Analysis of the Human Genome

We are interested in the structure of the human genome.  For example, we have identified long regions of shared heterozygosity that we call "compressions" of the reference genome and could be caused by reference assembly errors or polymorphisms.  In addition, we are working on analyses of mutation rate in the human genome and identifying hotspots of recombination in families.


Genomics of Human Populations

We are interested in examining the diversity of the human genome over the entirety of the human population. We utilize public genomes, such as from the Thousand Genomes Project, a carefully curated selection of variants from our private genomes, and compilations of human variation to generate resources that allow us to examine this in further detail. Please see our Kaviar: Known Variants paper for more information.


Genomics of Human Pedigrees

We are studying ways of utilizing pedigree information in genomic analysis, such as using family information to correct errors in whole genome sequencing.  We have created a method that will phase chromosomes across a nuclear family (Haploscribe) and are working to extend this over the entire pedigree.  We are working on several algorithms that compare two or more human genomes and provide an estimation of how related the individuals are.  These methods work well for second through fifth generation relationships with no cryptic relatedness and can also estimate up to ninth generation relatedness at a lower confidence.


Study of Human Diseases

We are currently studying around 10 different heritable disorders, including Huntington's Disease and Fanconi Anemia. Our collaborators have recruited excellent candidate families for these diseases and we have generated whole-genome sequences for individuals from each of these pedigrees. We are actively working on identifying candidate mutations for these diseases to test with our collaborators.