Proteomics data analysis, software and data sharing
Our group works on data standards, new statistical approaches and software for proteomics, for example used for large-scale re-processing of data. The group is developing approaches for understanding the profile, evolution and function of post-translational modification (PTMs) across animals, plants and eukaryotic pathogens.

I have contributed over many years leadership to the Proteomics Standards Initiative and have been involved with the development of data standards for proteomics, such as mzIdentML and for metabolomics such as mzTab-M.

Analysis of plant genomes and proteomes
My group is increasingly analysing and integrating multi-omics data on crops, particularly rice. Our first work in this area was via a BBSRC/Newton funded grant working with collaborators at BGI China, in which we developed and used a proteogenomics pipeline to provide protein-level evidence for ~8000 rice genes, discovery of over 100 novel genes not annotated in the canonical gene models, and suggested gene model revisions for ~700 genes (e.g. new splice junctions) through very large scale analysis of RNA Seq and mass spectrometry data in the public domain. We have also started to work on exploring the evolutionary conservartion of post-translational modifications (PTMs) in flowering plants, and potential discovery of crosstalk between different PTM sites.

In new work, we are interested in improving annotation of the rice genome, and particularly focussing on analysis of gene families involved in regulation (such as transcription factors) or signalling such as kinases.

Immunogenetics, adverse drug reactions and machine learning applied to biomedical data
Our group maintains and develops the popular Allele Frequency Net Database storing data on allele, haplotype and gene frequencies for immune-related genes (HLA, KIR, Cytokines) in healthy human populations, covering over 10M individuals. We have developed new portals for storing disease associations for KIR genes, and for capturing known associations between HLA alleles and adverse drug reactions (ADRs). We have also used molecular docking to understand associations between HLA protein structure and ADR mechanisms.

The group is also applying machine learning (ML) techniques for the analysis of clinical data, such as single antigen bead (SAB) technology for profiling patient antibodies prior to kidney transplants, with the Transplant Immunology lab at the Royal Liverpool hospital. We are also developing ML approaches for analysing flow cytometry data, used in blood cancer diagnosis with the Haemato Oncology Diagnostic Service (HODS).