Systems and methods for predicting homologous recombination deficiency status of a specimen
Methods, systems, and software are provided for an ensemble model trained to distinguish between cancers with homologous recombination pathway deficiencies (HRD positive cancers) and cancers without homologous recombination pathway deficiencies (HRD negative cancers) based on nucleic acid sequencing data, e.g., both RNA and DNA sequencing data, generated from a cancerous tissue sample of the subject.
1. A method of obtaining a homologous recombination pathway status of a cancer in a test subject, the method comprising:
at a computer system having one or more processors, and memory storing one or more programs for execution by the one or more processors:
(A) obtaining a first plurality of at least 10,000 sequence reads, in electronic form, of a DNA sample from the test subject, the DNA sample comprising DNA molecules from a cancerous tissue of the subject;
(B) obtaining a second plurality of at least 10,000 sequence reads, in electronic, of an RNA sample from the test subject, the RNA sample comprising RNA molecules from the cancerous tissue of the subject; and
(C) evaluating the first plurality of at least 10,000 sequence reads and the second plurality of at least 10,000 sequence reads using an ensemble model trained to distinguish between cancers with homologous recombination pathway deficiencies and cancers without homologous recombination pathway deficiencies, wherein the evaluating comprises:
determining, using the first plurality of at least 10,000 sequence reads, a first prediction for the homologous recombination pathway status of the cancerous tissue of the subject based on a measure of genome-wide loss of heterozygosity for the cancerous tissue of the subject;
determining, using the second plurality of at least 10,000 sequence reads, a second prediction for the homologous recombination pathway status of the cancerous tissue of the subject based on the expression levels of a plurality of genes in the cancerous tissue of the subject;
determining, using the second plurality of at least 10,000 sequence reads, a third prediction for the homologous recombination pathway status of the cancerous tissue of the subject based on a measure of transcriptome-wide rearrangements for the cancerous tissue of the subject; and
determining a final prediction for the homologous recombination pathway status based on the first prediction, the second prediction, and the third prediction,
thereby obtaining the homologous recombination pathway status of the test subject as an output of the ensemble model.
2. The method of claim 1 , wherein the measure of transcriptome-wide rearrangements is a measure of sequence insertions, sequence deletions, sequence inversions, and sequence translocations identified in the second plurality of sequence reads.
3. The method of claim 1 , wherein:
the method further comprises determining, based on the first plurality of at least 10,000 sequence reads, a fourth prediction for the homologous recombination pathway status of the cancerous tissue of the subject based on a measure of genome-wide rearrangements for the cancerous tissue of the subject; and
the final prediction is further based on the fourth prediction for the homologous recombination pathway status of the cancerous tissue of the subject.
4. The method of claim 3 , wherein the measure of genome-wide rearrangements is a measure of sequence insertions, sequence deletions, sequence inversions, and sequence translocations identified in the first plurality of sequence reads.
5. The method of claim 1 , wherein determining the third prediction is further based on a cancer type of the cancerous tissue of the subject.
6. The method of claim 1 , wherein:
the method further comprises determining, based on the second plurality of at least 10,000 sequence reads, a fourth prediction for the homologous recombination pathway status of the cancerous tissue of the subject based on a plurality of single-sample gene set enrichment analysis (ssGSEA) scores for the transcriptional profile of the cancerous tissue of the subject; and
the final prediction is further based on the fourth prediction for the homologous recombination pathway status of the cancerous tissue of the subject.
7. The method of claim 6 , wherein determining the fourth prediction is further based on a cancer type of the cancerous tissue of the subject.
8. The method of claim 1 , wherein:
the method further comprises obtaining a third plurality of at least 10,000 sequence reads comprising the methylation status of cytosine nucleotides, in electronic form, of a DNA sample from the test subject, the DNA sample comprising DNA molecules from a cancerous tissue of the subject;
determining, based on the third plurality of at least 10,000 sequence reads, a fourth prediction for the homologous recombination pathway status of the cancerous tissue of the subject based on a genomic methylation pattern of the cancerous tissue; and
the final prediction is further based on the fourth prediction for the homologous recombination pathway status of the cancerous tissue of the subject.
