14 6 10 10 13 15 13 26 11 0 0.030575606

[0158] The 3 discordant cases were further examined. A RAD51C methylation with loss of heterozygosity was found in one sample, fully explaining its OGM HRD status; the case was misclassified by the shallowHRDv2 test due to a noisy CNA profile. The two other samples, classifiedas HRD by shallowHRDv2, and as nonHRD by OGM, carried no detected HRR mutation (including BRCA1, BRCA2, PALB2, RAD51B, RAD51C, and RAD51D) and no methylation of BRCA1 or RAD51C promoters. One of the cases had a borderline score on its shallowHRDv2 test and low tumor content. The second case was a carcinosarcoma, a rare and aggressive form of ovarian carcinoma, and a borderline shallowHRD score. Notably, the HRD classification model achieved an accuracy of minimum 94%, with not a single discordance with the HRR genetic or epigenetic status.

[0159] The BRCA1 / 2 sub-classifier was tested on the 5 samples carrying known mutations, and the reclassified meRAD51C sample. Five samples were classified correctly, and one BRCA2 case was classified as borderline. The model achieved an accuracy of 83% (FIG. 12D).

[0160] Altogether, the OGM-based classifier correctly attributed HRD / nonHRD status to 94% of samples in the validation series, including correction of at least one case to confirmed HRD. Furthermore, the sub-classifier provided a correct assignment for the BRCA1 or BCRA2-like status for all HRD samples, but one borderline case.Tandem inverted duplications and inversions

[0161] It was believed that the enrichment in tandem inverted duplications (DUP INV) and inversions (INV) in nonHRD tumors had not been previously highlighted, while the number of inversions were used in HRDetect among other variables. To examine if DUP INV and INV reflected the frequent clustered alterations in nonHRD tumors, clustered and non-clustered SVs were defined and compared in HRD and nonHRD groups. No significant difference was found in non-clustered DUP INV and INV between HRD and nonHRD cases; the major differences appeared when comparing the number of inverted events clustered with more than 2 other SVs. Clustered DUP INV and INV were significantly enriched in nonHRD compared with HRD cases (median ± interquartile range [IQR] for DUP_INV 6.5 ± 12 vs 2 ± 3, respectively; P = 1.6 x 10‘3; for INV 4 ± 7 vs 1 ± 3, respectively; P = 3.5 x 1 O’4). No difference was found between TNBC and HGOC with respect to the inverted events (FIG. 13 A, FIG. 13B).Discussion

[0162] OGM was used to examine in TNBC and HGOC to identify HRD-associated structural variants. Informative SV features enriched in HRD were readily identified, such as small deletions (DEL<25kb) and duplications (DUP<100kb). Features including isolated translocations (Isolated TRA), and balanced translocations (Balanced TRA), both highly enriched in HRD tumors were also defined. Balanced translocations could not be detected from CNA profiles and representedan added value from OGM analysis. Finally, inverted duplications and inversions were found enriched in nonHRD tumors. Inversions were previously included into the output of WGS analyses but did not show independent predictive value for HRD detection, most probably because their amount correlated with clustered alterations. Tandem inverted duplications are probably more challenging to call from short reads WGS, which explained their absence in WGS-based HRD classifiers.

[0163] Embodiments provided herein include an OGM HRD classifier for TNBC and HGOC, using three selected features: Isolated TRA, Balanced TRA and DUP INVplus. Also disclosed are embodiments including a classifier to discriminate BRCA1 from B ?G42-like tumors, using two features: DUP<25kb (enriched mBRCAl cases) and DEL<25kb (enriched in BRCA2 ike cases). Validation of the classifiers in the independent validation series showed a high accuracy, with not a single discrepancy between the OGM HRD status and the genetic / epigenetic hallmarks of HRD. Two cases with discordant OGM and shallowHRDv2 status had shallowHRDv2 scores close to the borderline (including one case with low tumor content), and no detected HRR mutation or methylation, which support OGM predictions as the correct ones. The OGM-based HRD classifier had a high accuracy, with at least one sample re-classified to the confirmed HRD.

[0164] A more fundamental contribution of OGM approach is the detection of genomic rearrangements specifically associated with nonHRD tumors. Indeed, nonHRD TNBC and HGOC almost consistently harbor highly rearranged genome profiles, whereas underlying mechanisms explaining this instability are yet to unravel. Excess of tandem inverted duplications may represent a first clue, as they were recently associated with DNA polymerase 8-dependent double-strand break repair mechanism. The actual structure at the nucleotide resolution of the DUP INV detected by OGM is likely to provide important insights in the source of genomic instability involved in nonHRD breast and ovarian high-grade carcinomas.

