Building the Structural Evolutionary Toolkit of Tomorrow
Protein sequences lie. At the twilight zone of sequence identity, where homology detection fails and alignments collapse, evolution has quietly preserved something far more fundamental: the fold. In the post-AlphaFold era, where structural data is no longer a bottleneck, we now have an unprecedented opportunity to ask evolutionary questions at the resolution that actually matters. This talk argues that structure is not merely a complement to sequence, it is the primary currency of molecular evolution.
I will show how structural change maps onto functional divergence in ways sequence cannot capture, and how comparing protein architectures across the tree of life reveals deep evolutionary relationships invisible to traditional approaches. From there, I pivot to the therapeutic frontier: in an age of rapidly mutating pathogens, drug and vaccine design anchored to sequence is a losing game. Structure, by contrast, is conserved and understanding protein-pathogen interactions at the structural level opens new avenues for designing interventions that are robust to evolutionary escape.
Finally, I highlight the emerging power of structural alphabets, compact, discrete encodings of three-dimensional conformation, as a unifying language for structural evolution, enabling scalable comparison, phylogenetic inference, and machine learning across vast structural space.
Together, these threads point toward a new toolkit, one built not on letters, but on shape.
Bio
Ashar Malik is a computational structural biologist (postdoc) at the University of Queensland whose primary obsession is understanding molecular evolution through the lens of protein structure — and then telling anyone who’ll listen about it. When not building phylogenetic toolkits or wrangling protein folds, he can be found making the case for science communication or chasing the perfect shot through a forest somewhere near Brisbane.
