Current research

Gene conversions are a form of meiotic recombination distinct from crossover events, where a short tract of DNA is copied over from one homologous chromosome to the other. Alleles change on the donor chromosome only at positions inside the gene conversion tract in which the individual is heterozygous. This makes it difficult to guess where each tract starts and ends.

We extended a maximum likelihood model to estimate the mean length of gene conversion tracts using gene conversion events detected from biobank-scale human sequencing data.

Recombinant SARS-CoV-2 variants can emerge when an individual is simultaneously infected with multiple variants of the virus. However, it can be challenging to identify recombinant SARS-CoV-2 sequences among thousands of samples. 

In this project, we are developing a hidden Markov model that can identify recombinant SARS-CoV-2 sequences from a large set of sequences, and further guess the parental lineages for each detected recombinant sequence.