9. The method of claim 8 , wherein the first plurality of sequence reads and the third plurality of sequence reads were generated using different aliquots of the same DNA sample comprising the DNA molecules from the cancerous tissue of the subject.
10. The method of claim 8 , wherein the genomic methylation pattern of the cancerous tissue comprises a methylation pattern for a promoter region of a homologous recombination gene.
11. The method of claim 8 , wherein determining the fourth prediction is further based on a cancer type of the cancerous tissue of the subject.
12. The method of claim 1 , wherein:
the method further comprises obtaining a third plurality of at least 10,000 sequence reads, in electronic form, of a second DNA sample from the test subject, the second DNA sample comprising DNA molecules from a non-cancerous tissue of the subject; and
determining the first prediction is based on the first plurality of at least 10,000 sequence reads and the third plurality of at least 10,000 sequence reads.
13. The method of claim 1 , wherein determining the first prediction is further based on a cancer type of the cancerous tissue of the subject.
14. The method of claim 1 , wherein the cancerous tissue of the subject does not have a variant BRCA1 gene and does not have a variant BRCA2 gene.
15. The method of claim 1 , wherein the cancerous tissue of the subject has no more than one variant BRCA1 or BRCA2 gene.
16. The method of claim 1 , wherein the method further comprises:
when it is determined that the cancer in the test subject is homologous recombination deficiency (HRD) positive, treating the cancer by administering a poly ADP ribose polymerase (PARP) inhibitor to the test subject; and
when it is determined the cancer in the test subject is not homologous recombination deficiency (HRD) positive, treating the cancer with a therapy that does not include administration of a PARP inhibitor to the test subject.
17. The method of claim 1 , wherein the method further comprises:
when it is determined that the cancer in the test subject is homologous recombination deficiency (HRD) positive, treating the cancer by administering a platinum-containing neoadjuvant chemotherapy to the test subject; and
when it is determined the cancer in the test subject is not homologous recombination deficiency (HRD) positive, treating the cancer with a therapy that does not include administration of a platinum-containing neoadjuvant chemotherapy to the test subject.
18. The method of claim 17 , wherein the cancer is a triple-negative breast cancer.
19. A method of obtaining a homologous recombination pathway status of a cancer in a test subject, the method comprising:
at a computer system having one or more processors, and memory storing one or more programs for execution by the one or more processors:
(A) obtaining a first plurality of at least 10,000 sequence reads, in electronic form, of a DNA sample from the test subject, the DNA sample comprising DNA molecules from a cancerous tissue of the subject;
(B) obtaining a second plurality of at least 10,000 sequence reads, in electronic form, of an RNA sample from the test subject, the RNA sample comprising RNA molecules from the cancerous tissue of the subject; and
(C) evaluating the first plurality of at least 10,000 sequence reads and the second plurality of at least 10,000 sequence reads using an ensemble model trained to distinguish between cancers with homologous recombination pathway deficiencies and cancers without homologous recombination pathway deficiencies, wherein the evaluating comprises:
determining, based on the first plurality of at least 10,000 sequence reads, a first prediction for the homologous recombination pathway status of the cancerous tissue of the subject based on a measure of genome-wide loss of heterozygosity for the cancerous tissue of the subject;
determining, based on the second plurality of at least 10,000 sequence reads, a second prediction for the homologous recombination pathway status of the cancerous tissue of the subject based on the expression levels of a plurality of genes in the cancerous tissue of the subject;
determining, based on the first plurality of at least 10,000 sequence reads, a third prediction for the homologous recombination pathway status of the cancerous tissue of the subject based on a measure of genome-wide rearrangements for the cancerous tissue of the subject; and
determining a final prediction for the homologous recombination pathway status based on the first prediction, the second prediction, and the third prediction,
thereby obtaining the homologous recombination pathway status of the test subject as an output of the ensemble model.