[0165] OGM is a useful approach for HRD determination in high-grade breast and ovarian tumors. It is an affordable technique and highly efficient even with samples with low tumor content. It provides a very robust calling for large structural variants, even when involving low mappability regions (with the exception of centromeric and telomeric regions). Moreover, it does not require complex bioinformatic analyses or heavy data storage. Finally, the readily interpretable OGM genomic profile may also provide information about key molecular events occurring in the tumors, and to ultimately identify therapeutic targets.Materials and methodsOptical genome mapping

[0166] UHMW gDNA was extracted from frozen tissue by the Bionano Prep SP Tissue Kit (Bionano Genomics, Inc., San Diego, CA) following the manufacturer’ instructions. Samples with less than 0.75 pg of extracted DNA were excluded. DNA labeling was performed according to the Direct label stain (DLS) protocol with the Direct label enzyme 1 (DLE1) enzyme (CTTAAG sequence). Briefly, the DLS protocol includes an enzymatic labeling approach for direct fluorescent labeling of ultra-high molecular weight gDNA by DLE-1. The DLE-1 enzyme attaches a DL-Green fluorophore via covalent modification at a specific-sequence motif, such as the CTTAAG sequence. This labeling process is non-damaging to the gDNA, allowing the generation of highly contiguous genome maps and providing high sensitivity to detection of structural variants. After sequencespecific labeling with DLE-1, the labeled DNA was stained for backbone visualization.

[0167] Labelled molecules were linearized and guided on the Saphyr Chip® into nanochannels to be imaged. Samples were analyzed 3 by 3 for 3 days to obtain the highest possible coverage. Molecules were aligned together and compared to the hg38 reference map. Quality control (QC) was performed. Samples were excluded that failed to have an effective coverage > 100 X; ‘N50 > 20 kb’ (i.e. median size of molecules > 20 kb) > 100 kb; and map rate > 50%. TABLE 1 lists samples used in the training series and validation series, along with certain characteristics.Mining PGM structural variants

[0168] A RVA was performed using Bionano Access® software version 1.7 (Bionano Genomics, Inc., San Diego, CA) and Solve software version 3.7 (Bionano Genomics, Inc., San Diego, CA). Smap files were generated and data analyses were performed with a custom R script using the tidyverse and GenomicRanges libraries. The RVA can include detection of SVs by a “split-read” analysis, where initial molecule alignment and molecule extension refinement allow identification and detection of somatic variants and identifies variants that are 5,000 bp or longer in size. In some embodiments, a rare variant is present in low abundance in a sample and is not present in a reference molecule.

[0169] Common structural variants were filtered out using the PGM SVs database (RVA pipeline), as the corresponding germline samples were not analyzed. Variants with reported confidence <0.1 and SVs supported by less than 5 / 500 multiplied by effective coverage of the sample were filtered out.

[0170] Duplicates defined as SVs with genomic overlapping positions that differed less than 30 kb were detected. For translocations, duplicates had genomic positions that differed less than 30 kb, the same orientation and similar VAF. The duplicate with the highest confidence was kept.

[0171] Isolated translocations were detected using the function countOverlaps from the GenomicRanges R package. Translocations without any other overlapping translocation in a fixed neighborhood were considered as isolated. Balanced translocations were detected from translocation pairs with both positions overlapping within 30 kb and having the same orientation + / + or - / - or coupled orientations + / - and - / +, or vice versa.

[0172] Two SVs were named ‘coupled’ if they had less than 1 Mb distance between their closest positions. Clustered N alterations were called if they could be arranged into the chain of coupled SVs.Methylation-sensitive high-resolution melting (MS-HRM) analysis

[0173] BRCA1 and RAD51C promoter methylation was assessed with MS-HRM analysis.100 ng of DNA from each sample were treated with sodium bisulfite with the EZ DNA Methylation-Lightning Kit (Zymo Research, D5030-E, Freibourg, Germany), following the manufacturer's instructions. MS-HRM analysis was performed in a Roche LC480 instrument (Roche Diagnostics, Meylan, France). The promoter regions of BRCA1 and RAD51C were analyzed using specifically designed primers by Methyldetect (Aalborg, Denmark).

[0174] All the analyses were run as follows: 1 cycle of 95 °C for 10 min, 50 cycles of 95 °C for 15 s, 56 °C for BRCA1 and 62°C for RAD51C for 10 s and 72 °C for 15 s; followed by an HRM step of 95 °C for 15 s, 60 °C for 1 min and continuous acquisition to 95°. PCR was performed in a final volume of 20 pL, containing 10 pL LightCycler 480 HRM Master (Roche Diagnostics), 1 pL EpiMelt primer Mix, 2.4 pL MgCh, 0.6 pL H2O, and 6 pL of bisulfite modified DNA template. A methylation positive, an assay calibration, and a methylation negative control were supplied with the EpiMelt Kit (Methyldetect, Aalborg, Denmark). The assay calibration control was included to ensure the assay sensitivity to detect methylation of 1% in each experiment.Data and statistical analyses

[0175] A Wilcoxon Mann-Whitney nonparametric test was used to assess the SV count differences between two groups of samples. Correlation analyses were performed with a Spearman's test to calculate R2and P. P values < 0.05 were considered significant. The significance of the Pvalues was annotated on graphs as * if P < 0.05, ** if P < 0.005, *** if P < 0.001 and **** if P < 0.0001.

