Developing Darwinian NGS Tools: Advanced bioinformatics approaches for mapping next-generation sequencing data across phylogenetic distances

Co-supervisor 2 - Prof Leszek Gąsieniec, Department of Computer Science
***********************

Advances in next-generation sequencing (NGS) are having enormous impact in all areas of the Life Sciences, through applications in genome sequencing and assembly, discovery of variants and measuring gene expression. The latest instruments are able to generate 100s of Gigabases in a single run. The instruments that produce the most data typically produce “short reads” (< 100bp in length). Data analysis involves mapping NGS reads against a reference sequence, for which a variety of software packages exist. Mapping these huge collections of data against a reference genome is computationally demanding and recent approaches employ novel indexing or data compression techniques. The downside is that they are based on an initial exact (or near exact) match from the read to the reference. Allowing for a larger number of changes, their performance downgrades rapidly. This is a major problem in sequencing new genomes where mapping software cannot be employed.

We are developing techniques that enable short reads to be mapped across larger evolutionary distances than previously possible. We are employing a technique that makes use of underlying evolutionary conserved signals in DNA data to allow existing mappers (e.g. BWA or Bowtie) to be used for species that are not closely related. In this project, we would like to develop and implement new mapping software that makes use of signals in the data to map short read data across wider evolutionary ranges than previously possible, without degradation in performance. We can expect the outputs of the project to be high-profile, since NGS workflows are becoming ubiquitous across the Life Sciences, and sequence mapping is a core function.

We expect this project could suit someone from a background in Computer Science, biology or mathematics, with an interest in bioinformatics.

Training:
You will take modules from the Masters in Post-genomic science (on which Andy Jones is director of studies) to learn genomics, bioinformatics, statistics and Perl programming, as appropriate. Additionally, the research group of Prof Gąsieniec will provide respective PhD training in object-oriented (Java) programming, algorithmic and optimisation methods. Members of this group will also contribute to research explorations and assist in software development that will form an integral part of this project.

You will also gain exposure to cutting edge facility in the Centre for Genomic Research (CGR), which is led by Prof Hall. The CGR acts as an MRC and NERC sequencing hub, and contains state-of-the-art next generation sequencers, including Roche 454, SOLiD 5500, PGM Ion Torrent machine and Illimuna MiSeq, GAIIx and HiSeq 2500 sequencers, as well as a bioinformatics team for data analysis.

We have regular journal clubs, group meetings and training sessions to learn new skills, and thus is an excellent training environment for new students joining.