The lab headed by Dr. Yongtao Guan is a computational lab with primary research interests in developing statistical and computational methods to address problems arising from genetics and genomics. Current projects include:
- Local ancestry inference
Motivated by approximating coalescent with recombination, we developed a two-layer clustering model that can model two scales of LD. An important application is to infer local ancestry of admixed individuals. Compared with other methods, our method 1) directly works with diplotypes and requires no recombination map; 2) cleanly handles missing data; 3) has a higher resolution --- can detects ancestry track length of a few tenths of a centimorgan; and 4) can handle multiple source populations.
- Haplotype association
We can infer local haplotype sharing (LHS) between a pair of individuals at an arbitray marker. We may then link the LHS with phenotypes to perform association mapping. Our method reinvent haplotype association mapping to provide several benefits --- no phasing requirement, no sliding-window requirement, an ability to work directly with next-generation sequencing data, and enhanced interpretability of association findings.
- De novo assembly
Identifying difficulties associated with the de Bruijn graph based approach for de novo assembly, we are currently developing algorithms and software packages to perform de novo assembly using Monte Carlo approach. We expect that the longer contigs produce better alignment-based variants call.
- Directed acyclic graphs
A directed acylic graph (DAG) specifies a joint distribution on all nodes in the graph. The joint distribution itself is of interest in many genetic applications. In addition, the joint distribution can be latent in some applications and need to be integrated out. Current order-graph based sampling approach cannot handle large number of nodes. We are developing new methods that can sample DAGs with very large number of nodes.
We have two USDA and NIH-funded positions open for motivated individuals with computational skills and genetics background. A position can be filled by a student, a programmer, a statistician, or a postdoc.
Students from genetics and SCBCM are strongly encouraged to inquire about potential rotation projects.