20. A method of obtaining a homologous recombination pathway status of a cancer in a test subject, the method comprising:
at a computer system having one or more processors, and memory storing one or more programs for execution by the one or more processors:
(A) obtaining a first plurality of at least 10,000 sequence reads, in electronic form, of a DNA sample from the test subject, the DNA sample comprising DNA molecules from a cancerous tissue of the subject;
(B) obtaining a second plurality of at least 10,000 sequence reads, in electronic form, of an RNA sample from the test subject, the RNA sample comprising RNA molecules from the cancerous tissue of the subject;
(C) obtaining a third plurality of at least 10,000 sequence reads, in electronic form, of a second DNA sample from the test subject, the second DNA sample comprising DNA molecules from a non-cancerous tissue of the subject;
(D) evaluating the first plurality of at least 10,000 sequence reads, the second plurality of at least 10,000 sequence reads, and the third plurality of sequence reads using an ensemble model trained to distinguish between cancers with homologous recombination pathway deficiencies and cancers without homologous recombination pathway deficiencies, wherein the evaluating comprises:
determining, based on the first plurality of at least 10,000 sequence reads and the third plurality of at least 10,000 sequence reads, a first prediction for the homologous recombination pathway status of the cancerous tissue of the subject based on a measure of genome-wide loss of heterozygosity for the cancerous tissue of the subject; and
determining, based on the second plurality of at least 10,000 sequence reads, a second prediction for the homologous recombination pathway status of the cancerous tissue of the subject based on the expression levels of a plurality of genes in the cancerous tissue of the subject;
thereby obtaining the homologous recombination pathway status of the test subject.
21. A computer system comprising:
one or more processors; and
a non-transitory computer-readable medium including computer-executable instructions that, when executed by the one or more processors, cause the processors to perform a method for obtaining a homologous recombination pathway status of a cancer in a test subject, the method comprising:
(A) obtaining a first plurality of at least 10,000 sequence reads, in electronic form, of a DNA sample from the test subject, the DNA sample comprising DNA molecules from a cancerous tissue of the subject;
(B) obtaining a second plurality of at least 10,000 sequence reads, in electronic form, of an RNA sample from the test subject, the RNA sample comprising RNA molecules from the cancerous tissue of the subject; and
(C) evaluating the first plurality of at least 10,000 sequence reads and the second plurality of at least 10,000 sequence reads using an ensemble model trained to distinguish between cancers with homologous recombination pathway deficiencies and cancers without homologous recombination pathway deficiencies, wherein the evaluating comprises:
determining, using the first plurality of at least 10,000 sequence reads, a first prediction for the homologous recombination pathway status of the cancerous tissue of the subject based on a measure of genome-wide loss of heterozygosity for the cancerous tissue of the subject;
determining, using the second plurality of at least 10,000 sequence reads, a second prediction for the homologous recombination pathway status of the cancerous tissue of the subject based on the expression levels of a plurality of genes in the cancerous tissue of the subject;
determining, using the second plurality of at least 10,000 sequence reads, a third prediction for the homologous recombination pathway status of the cancerous tissue of the subject based on a measure of transcriptome-wide rearrangements for the cancerous tissue of the subject; and
determining a final prediction for the homologous recombination pathway status based on the first prediction, the second prediction, and the third prediction,
thereby obtaining the homologous recombination pathway status of the test subject.