[0176] A least absolute shrinkage and selection operator (LASSO) logistic regression model was used to build the HRD classifier and the BRCAHBRCA2 sub-classifier from counts of all significant SV features / types / subtypes. Leave-one-out cross-validation was used to select the penalty X, using the cv.glmnet function from the R glmnet package. The minimizing the prediction errors (lambda.min) were 0.0028 and 0.0031, respectively. A glmnet function was used with the optimal X to calculate each coefficient.

[0177] The HRD classifier returns the probability (P) of a sample being HRD:where y = -8.7302 + (Q.6^>19-Isolated_TRA) + (0.1720 -Balanced TRA) + (-0.1764 -D UP INVplus) .

[0178] The sample is declared HRD if P(HRD) > 0.55, nonHRD if P(HRD) < 0.45 and borderline otherwise.

[0179] The BRCAU2 sub-classifier returns the probability of a sample being BRCA1 inactivated:1P(BRG41) =where y = 0.8959 + (0 A1W-DUP < 25 kb) + (-0.1319-DEL < 25 kb).

[0180] The sample is classified as BRCA1 inactivated ifP ~ (BRCAP) > 0.55, BRCA2-\\ e if ?(PjRCAl) < 0.45 and borderline otherwise. Predictions for both models were performed using the predict function from the R stats package.Additional Considerations

[0181] In at least some of the previously described embodiments, one or more elements used in an embodiment can interchangeably be used in another embodiment unless such a replacement is not technically feasible. It will be appreciated by those skilled in the art that various other omissions, additions and modifications may be made to the methods and structures described above without departing from the scope of the claimed subject matter. All such modifications and changes are intended to fall within the scope of the subject matter, as defined by the appended claims.

[0182] One skilled in the art will appreciate that, for this and other processes and methods disclosed herein, the functions performed in the processes and methods can be implemented indiffering order. Furthermore, the outlined steps and operations are only provided as examples, and some of the steps and operations can be optional, combined into fewer steps and operations, or expanded into additional steps and operations without detracting from the essence of the disclosed embodiments.

[0183] With respect to the use of substantially any plural and / or singular terms herein, those having skill in the art can translate from the plural to the singular and / or from the singular to the plural as is appropriate to the context and / or application. The various singular / plural permutations may be expressly set forth herein for sake of clarity. As used in this specification and the appended claims, the singular forms “a,” “an,” and “the” include plural references unless the context clearly dictates otherwise. Accordingly, phrases such as “a device configured to” are intended to include one or more recited devices. Such one or more recited devices can also be collectively configured to carry out the stated recitations. For example, “a processor configured to carry out recitations A, B and C can include a first processor configured to carry out recitation A and working in conjunction with a second processor configured to carry out recitations B and C. Any reference to “or” herein is intended to encompass “and / or” unless otherwise stated.

[0184] It will be understood by those within the art that, in general, terms used herein, and especially in the appended claims (e.g., bodies of the appended claims) are generally intended as “open” terms (e.g., the term “including” should be interpreted as “including but not limited to,” the term “having” should be interpreted as “having at least,” the term “includes” should be interpreted as “includes but is not limited to,” etc.). It will be further understood by those within the art that if a specific number of an introduced claim recitation is intended, such an intent will be explicitly recited in the claim, and in the absence of such recitation no such intent is present. For example, as an aid to understanding, the following appended claims may contain usage of the introductory phrases “at least one” and “one or more” to introduce claim recitations. However, the use of such phrases should not be construed to imply that the introduction of a claim recitation by the indefinite articles “a” or “an” limits any particular claim containing such introduced claim recitation to embodiments containing only one such recitation, even when the same claim includes the introductory phrases “one or more” or “at least one” and indefinite articles such as “a” or “an” (e.g., “a” and / or “an” should be interpreted to mean “at least one” or “one or more”); the same holds true for the use of definite articles used to introduce claim recitations. In addition, even if a specific number of an introduced claim recitation is explicitly recited, those skilled in the art will recognize that such recitation should be interpreted to mean at least the recited number (e.g., the bare recitation of “two recitations,” without other modifiers, means at least two recitations, or two or more recitations). Furthermore, in those instances where aconvention analogous to “at least one of A, B, and C, etc.” is used, in general such a construction is intended in the sense one having skill in the art would understand the convention (e.g., “ a system having at least one of A, B, and C” would include but not be limited to systems that have A alone, B alone, C alone, A and B together, A and C together, B and C together, and / or A, B, and C together, etc.). In those instances where a convention analogous to “at least one of A, B, or C, etc.” is used, in general such a construction is intended in the sense one having skill in the art would understand the convention (e.g., “ a system having at least one of A, B, or C” would include but not be limited to systems that have A alone, B alone, C alone, A and B together, A and C together, B and C together, and / or A, B, and C together, etc.). It will be further understood by those within the art that virtually any disjunctive word and / or phrase presenting two or more alternative terms, whether in the description, claims, or drawings, should be understood to contemplate the possibilities of including one of the terms, either of the terms, or both terms. For example, the phrase “A or B” will be understood to include the possibilities of “A” or “B” or “A and B.”

[0185] In addition, where features or aspects of the disclosure are described in terms of Markush groups, those skilled in the art will recognize that the disclosure is also thereby described in terms of any individual member or subgroup of members of the Markush group.

