System and method for the computational prediction of expression of single-gene phenotypes
In accordance with an embodiment of the invention, a system and method is provided for determining a probability of a progeny having one or more phenotypes Ph j each associated with a single gene Q j . A score sip may be assigned to each allele hip at a plurality of genetic loci (i) in a haploid genome profile H p of a parent (p). A plurality (Nj) of the alleles hkp (k=1, . . . , Nj) associated with the gene Q j may be identified. The scores sip may be mapped or indexed to gene-specific scores ŝj,kp associated with gene Q j for the plurality of (Nj) alleles hkp. A probability may be computed for altering the gene product from gene Q j in a progeny of the parent (p) to be a function of the gene-specific scores ŝj,kp.
1. A computer-implemented method for accounting for genetic recombination in determining likelihood of unconceived progenies of two potential parents having a phenotype, the computer-implemented method, comprising:
receiving, by a computer, a first potential parent's diploid DNA profile and a second potential parent's diploid DNA profile, the first potential parent's diploid DNA profile and the second potential parent's diploid DNA profile generated respectively by DNA sequencing of a first biological sample of the first potential parent and a second biological sample of the second potential parent;
generating, by the computer, a plurality of virtual progenies of the first potential parent and the second potential parent, generating the plurality of virtual progenies accounted for genetic recombination and comprising:
for each virtual progeny:
generating a first virtual haplopath from the first potential parent's diploid DNA profile;
generating a second virtual haplopath from the second potential parent's diploid DNA profile, wherein each of the first virtual haplopath and the second virtual haplopath representing a virtual gamete simulated with genetic recombination, the virtual gamete comprising a plurality of genetic loci, each genetic locus having a haplotype and generated by progressing locus-by-locus of the first potential parent's diploid DNA profile and the second potential parent's diploid DNA profile to select, at least partially randomly and also based on linkage probability generated from linkage disequilibrium information, one of two alleles of the first potential parent's diploid DNA profile and the second potential parent's diploid DNA profile at each genetic locus; and
combining the first virtual haplopath and the second virtual haplopath to generate the virtual progeny;
determining, for each virtual progeny of the plurality of the virtual progenies that account for genetic recombination, a probability of the virtual progeny having the phenotype associated with a gene, the determining comprising:
a. assigning a score s i p to each allele h i p at the plurality of genetic loci (i) in each of the first virtual haplopath and the second virtual haplopath;
b. identifying a plurality of the alleles associated with the gene Q j , the plurality of N j of the alleles h i p (k=1, . . . , N j ) identified from the plurality of genetic loci (i) in the first and second virtual haplopaths;
c. mapping the scores s i p to gene-specific scores s j,k p associated with gene Q j for the plurality of (N j ) alleles h k p ; and
d. computing the probability of the virtual progeny having the phenotype associated with the gene to be a function of the gene-specific scores s j,k p ;
combining the determined probability of each virtual progeny of the plurality of the virtual progenies to generate a distribution; and
outputting a result of the likelihood of unconceived progenies of two potential parents having the phenotype based on the distribution, the result accounted for genetic recombination.
2. The method of claim 1 , wherein the probability of having the phenotype is determined based on P j p =1−Π k=1 Nj [1− s j,k p ].
3. The method of claim 1 , wherein a probability of having a recessive phenotype is determined based on p j p1,p2 =(1−Π k=1 Nj [1− s j,k p1 ])(1−Π k=1 Nj [1− s j,k p2 ]).
4. The method of claim 1 , wherein a probability of having a dominant phenotype is determined based on p j p1,p2 =(1−Π k=1 Nj [1−s j,k p1 ])(1−Π k=1 Nj [1− s j,k p2 ]).
5. The method of claim 1 , wherein each of the scores s i p defines a likelihood that the allele h i p is an amino acid substitution at the locus (i) that will damage protein function.
6. The method of claim 1 , wherein each of the scores s i p defines a probability that for a change in one or more amino acids that the change will occur randomly based on natural selection.
7. The method of claim 1 , wherein the probability indicates a degree of expressivity of the phenotype in the virtual progeny.
8. The method of claim 1 further comprising comparing the probability to one or more thresholds to determine a category of expressivity of the phenotype.
9. The method of claim 1 , wherein the probability corresponds to a probability of protein damage in one or more gene products, each associated with a single gene Q j .