22. A non-transitory computer-readable storage medium having stored thereon program code instructions that, when executed by a processor, cause the processor to perform a method for obtaining a homologous recombination pathway status of a cancer in a test subject, the method comprising:
(A) obtaining a first plurality of at least 10,000 sequence reads, in electronic form, of a DNA sample from the test subject, the DNA sample comprising DNA molecules from a cancerous tissue of the subject;
(B) obtaining a second plurality of at least 10,000 sequence reads, in electronic form, of an RNA sample from the test subject, the RNA sample comprising RNA molecules from the cancerous tissue of the subject; and
(C) evaluating the first plurality of at least 10,000 sequence reads and the second plurality of at least 10,000 sequence reads using an ensemble model trained to distinguish between cancers with homologous recombination pathway deficiencies and cancers without homologous recombination pathway deficiencies, wherein the evaluating comprises:
determining, based on the first plurality of at least 10,000 sequence reads, a first prediction for the homologous recombination pathway status of the cancerous tissue of the subject based on a measure of genome-wide loss of heterozygosity for the cancerous tissue of the subject;
determining, based on the second plurality of at least 10,000 sequence reads, a second prediction for the homologous recombination pathway status of the cancerous tissue of the subject based on the expression levels of a plurality of genes in the cancerous tissue of the subject;
determining, based on the first plurality of at least 10,000 sequence reads, a third prediction for the homologous recombination pathway status of the cancerous tissue of the subject based on a measure of genome-wide rearrangements for the cancerous tissue of the subject; and
determining a final prediction for the homologous recombination pathway status based on the first prediction, the second prediction, and the third prediction,
thereby obtaining the homologous recombination pathway status of the test subject as an output of the ensemble model.
23. A computer system comprising:
one or more processors; and
a non-transitory computer-readable medium including computer-executable instructions that, when executed by the one or more processors, cause the processors to perform a method for obtaining a homologous recombination pathway status of a cancer in a test subject, the method comprising:
(A) obtaining a first plurality of at least 10,000 sequence reads, in electronic form, of a DNA sample from the test subject, the DNA sample comprising DNA molecules from a cancerous tissue of the subject;
(B) obtaining a second plurality of at least 10,000 sequence reads, in electronic form, of an RNA sample from the test subject, the RNA sample comprising RNA molecules from the cancerous tissue of the subject; and
(C) evaluating the first plurality of at least 10,000 sequence reads and the second plurality of at least 10,000 sequence reads using an ensemble model trained to distinguish between cancers with homologous recombination pathway deficiencies and cancers without homologous recombination pathway deficiencies, wherein the evaluating comprises:
determining, based on the first plurality of at least 10,000 sequence reads, a first prediction for the homologous recombination pathway status of the cancerous tissue of the subject based on a measure of genome-wide loss of heterozygosity for the cancerous tissue of the subject;
determining, based on the second plurality of at least 10,000 sequence reads, a second prediction for the homologous recombination pathway status of the cancerous tissue of the subject based on the expression levels of a plurality of genes in the cancerous tissue of the subject;
determining, based on the first plurality of at least 10,000 sequence reads, a third prediction for the homologous recombination pathway status of the cancerous tissue of the subject based on a measure of genome-wide rearrangements for the cancerous tissue of the subject; and
determining a final prediction for the homologous recombination pathway status based on the first prediction, the second prediction, and the third prediction,
thereby obtaining the homologous recombination pathway status of the test subject as an output of the ensemble model.
24. A non-transitory computer-readable storage medium having stored thereon program code instructions that, when executed by a processor, cause the processor to perform a method for obtaining a homologous recombination pathway status of a cancer in a test subject, the method comprising:
(A) obtaining a first plurality of at least 10,000 sequence reads, in electronic form, of a DNA sample from the test subject, the DNA sample comprising DNA molecules from a cancerous tissue of the subject;
(B) obtaining a second plurality of at least 10,000 sequence reads, in electronic form, of an RNA sample from the test subject, the RNA sample comprising RNA molecules from the cancerous tissue of the subject; and
(C) evaluating the first plurality of at least 10,000 sequence reads and the second plurality of at least 10,000 sequence reads using an ensemble model trained to distinguish between cancers with homologous recombination pathway deficiencies and cancers without homologous recombination pathway deficiencies, wherein the evaluating comprises:
determining, based on the first plurality of at least 10,000 sequence reads, a first prediction for the homologous recombination pathway status of the cancerous tissue of the subject based on a measure of genome-wide loss of heterozygosity for the cancerous tissue of the subject;
determining, based on the second plurality of at least 10,000 sequence reads, a second prediction for the homologous recombination pathway status of the cancerous tissue of the subject based on the expression levels of a plurality of genes in the cancerous tissue of the subject;
determining, based on the first plurality of at least 10,000 sequence reads, a third prediction for the homologous recombination pathway status of the cancerous tissue of the subject based on a measure of genome-wide rearrangements for the cancerous tissue of the subject; and
determining a final prediction for the homologous recombination pathway status based on the first prediction, the second prediction, and the third prediction,
thereby obtaining the homologous recombination pathway status of the test subject as an output of the ensemble model.