[0186] As will be understood by one skilled in the art, for any and all purposes, such as in terms of providing a written description, all ranges disclosed herein also encompass any and all possible sub-ranges and combinations of sub-ranges thereof. Any listed range can be easily recognized as sufficiently describing and enabling the same range being broken down into at least equal halves, thirds, quarters, fifths, tenths, etc. As a non-limiting example, each range discussed herein can be readily broken down into a lower third, middle third and upper third, etc. As will also be understood by one skilled in the art all language such as “up to,” “at least,” “greater than,” “less than,” and the like include the number recited and refer to ranges which can be subsequently broken down into sub-ranges as discussed above. Finally, as will be understood by one skilled in the art, a range includes each individual member. Thus, for example, a group having 1-3 articles refers to groups having 1, 2, or 3 articles. Similarly, a group having 1-5 articles refers to groups having 1, 2, 3, 4, or 5 articles, and so forth.

[0187] It will be appreciated that various embodiments of the present disclosure have been described herein for purposes of illustration, and that various modifications may be made without departing from the scope and spirit of the present disclosure. Accordingly, the various embodiments disclosed herein are not intended to be limiting, with the true scope and spirit being indicated by the following claims.

[0188] It is to be understood that not necessarily all objects or advantages may be achieved in accordance with any particular embodiment described herein. Thus, for example, those skilled in the art will recognize that certain embodiments may be configured to operate in a manner that achieves or optimizes one advantage or group of advantages as taught herein without necessarily achieving other objects or advantages as may be taught or suggested herein.

[0189] All of the processes described herein may be embodied in, and fully automated via, software code modules executed by a computing system that includes one or more computers or processors. The code modules may be stored in any type of non-transitory computer-readable medium or other computer storage device. Some or all the methods may be embodied in specialized computer hardware.

[0190] Many other variations than those described herein will be apparent from this disclosure. For example, depending on the embodiment, certain acts, events, or functions of any of the algorithms described herein can be performed in a different sequence, can be added, merged, or left out altogether (for example, not all described acts or events are necessary for the practice of the algorithms). Moreover, in certain embodiments, acts or events can be performed concurrently, for example through multi-threaded processing, interrupt processing, or multiple processors or processor cores or on other parallel architectures, rather than sequentially. In addition, different tasks or processes can be performed by different machines and / or computing systems that can function together.

[0191] The various illustrative logical blocks and modules described in connection with the embodiments disclosed herein can be implemented or performed by a machine, such as a processing unit or processor, a digital signal processor (DSP), an application specific integrated circuit (ASIC), a field programmable gate array (FPGA) or other programmable logic device, discrete gate or transistor logic, discrete hardware components, or any combination thereof designed to perform the functions described herein. A processor can be a microprocessor, but in the alternative, the processor can be a controller, microcontroller, or state machine, combinations of the same, or the like. A processor can include electrical circuitry configured to process computer-executable instructions. In another embodiment, a processor includes an FPGA or other programmable device that performs logic operations without processing computer-executable instructions. A processor can also be implemented as a combination of computing devices, for example a combination of a DSP and a microprocessor, a plurality of microprocessors, one or more microprocessors in conjunction with a DSP core, or any other such configuration. Although described herein primarily with respect to digital technology, a processor may also include primarily analog components. For example, someor all of the signal processing algorithms described herein may be implemented in analog circuitry or mixed analog and digital circuitry. A computing environment can include any type of computer system, including, but not limited to, a computer system based on a microprocessor, a mainframe computer, a digital signal processor, a portable computing device, a device controller, or a computational engine within an appliance, to name a few.

[0192] Any process descriptions, elements or blocks in the flow diagrams described herein and / or depicted in the attached figures should be understood as potentially representing modules, segments, or portions of code which include one or more executable instructions for implementing specific logical functions or elements in the process. Alternate implementations are included within the scope of the embodiments described herein in which elements or functions may be deleted, executed out of order from that shown, or discussed, including substantially concurrently or in reverse order, depending on the functionality involved as would be understood by those skilled in the art.

[0193] It should be emphasized that many variations and modifications may be made to the above-described embodiments, the elements of which are to be understood as being among other acceptable examples. All such modifications and variations are intended to be included herein within the scope of this disclosure and protected by the following claims.

Claims

WHAT IS CLAIMED IS:

1. A method for determining the presence or absence of a homologous recombination deficiency (HRD) in a sample, comprising:under control of a hardware processor:receiving genome mapping (GM) data generated from a sample from a subject; determining a plurality of structural variants (SVs) in the GM data by comparing the GM data with a reference genome; anddetermining the presence or absence of an HRD in the sample based on the plurality of SVs.

2. The method of claim 1, wherein the determining a plurality of SVs in the GM data comprises generating from the GM data a list comprising the plurality of SVs, and / or a map comprising the plurality of SVs.

3. The method of claim 1 or 2, wherein the determining a plurality of SVs in the GM data comprises identifying SV types in the plurality of SVs; optionally, wherein the SV types are selected from deletions, duplications, inversions, and / or translocations.

