OPIG recently celebrated its 20th year; and on 10 January 2023 I gave a talk just a day before the 10th anniversary of BLOPIG’s first blog post. It’s worth reflecting on what’s stayed the same and what’s changed since then.

Research focus has shifted from the areas of membrane protein prediction and protein-protein networks, and much more towards immunoinformatics and small molecule drug discovery, especially fragment-based. Emphasis on protein folding has shifted towards coarse grained modelling, and away from protein structure prediction. Another new strand of research being pursued in OPIG is around statistical methods for mass spec.

There has been a dramatic shift from conventional computing—where we devise algorithms that encode what we know—towards using data and algorithms to figure out what we don’t know from patterns in data. Applications of machine learning and artificial intelligence dominate our intellectual activities these days—and the impressive local GP-GPU compute that OPIG continues to grow is another significant change from a decade ago. Speaking of alternative computational paradigms, we have also explored the potential of quantum computing in computational biology.

At the start of 2020, the WHO declared a global pandemic, and OPIG immediately shifted our focus: to trying to understand the molecular machines of SARS-CoV-2, in particular main protease and the antibodies that bind to the virus that causes COVID-19. OPIG’s leader, Prof. Charlotte M. Deane was recognized with an M.B.E. for the key role she played while seconded to UKRI.

The introduction of the transformer in 2017 has catalyzed some incredible developments in AI, not least large language models. When I spoke about LLMs and their intersection with scientific research, it prompted lots of discussion, and raised lots of questions including ethical and legal. It’s worth thinking about where our research will be in another ten year’s time…