25. A computer system comprising:
one or more processors; and
a non-transitory computer-readable medium including computer-executable instructions that, when executed by the one or more processors, cause the processors to perform a method for obtaining a homologous recombination pathway status of a cancer in a test subject, the method comprising:
(A) obtaining a first plurality of at least 10,000 sequence reads, in electronic form, of a DNA sample from the test subject, the DNA sample comprising DNA molecules from a cancerous tissue of the subject;
(B) obtaining a second plurality of at least 10,000 sequence reads, in electronic form, of an RNA sample from the test subject, the RNA sample comprising RNA molecules from the cancerous tissue of the subject;
(C) obtaining a third plurality of at least 10,000 sequence reads, in electronic form, of a second DNA sample from the test subject, the second DNA sample comprising DNA molecules from a non-cancerous tissue of the subject;
(D) evaluating the first plurality of at least 10,000 sequence reads, the second plurality of at least 10,000 sequence reads, and the third plurality of sequence reads using an ensemble model trained to distinguish between cancers with homologous recombination pathway deficiencies and cancers without homologous recombination pathway deficiencies, wherein the evaluating comprises:
determining, based on the first plurality of at least 10,000 sequence reads and the third plurality of at least 10,000 sequence reads, a first prediction for the homologous recombination pathway status of the cancerous tissue of the subject based on a measure of genome-wide loss of heterozygosity for the cancerous tissue of the subject; and
determining, based on the second plurality of at least 10,000 sequence reads, a second prediction for the homologous recombination pathway status of the cancerous tissue of the subject based on the expression levels of a plurality of genes in the cancerous tissue of the subject;
thereby obtaining the homologous recombination pathway status of the test subject.
26. A non-transitory computer-readable storage medium having stored thereon program code instructions that, when executed by a processor, cause the processor to perform a method for obtaining a homologous recombination pathway status of a cancer in a test subject, the method comprising:
(A) obtaining a first plurality of at least 10,000 sequence reads, in electronic form, of a DNA sample from the test subject, the DNA sample comprising DNA molecules from a cancerous tissue of the subject;
(B) obtaining a second plurality of at least 10,000 sequence reads, in electronic form, of an RNA sample from the test subject, the RNA sample comprising RNA molecules from the cancerous tissue of the subject;
(C) obtaining a third plurality of at least 10,000 sequence reads, in electronic form, of a second DNA sample from the test subject, the second DNA sample comprising DNA molecules from a non-cancerous tissue of the subject;
(D) evaluating the first plurality of at least 10,000 sequence reads, the second plurality of at least 10,000 sequence reads, and the third plurality of sequence reads using an ensemble model trained to distinguish between cancers with homologous recombination pathway deficiencies and cancers without homologous recombination pathway deficiencies, wherein the evaluating comprises:
determining, based on the first plurality of at least 10,000 sequence reads and the third plurality of at least 10,000 sequence reads, a first prediction for the homologous recombination pathway status of the cancerous tissue of the subject based on a measure of genome-wide loss of heterozygosity for the cancerous tissue of the subject; and
determining, based on the second plurality of at least 10,000 sequence reads, a second prediction for the homologous recombination pathway status of the cancerous tissue of the subject based on the expression levels of a plurality of genes in the cancerous tissue of the subject;
thereby obtaining the homologous recombination pathway status of the test subject.