4. The method of any one of claims 1-3, wherein the determining a plurality of SVs in the GM data comprises removing common SVs from the plurality of SVs by removing from the plurality of SVs an SV having a variant allele frequency (VAF) > 001 in a plurality of reference genomes; optionally, wherein the plurality of references genomes is from subjects lacking a reported disease phenotype.

5. The method of any one of claims 1-4, wherein the determining a plurality of SVs in the GM data comprises determining the number of SVs in one or more SV types in the plurality of SVs.

6. The method of claim 5, wherein the one or more SV types is selected from a deletion, a duplication, an inversion, and / or a translocation.

7. The method of claim 5 or 6, wherein the one or more SV types is selected from an isolated translocation, a balanced translocation, an inverted duplication with a size in a range from 30 kb to 300 kb, an inversion, a deletion < 25 kb, and / or a duplication < 100 kb.

8. The method of any one of claims 5-7, wherein the one or more SV types is selected from an isolated translocation, a balanced translocation, an inverted duplication with a size in a range from 30 kb to 300 kb, and / or an inversion.

9. The method of claim 7 or 8, wherein each end of an isolated translocation is located at a distance from another translocation in an assembled map of the genome mapping data of at least 1.5 Mb, 2.0 Mb, 2.5 Mb, or 3.0 Mb; optionally, wherein the distance is at least 2.0 Mb.

10. The method of any one of claims 1-9, wherein the determining the presence or absence of an HRD comprises applying to an HRD classifier the number of SVs in each of the one of more SV types to provide a probability that a sample comprises an HRD.

11. The method of claim 10, wherein the number of SVs in each of the one of more SV types comprises the number of: (i) isolated translocations, (ii) balanced translocations, and (iii) inverted duplications with a size in a range from 30 kb to 300 kb together with the number of inversions.

12. The method of claim 10 or 11, wherein the HRD classifier is obtained by performing a least absolute shrinkage and selection operator (LASSO) logistic regression on the number of SVs in each SV type in a plurality of different SV types obtained from GM data from a plurality of control samples comprising an HRD and a plurality of control samples lacking an HRD.

13. The method of claim 12, wherein, the SVs obtained from GM data from the plurality of control samples lack an SV having a VAF > 001 in a plurality of reference genomes; optionally, wherein the plurality of references genomes is from subjects lacking a reported disease phenotype.

14. The method of claim 12 or 13, wherein the different SV types are selected from duplications, inverted duplications, inversions, translocations, and / or deletions; optionally isolated translocations, balanced translocations, inverted duplications with a size in a range from 30 kb to 300 kb, deletions < 25 kb, and / or duplications < 100 kb.

15. The method of any one of claims 1-14, wherein the determining the presence or absence of an HRD comprises comparing the number of SVs in one or more SV types in the plurality of SVs with the number of SVs in the one or more SV types in a control obtained from GM data from a sample lacking an HRD.

16. The method of claim 15, wherein an increase in the number of the SVs in the one or more SV types in the plurality of SVs compared to the number of SVs in the one or more SV types in the control is indicative of the sample having an HRD, wherein the one or more SV types is selected from an isolated translocation, a balanced translocation, and / or a duplication < 100 kb.

17. The method of claim 15, wherein a decrease in the number of the SVs in the one or more SV types in the plurality of SVs compared to the number of SVs in the one or more SV types in the control is indicative of the sample having an HRD, wherein the one or more SV types is selectedfrom an inverted duplication with a size in a range from 30 kb to 300 kb, an inversion, and / or a deletion < 25 kb.

18. The method of any one of claims 15-17, further comprising determining the presence or absence of an (i) inactive BRCA1 gene or (ii) inactive BRCA2 or RAD51C gene in the sample; optionally, wherein the sample comprises an HRD.

19. The method of claim 18, wherein the determining the presence or absence of an (i) inactive BRCA1 gene or (ii) inactive BRCA2 or RAD51C gene in the sample comprises applying to a BRCAHBRCA2-\ Q classifier the number of SVs in each of the one of more SV types to provide a probability that a sample comprises an (i) inactive BRCA1 gene or (ii) inactive BRCA2 or RAD51C gene; optionally, wherein the one of more SV types comprises a deletion < 25 kb, and / or a duplication < 100 kb.

20. The method of claim 18, comprising comparing the number of SVs in one or more SV types in the plurality of SVs with the number of SVs in the one or more SV types in a control obtained from GM data from a sample lacking inactive BRCA1, BRCA2 and RAD51C genes.

21. The method of claim 20, wherein an increase in the number of the SVs in the one or more SV types in the plurality of SVs compared to the number of SVs in the one or more SV types in the control is indicative of the sample having an HRD, wherein the one or more SV types comprise a duplication < 100 kb.

22. The method of claim 20, wherein a decrease in the number of the SVs in the one or more SV types in the plurality of SVs compared to the number of SVs in the one or more SV types in the control is indicative of the sample having an HRD, wherein the one or more SV types comprise a deletion < 25 kb.

