MRC DTP in Precision Medicine

Precision serology – developing new statistical methods for the analysis of PhIP-Seq (Phage ImmunoPrecipitation Sequencing) data

Supervisors

Prof Emma Thomson, School of Infection and Immunity, University of Glasgow
Dr Richard Orton, School of Infection and Immunity, University of Glasgow
Dr Craig Wilkie, School of Mathematics & Statistics, University of Glasgow

Summary

PhIP-Seq (Phage ImmunoPrecipitation Sequencing) offers the potential for ‘precision serology’ by combining phage display, immunoprecipitation, and high-throughput sequencing (HTS) to measure which antigens a person’s antibodies recognize at a massive scale. Instead of measuring one antigen at a time (like in ELISA), PhIP-Seq uses HTS to quantify which library peptides are pulled down by a patient’s antibodies. As hundreds of thousands of distinct peptides can be in the library, and hundreds of samples can be analysed on a single sequencing run, PhIP-Seq offers the potential to perform serological analyses at an unprecedented scale.

This project will develop novel bioinformatic and statistical methods to analyse PhIP-Seq data. Control data will be created to thoroughly evaluate key aspects of the PhIP-Seq methodology itself, such as replicability, cross reactivity, and epitope distance. Approaches will be developed capable of utilising observed counts of peptides with information on sequence composition, replicates (peptide and sample), sample type, peptide similarity, and viral phylogeny; methods include hierarchical Bayesian models and machine learning approaches utilising graph base models or transformer-based protein embeddings. The student will be trained in advanced bioinformatic and statistical techniques, within research groups actively generating PhIP-Seq data and with supervisors from clinical virology, bioinformatics, and statistics.