Computational modeling of loss of function based on allelic frequency
The disclosure relates to computer technology for precision diagnosis of various states of genetic material such as a gene sequenced from cell-free DNA in a sample. The state may include a somatic homozygous deletion, a somatic heterozygous deletion, a copy number variation, or other states. A computer system may generate competing probabilistic models that each output a probability that the genetic material is in a certain state. Each model may be trained on a training sample set to output a probability that the genetic material is in a respective state. In some embodiments, the computer system may use various probabilistic distributions to generate the models. For example, the computer system may use a beta-binomial distribution, a binomial distribution, a normal (also referred to as “Gaussian”) distribution, or other type of probabilistic modeling techniques.
1. A method for treating a subject having a cancer associated with homologous recombination deficiency (HRD), the method comprising:
(a) determining that the subject has a somatic homozygous deletion for a BRCA1 gene or BRCA2 gene in a sample of cell-free nucleic acid molecules by:
(i) obtaining the sample of cell-free nucleic acid molecules from the subject;
(ii) performing a diagnostic assay on the sample to detect the somatic homozygous deletion for the BRCA1 gene or BRCA2 gene, wherein the diagnostic assay comprises:
A) generating, via a first probabilistic distribution, a first model of allelic counts based on one or more germline single nucleotide polymorphism (SNP) positions associated with the BRCA1 gene or BRCA2 gene, the first model representing the somatic homozygous deletion;
B) generating, via a second probabilistic distribution, a second model of allelic counts in the sample based on the one or more germline SNP positions, the second model representing the somatic heterozygous deletion;
C) comparing a first output of the first model and a second output of the second model;
D) generating a prediction that the somatic homozygous deletion for the BRCA1 gene or BRCA2 gene exists in the sample based on the comparison; and
(b) administering a therapy comprising a poly ADP ribose polymerase (PARP) inhibitor to the subject to treat the cancer.
2. The method of claim 1 , wherein the first model represents a first probability that the sample includes the somatic homozygous deletion, and the second model represents a second probability that the sample includes the somatic heterozygous deletion.
3. The method of claim 1 , wherein to generate the first model, one or more parameters are determined for input to the first probabilistic distribution.
4. The method of claim 3 , wherein the first probabilistic distribution comprises a type of probabilistic distribution comprising one of: a beta-binomial distribution, a binomial distribution, or a normal distribution.
5. The method of claim 3 , wherein to generate the first model of allelic counts, a prevalence of heterozygosity of the one or more germline SNPs in a training set of samples is determined, for input to the first probabilistic distribution.
6. The method of claim 5 , wherein the training set of samples comprises a plurality of samples in which tumor is not detected (TND).
7. The method of claim 5 , wherein to generate the first model of allelic counts, a standard deviation of a minor allele frequency (MAF) associated with the one or more germline SNPs in the training set of samples is determined, for input to the first probabilistic distribution.
8. The method of claim 7 , wherein to generate the first model, a number of molecules in the sample that supports a mutant allele is determined, for input to the first probabilistic distribution.
9. The method of claim 8 , wherein to generate the first model, a total number of molecules in the sample is determined, for input to the first probabilistic distribution.
10. The method of claim 9 , wherein to generate the first model, a first likelihood of the allelic counts of the one or more germline SNP positions in the sample is calculated assuming somatic homozygous deletion based on molecule coverage associated with the somatic homozygous deletion.
11. The method of claim 10 , wherein to generate the second model, a second likelihood of the allelic counts of the one or more germline SNP positions in the sample is calculated assuming somatic heterozygous deletion based on molecule coverage associated with the somatic heterozygous deletion.
12. The method of claim 3 , wherein to generate the second model, a mean of tumor fraction is estimated from the sample, for input to the second probabilistic distribution for the second model.
13. The method of claim 12 , wherein the tumor fraction is estimated based on sequence coverage information.
14. The method of claim 12 , wherein to generate the second model, a standard deviation of tumor fraction estimated from the sample, for input to the second probabilistic distribution for the second model.
15. The method of claim 1 , further comprising: accessing a plurality of samples; identifying a set of samples from among the plurality of samples that include a germline deletion; and filtering out the set of samples from the plurality of samples; and identifying, from among the filtered plurality of samples, a presence of the somatic homozygous deletion or the somatic heterozygous deletion.
16. The method of claim 1 , wherein to compare the first output of the first model and the second output of the second model, a log likelihood function is computed based on the first output and the second output.
17. The method of claim 1 , wherein the PARP inhibitor comprises at least one of: OLAPARIB, TALAZOPARIB, RUCAPARIB, and NIRAPARIB.
18. The method of claim 17 , wherein the cancer comprises breast, ovarian, prostate, or pancreatic cancer.
19. The method of claim 1 , wherein the cancer comprises breast, ovarian, prostate, or pancreatic cancer.