10. A computer-implemented method for accounting for genetic recombination in determining likelihood of unconceived progenies of two potential parents having a phenotype, the computer-implemented method, comprising:
receiving, by a computer, a first potential parent's diploid DNA profile and a second potential parent's diploid DNA profile, the first potential parent's diploid DNA profile and the second potential parent's diploid DNA profile generated respectively by DNA sequencing of a first biological sample of the first potential parent and a second biological sample of the second potential parent;
generating, by the computer, a plurality of virtual progenies of the first potential parent and the second potential parent, generating the plurality of virtual progenies accounted for genetic recombination and comprising:
for each virtual progeny:
generating a first virtual haplopath from the first potential parent's diploid DNA profile;
generating a first virtual haplopath from the first potential parent's diploid DNA profile;
generating a second virtual haplopath from the second potential parent's diploid DNA profile, wherein each of the first virtual haplopath and the second virtual haplopath representing a virtual gamete simulated with genetic recombination, the virtual gamete comprising a plurality of genetic loci, each genetic locus having a haplotype and generated by progressing locus-by-locus of the first potential parent's diploid DNA profile and the second potential parent's diploid DNA profile to select, at least partially randomly and also based on linkage probability generated from linkage disequilibrium information, one of two alleles of the first potential parent's diploid DNA profile and the second potential parent's diploid DNA profile at each genetic locus; and
combining the first virtual haplopath and the second virtual haplopath to generate the virtual progeny;
determining, for each virtual progeny of the plurality of the virtual progenies that account for genetic recombination, a probability of the virtual progeny having the phenotype associated with a gene, the determining comprising:
a. comparing genotypes of the virtual progeny to one or more databases of genotype-phenotype associations to determine a phenotype associated with database genotypes matching genotypes of the virtual progeny genome sampling G, wherein the phenotype is associated with a penetrance value;
b. generating, by a random number generator of the computer, a random number that is weighted by a bias factor that is biased to express or non-express with a non-equal probability according to a predefined bias ratio;
c. adjusting the penetrance value with a degree of expression of the phenotype in the virtual progeny based on the random number; and
d. determining the probability of the virtual progeny having the phenotype based on the penetrance value;
combining the determined probability of each virtual progeny of the plurality of the virtual progenies to generate a distribution; and
outputting a result of the likelihood of unconceived progenies of two potential parents having the phenotype based on the distribution, the result accounted for genetic recombination.
11. The method of claim 1 , wherein progressing locus-by-locus of the first potential parent's diploid DNA profile and the second potential parent's diploid DNA profile to select, at least partially randomly and also based on the linkage probability generated from the linkage disequilibrium information, one of two alleles of the first potential parent's diploid DNA profile and the second potential parent's diploid DNA profile at each genetic locus comprises using an iteration of a Monte Carlo simulation or a chain generated through a Markov process.
12. The method of claim 1 , wherein the first potential parent's diploid DNA profile is the first potential parent's genome profile and the second potential parent's diploid DNA profile is the second potential parent's genome profile.
13. The method of claim 10 , wherein progressing locus-by-locus of the first potential parent's diploid DNA profile and the second potential parent's diploid DNA profile to select, at least partially randomly and also based on the linkage probability generated from the linkage disequilibrium information, one of two alleles of the first potential parent's diploid DNA profile and the second potential parent's diploid DNA profile at each genetic locus comprises using an iteration of a Monte Carlo simulation or a chain generated through a Markov process.
14. The method of claim 10 , wherein the first potential parent's diploid DNA profile is the first potential parent's genome profile and the second potential parent's diploid DNA profile is the second potential parent's genome profile.
15. The method of claim 10 , wherein a probability of having a recessive phenotype is determined based on p j p1,p2 =(1−Π k=1 Nj [1− s j,k p1 ])(1−Π k=1 Nj [1− s j,k p2 ]).
16. The method of claim 10 , wherein a probability of having a dominant phenotype is determined based on p j p1,p2 =(1−Π k=1 Nj [1− s j,k p1 ])(1−Π k=1 Nj [1− s j,k p2 ]).
17. The method of claim 10 , wherein the probability indicates a degree of expressivity of the phenotype in the progeny.
18. The method of claim 10 further comprising comparing the probability to one or more thresholds to determine a category of expressivity of the phenotype.
19. The method of claim 10 , wherein the probability corresponds to a probability of protein damage in one or more gene products, each associated with a single gene.
20. The method of claim 1 , wherein the phenotype is one of the following: Thalassemia, cystic fibrosis, anemia, or Mucopolysaccharidosis.
21. The method of claim 1 , wherein each of the first virtual haplopath and the second virtual haplopath includes at least 10,000 base pairs.
22. The method of claim 1 , wherein determining, for each virtual progeny of the plurality of the virtual progenies, the probability of the virtual progeny having the phenotype associated with a gene, further comprising:
generating, by a random number generator of the computer, a random number that is weighted by a bias factor that is biased to express or non-express with a non-equal probability according to a predefined bias ratio; and
adjusting the probability of the virtual progeny having the phenotype associated with the gene based on the random number.
23. The method of claim 10 , wherein the phenotype is one of the following: Thalassemia, cystic fibrosis, anemia, or Mucopolysaccharidosis.
24. The method of claim 10 , wherein each of the first virtual haplopath and the second virtual haplopath includes at least 10,000 base pairs.