My PhD Project

I started my PhD in March 2015, aiming to devise and implement methods to accelerate the analysis of population resequencing data. This is a summary of my work thus far, and for the remainder of my time as a PhD candidate.

Improved tools and pipelines for GBS analysis

Genotyping By Sequencing is a common, reasonably priced sequencing method to assay large numbers of samples at a subset of their genomes. I have created a series of small tools to aid the analysis of GBS data. In particular, Axe is one of the few sequence read demultiplexers which can demultiplex GBS reads with combinatorial index sequences of differing length, while tolerating mismatches.

Reference- and Alignment-free estimation of genetic distance

One important output of population re-sequencing experiments is the genetic distance between samples. Often, this is an initial result which guides further analysis, and in some cases this may even be the only desired output. kWIP, the k-mer Weighted Inner Product is a method to estimate genetic distance directly from sequencing data, without reliance on a reference genome. In our PLoS Comp. Biol. paper, we show that kWIP performs well even at modest coverage or low divergence.

links

social