Imagine you're a parent of twins in the same grade-school class, and that their teacher issued a single report card for both, with the marks an even mix of As and Bs. Reading through it, you'd have to wonder: Are both children earning As and Bs? Is one an A-student and the other a B-student?
That, the Institute for Systems Biology's Jared Roach says, is "the way most genome analysis is being done right now — we get genotypes, which are sort of a mixed -sequence of the genome, and in some cases it's hard to interpret what the meaning is." Being able to genetically phase individuals, or generate haplotypes for them, could help researchers pinpoint disease-causing mutations within pedigrees, he adds, and perhaps even enhance their understanding of fundamental biology.
In a September American Journal of Human Genetics paper, Roach and his colleagues present full-chromosome haplotypes for two nuclear families with four children each, which they generated using whole-genome data on Haploscribe — an algorithm the researchers designed for phasing by genetic analysis.
"In some sense, what we're doing is taking basic genetics — concepts like linkage and knowledge of how recombination works — and taking advantage of whole-genome sequence and saying, 'Look what we can do with really good data and [the] application of basic algorithms that have been on the shelf, getting dusty for 20 or 30 years,'" Roach says. Having whole-genome sequences for both parents and more than one child in a pedigree enables researchers to ascertain "exceptionally precise determinations of where recombinations have occurred," he adds.
Using Haploscribe, "we're able to do haplotyping with just two generations of data," Roach says. He and his colleagues used grandparental sequence data for both pedigrees only after they'd phased the family members to verify the accuracy of the haplotypes they generated.
Roach adds that computational analyses on familial genomes, such as the one his team reports in its paper, could also propel advances in molecular haplotyping.
"Being able to use family haplotypes as gold standards will allow other technologies to improve," he says. Molecular haplotyping could be of clinical importance in the future, Roach says, particularly as not every patient has access to family history or familial genetic data.
Family sequences and haplotypes could be important for personalized medicine. In the meantime, Roach says that "most genome analyses, including those employing our haplotyping algorithm, are only as good as the reference sequence" and its current blind spots. Forthcoming generations of the reference sequence "will vastly empower this and all other algorithms working on haplotypes," he adds.