23. A method for determining the presence or absence of an (i) inactive BRCA1 gene or (ii) inactive BRCA2 o RAD51C gene in a sample comprising a homologous recombination deficiency (HRD), comprising:under control of a hardware processor:receiving genome mapping (GM) data generated from the sample from a subject; determining a plurality of structural variants (SVs) in the GM data by comparing the GM data with a reference genome; anddetermining the presence or absence of an (i) inactive BRCA1 gene or (ii) inactive BRCA2 or RAD51C gene in the sample based on the plurality of SVs.

24. The method of claim 23, wherein the determining a plurality of SVs in the GM data comprises one or more of:(a) generating from the GM data a list comprising the plurality of SVs, and / or a map comprising the plurality of SVs;(b) identifying SV types in the plurality of SVs, optionally, wherein the SV types are selected from deletions, duplications, inversions, and / or translocations;(c) removing from the plurality of SVs an SV having a variant allele frequency (VAF) > 001 in a plurality of reference genomes; optionally, wherein the plurality of references genomes is from subjects lacking a reported disease phenotype; and / or(d) determining the number of SVs in one or more SV types in the plurality of SVs.

25. The method of claim 24, wherein the one or more SV types is selected from a deletion, a duplication; optionally, wherein the one or more SV types is selected from a deletion < 25 kb, and / or a duplication < 100 kb.

26. The method of claim 24 or 25, wherein the determining the presence or absence comprises applying to a BRCA1 / BRCA2 -like classifier the number of SVs in each of the one of more SV types to provide a probability that a sample comprises an (i) inactive BRCA1 gene or (ii) inactive BRCA2 or RAD51C gene; optionally, wherein the one of more SV types comprises a deletion < 25 kb, and / or a duplication < 100 kb.

27. The method of claim 24 or 25, wherein the determining the presence or absence comprises comparing the number of SVs in one or more SV types in the plurality of SVs with the number of SVs in the one or more SV types in a control obtained from GM data from a sample lacking inactive BRCA1, BRCA2 and RAD51C genes; optionally, wherein:(i) an increase in the number of the SVs in the one or more SV types in the plurality of SVs compared to the number of SVs in the one or more SV types in the control is indicative of the sample having an HRD, wherein the one or more SV types comprise a duplication < 100 kb; or(ii) a decrease in the number of the SVs in the one or more SV types in the plurality of SVs compared to the number of SVs in the one or more SV types in the control is indicative of the sample having an HRD, wherein the one or more SV types comprise a deletion < 25 kb.

28. The method of any one of claims 1-27, wherein the sample comprises cells, cell-free DNA, cell-free fetal DNA, amniotic fluid, a blood sample, a bone marrow sample, a biopsy sample, or a combination thereof.

29. The method of any one of claims 1-28, wherein the sample comprises a cancer cell or sample.

30. The method of any one of claims 1-29, wherein the sample comprises an ovarian cancer cell or sample or a breast cancer cells or sample.

31. The method of any one of claims 1 -30, wherein the sample comprises a triple negative breast cancer (TNBC) cell or sample or a high grade ovarian cancer (HGOC) cell or sample.

32. The method of any one of claims 1-31, wherein the sample comprises a formalin-fixed paraffin-embedded (FFPE) sample.

33. The method of any one of claims 1-32, wherein the GM data comprises a consensus map for the sample from the subject.

34. The method of any one of claims 1-33, wherein the GM data is obtained by labelling genomic DNA from the sample from the subject with a detectable tag comprising a fluorophore, a quantum dot, a dendrimer, a nanowire, a bead, a hapten, streptavidin, avidin, neutravidin, biotin, a stabilized reactive group a radiolabel, an electromagnetic label, an azide, Dibenzocyclooctyne (DBCO), or an alkyne.

35. The method of any one of claims 1-34, wherein the GM data comprises optical genome mapping (OGM) data.

36. The method of any one of claims 1-35, wherein the GM data comprises electronic genome mapping (EGM) data.

37. The method of any one of claims 1-36, comprising: creating a file or a report and / or generating a user interface (UI) comprising a UI element representing or comprising (i) the SV types in the GM data from the sample; (ii) the number of SVs in each SV type in the GM data; (iii) a probability that the sample comprises an HRD; and / or (iv) a probability that an HRD sample comprises an (a) inactive BRCA1 gene or (b) inactive BRCA2 or RAD51C gene.

38. A system for determining the presence or absence of a homologous recombination deficiency (HRD) in a sample, comprising:non-transitory memory configured to store:executable instructions;genome mapping (GM) data generated from a sample from a subject;a hardware processor in communication with the non-transitory memory, wherein the hardware processor is programmed by the executable instructions to perform:determining a plurality of structural variants (SVs) in the GM data by comparing the GM data with a reference genome; anddetermining the presence or absence of an HRD in the sample based on the plurality of SVs.

39. The system of claim 38, wherein the hardware processor is programmed by the executable instructions to perform: generating from the GM data a list comprising the plurality of SVs, and / or a map comprising the plurality of SVs .

40. The system of claim 39, wherein the non-transitory memory is configured to store: the list and / or the map.

41. The system of claim 39 or 40, wherein the hardware processor is programmed by the executable instructions to perform: identifying SV types in the plurality of SVs; optionally, wherein the SV types are selected from deletions, duplications, inversions, and / or translocations.

