This webinar is by the ELIXIR 3D-BioInfo Community chaired by Prof. Shoshana Wodak(Group Leader at VIB Structural Biology Research Center) |
 |
Is this yet another multiverse talk?! Exploring murky regions of the protein multiverse with ancestral fragments and deep generative models
Dr. Eli Draizen
(Postdoctoral Scholar, UC San Francisco)
We are finally in the era of the protein multiverse! The ‘protein universe’ is commonly used to describe the collection of all possible proteins. ‘Possible’ refers to all modern, full-sized proteins found in public databases and could even include novel proteins designed with deep generative models that explore the ‘dark’ regions of the learned protein space.
However, modern proteins did not arise abruptly, as singular events, but rather over the course of at least 3.5 billion years of evolution. The molecular evolutionary processes that yielded their intricate 3D structures involve duplication, recombination and mutation of genetic elements, corresponding to short peptide fragments. This process can be viewed as corresponding to evolutionarily time-resolved protein universes, or protein multiverse.
Identifying ancestral fragments is crucial to deciphering the interrelationships amongst proteins, as well as how evolution acts upon protein sequences, structures and functions. Traditionally, common fragments have been found using alignment approaches, which becomes challenging when proteins have undergone extensive permutations—allowing for architecture similarity despite topological variability, a phenomenon we term the Urfold. We have created a framework to identify compact, potentially-discontinuous peptide fragments by combining deep generative models of protein superfamilies with explainable AI to identify relevant atoms.
Our approach recapitulates known relationships (established via manual analyses) amongst the evolutionarily ancient small β-barrels and amongst P-loop–containing proteins. We are now applying our approach to every CATH superfamily, including CATH-AlphaFold2 predicted domain structures. Because of the generality of our model’s approach, we anticipate that it can enable the discovery of new ancestral peptides. Alternative views of the protein universe, aka protein multiverse—such as the Urfold—offer new ways to explore exceedingly remote protein relationships, beyond traditional hierarchical classification systems, and could allow for finer grained functional annotations.
SWISH-X, an expanded approach to detect cryptic pockets in proteins and at protein-protein interfaces
Alberto Borsatto
(PhD Student - Gervasio Lab, University of Geneva)
Protein-protein interactions mediate most molecular processes in the cell, offering a significant opportunity to expand the set of known druggable targets. Unfortunately, targeting these interactions can be challenging due to their typically flat and featureless interaction surfaces, which often change as the complex forms. Such surface changes may reveal hidden (cryptic) druggable pockets.
Here, we analyse a set of well-characterised protein-protein interactions harbouring cryptic pockets and investigate the predictive power of current computational methods. Based on our observations, we develop a new computational strategy, SWISH-X (SWISH Expanded), which combines the established cryptic pocket identification capabilities of SWISH with the rapid temperature range exploration of OPES MultiThermal. SWISH-X is able to reliably identify cryptic pockets at protein-protein interfaces while retaining its predictive power for revealing cryptic pockets in isolated proteins, such as TEM-1 β-lactamase
You can find previous webinars from the 3D-BioInfo Community on the Community webinars page.