84 Bioinformatics in the noughties with Mark Pallen

Mark Pallen explains how exciting it was to be in microbial bioinformatics around the turn of the millennium, as we gained genomes for the first time from model organisms and fearsome pathogens. He recounts working with his hero David Relman on the genome sequencing of the strange slow-growing organism called Tropheryma whipplei in competition with a French team. Mark moved to Belfast in late 1999 collaborating with another Englishman working on the island of Ireland, Tim Foster in Dublin. Pallen describes the addictive exhilaration of using PSI-BLAST to find new sortases and sortase substrates across a range of new genomes—for him this was the bioinformatics equivalent of crack cocaine. He quotes the philosopher Alfred North Whitehead in saying that the goal of every scientist is to seek simplicity but distrust it. What Mark found was that in most organisms sortases were behaving quite differently from the rather simple scenario seen in Staphylococcus aureus. He made similar observations on the WXG100 proteins and type VII secretion, which he found in many new contexts quite different from the original context of ESAT-6 as an antigen in Mycobacterium tuberculosis. Mark makes clear that we still don't really fully understand the role of ESAT-6 twenty years on. The focus of Pallen's work then shifted to E. coli, where he described vestigial gene clusters for non-functional type III secretion systems in this model organism. He came to realize that E. coli K-12 was not handed to microbiologist by God as a model organism but was just another strain of E. coli and nothing special. Many of the earliest genomes to be sequenced came from worn-out lab strains. To counter this problem, Gordon Dougan at the Wellcome Trust Sanger Institute moved the focus to genome-sequencing freshly isolated minimally passaged isolates. With Brendan Wren, Pallen wrote a review article for nature, emphasizing the importance of adopting an eco-evo perspective when trying to interpret bacterial genomes. Around that time, Scott Beatson joined Pallen's group. Mark managed to persuade Scott to work on type III secretion in E. coli rather than Pseudomonas aeruginosa. The result was the discovery of dozens of new type III secretion effectors, tying together bioinformatics and lab work to culminate in a PNAS paper. References - https://microbinfie.github.io/2022/06/09/bioinformatics-in-the-noughties.html