42. The system of any one of claims 38-41, wherein the hardware processor is programmed by the executable instructions to perform: removing common SVs from the plurality of SVs by removing from the plurality of SVs an SV having a variant allele frequency (VAF) > 001 in a plurality of reference genomes; optionally, wherein the plurality of references genomes is from subjects lacking a reported disease phenotype.

43. The system of any one of claims 38-42, wherein the hardware processor is programmed by the executable instructions to perform: determining the number of SVs in one or more SV types in the plurality of SVs.

44. The system of claim 43, wherein the one or more SV types is selected from a deletion, a duplication, an inversion, and / or a translocation.

45. The system of claim 43 or 44, wherein the one or more SV types is selected from an isolated translocation, a balanced translocation, an inverted duplication with a size in a range from 30 kb to 300 kb, an inversion, a deletion < 25 kb, and / or a duplication < 100 kb.

46. The system of any one of claims 43-45, wherein the one or more SV types is selected from an isolated translocation, a balanced translocation, an inverted duplication with a size in a range from 30 kb to 300 kb, and / or an inversion.

47. The system of claim 45 or 46, wherein each end of an isolated translocation is located at a distance from another translocation in an assembled map of the genome mapping data of at least 1.5 Mb, 2.0 Mb, 2.5 Mb, or 3.0 Mb; optionally, wherein the distance is at least 2.0 Mb.

48. The system of claim 47, wherein the non-transitory memory configured to store an HRD classifier.

49. The system of claim 47, wherein the hardware processor is programmed by the executable instructions to perform: applying to an HRD classifier the number of SVs in each of the one of more SV types to provide a probability that a sample comprises an HRD.

50. The system of claim 49, wherein the number of SVs in each of the one of more SV types comprises the number of: (i) isolated translocations, (ii) balanced translocations, and (iii) inverted duplications with a size in a range from 30 kb to 300 kb together with the number of inversions.

51. The system of claim 49 or 50, wherein the HRD classifier is obtained by performing a least absolute shrinkage and selection operator (LASSO) logistic regression on the number of SVs in each SV type in a plurality of different SV types obtained from GM data from a plurality of control samples comprising an HRD and a plurality of control samples lacking an HRD.

52. The system of claim 51, wherein the SVs obtained from GM data from the plurality of control samples lack an SV having a VAF > 001 in a plurality of reference genomes; optionally, wherein the plurality of references genomes is from subjects lacking a reported disease phenotype.

53. The system of claim 51 or 52, wherein the different SV types are selected from duplications, inverted duplications, inversions, translocations, and / or deletions; optionally isolated translocations, balanced translocations, inverted duplications with a size in a range from 30 kb to 300 kb, deletions < 25 kb, and / or duplications < 100 kb.

54. The system of any one of claims 38-53, wherein the hardware processor is programmed by the executable instructions to perform: comparing the number of SVs in one or more SV types in the plurality of SVs with the number of SVs in the one or more SV types in a control obtained from GM data from a sample lacking an HRD.

55. The system of claim 54, wherein an increase in the number of the SVs in the one or more SV types in the plurality of SVs compared to the number of SVs in the one or more SV types in the control is indicative of the sample having an HRD, wherein the one or more SV types is selected from an isolated translocation, a balanced translocation, and / or a duplication < 100 kb.

56. The system of claim 54, wherein a decrease in the number of the SVs in the one or more SV types in the plurality of SVs compared to the number of SVs in the one or more SV types in the control is indicative of the sample having an HRD, wherein the one or more SV types is selected from an inverted duplication with a size in a range from 30 kb to 300 kb, an inversion, and / or a deletion < 25 kb.

57. The system of any one of claims 38-56, wherein the hardware processor is programmed by the executable instructions to perform: determining the presence or absence of an (i) inactive BRCA1 gene or (ii) inactive BRCA2 or RAD51C gene in the sample; optionally, wherein the sample comprises an HRD.

58. The system of any one of claims 38-57, wherein the non-transitory memory configured to store: a BRCAl / BRCA2-like classifier.

59. The system of claim 57, wherein the hardware processor is programmed by the executable instructions to perform: applying to a BRCAl / BRC A 2 -like classifier the number of SVs in each of the one of more SV types to provide a probability that a sample comprises an (i) inactive BRCAl gene or (ii) inactive BRCA2 or RAD51C gene; optionally, wherein the one of more SV types comprises a deletion < 25 kb, and / or a duplication < 100 kb.

60. The system of claim 57, wherein the hardware processor is programmed by the executable instructions to perform: comparing the number of SVs in one or more SV types in the plurality of SVs with the number of SVs in the one or more SV types in a control obtained from GM data from a sample lacking inactive BRCAl, BRCA2 and RAD51C genes.

61. The system of claim 60, wherein an increase in the number of the SVs in the one or more SV types in the plurality of SVs compared to the number of SVs in the one or more SV types in the control is indicative of the sample having an HRD, wherein the one or more SV types comprise a duplication < 100 kb.

62. The system of claim 60, wherein a decrease in the number of the SVs in the one or more SV types in the plurality of SVs compared to the number of SVs in the one or more SV types in the control is indicative of the sample having an HRD, wherein the one or more SV types comprise a deletion < 25 kb.

63. A system for determining the presence or absence of an (i) inactive BRCAl gene or (ii) inactive BRCA2 o RAD51C gene in a sample comprising a homologous recombination deficiency (HRD), comprising:non-transitory memory configured to store:executable instructions;genome mapping (GM) data generated from a sample from a subject;a hardware processor in communication with the non-transitory memory, wherein the hardware processor is programmed by the executable instructions to perform:determining a plurality of structural variants (SVs) in the GM data by comparing the GM data with a reference genome; anddetermining the presence or absence of an (i) inactive BRCAl gene or (ii) inactive BRCA2 or RAD51C gene in the sample based on the plurality of SVs.

64. The system of claim 63, wherein the hardware processor is programmed by the executable instructions to perform: one or more of:(a) generating from the GM data a list comprising the plurality of SVs, and / or a map comprising the plurality of SVs; optionally, wherein the non-transitory memory is configured to store the list and / or the map.(b) identifying SV types in the plurality of SVs, optionally wherein the SV types are selected from deletions, duplications, inversions, and / or translocations;(c) removing common SVs from the plurality of SVs by removing from the plurality of SVs an SV having a variant allele frequency (VAF) > 001 in a plurality of reference genomes; optionally, wherein the plurality of references genomes is from subjects lacking a reported disease phenotype; and / or(d) determining the number of SVs in one or more SV types in the plurality of SVs.

65. The system of claim 64, wherein the one or more SV types is selected from a deletion, a duplication; optionally, wherein the one or more SV types is selected from a deletion < 25 kb, and / or a duplication < 100 kb.

66. The system of any one of claims 63-65, wherein the non-transitory memory configured to store: a BRCA1 / BRCA2-like classifier.

67. The system of claim 64 or 65, wherein the hardware processor is programmed by the executable instructions to perform: applying to aBRCAJ / BRCA2-like classifier the number of SVs in each of the one of more SV types to provide a probability that a sample comprises an (i) inactive BRCA1 gene or (ii) inactive BRCA2 or RAD5IC gene; optionally, wherein the one of more SV types comprises a deletion < 25 kb, and / or a duplication < 100 kb.

68. The system of claim 64 or 65, wherein the hardware processor is programmed by the executable instructions to perform: comparing the number of SVs in one or more SV types in the plurality of SVs with the number of SVs in the one or more SV types in a control obtained from GM data from a sample lacking inactive BRCA1, BRCA2 and RAD51C genes; optionally, wherein:(i) an increase in the number of the SVs in the one or more SV types in the plurality of SVs compared to the number of SVs in the one or more SV types in the control is indicative of the sample having an HRD, wherein the one or more SV types comprise a duplication < 100 kb; or(ii) a decrease in the number of the SVs in the one or more SV types in the plurality of SVs compared to the number of SVs in the one or more SV types in the control is indicative of the sample having an HRD, wherein the one or more SV types comprise a deletion < 25 kb.

69. The system of any one of claims 38-68, wherein the sample comprises cells, cell-free DNA, cell-free fetal DNA, amniotic fluid, a blood sample, a bone marrow sample, a biopsy sample, or a combination thereof.

70. The system of any one of claims 38-69, wherein the sample comprises a cancer cell or sample.

71. The system of any one of claims 38-70, wherein the sample comprises an ovarian cancer cell or sample or a breast cancer cell or sample.

72. The system of any one of claims 38-71, wherein the sample comprises a triple negative breast cancer (TNBC) cell or sample or a high grade ovarian cancer (HGOC) cell or sample.

73. The system of any one of claims 38-72, wherein the sample comprises a formalin-fixed paraffin-embedded (FFPE) sample.

74. The system of any one of claims 38-73, wherein the GM data comprises a consensus map for the sample from the subject.

75. The system of any one of claims 38-74, wherein the GM data is obtained by labelling genomic DNA from the sample from the subject with a detectable tag comprising a fluorophore, a quantum dot, a dendrimer, a nanowire, a bead, a hapten, streptavidin, avidin, neutravidin, biotin, a stabilized reactive group a radiolabel, an electromagnetic label, an azide, dibenzocyclooctyne (DBCO), or an alkyne.

76. The system of any one of claims 38-75, wherein the GM data comprises optical genome mapping (OGM) data.

77. The system of any one of claims 38-76, wherein the GM data comprises electronic genome mapping (EGM) data.

78. The system of any one of claims 38-77, wherein the hardware processor is programmed by the executable instructions to perform: creating a file or a report and / or generating a user interface (UI) comprising a UI element representing or comprising (i) the SV types in the GM data from the sample; (ii) the number of SVs in each SV type in the GM data; (iii) a probability that the sample comprises an HRD; and / or (iv) a probability that an HRD sample comprises an (a) inactive BRCA1 gene or (b) inactive BRCA2 or RAD51C gene.

Images