Structure prediction and design of intrinsically disordered proteome [IRB Research Nodes Seminar]

Intrinsically disordered proteins and regions (collectively IDRs) are pervasive across proteomes in all kingdoms of life, help shape biological functions and are involved in numerous diseases. IDRs populate a diverse set of transiently formed structures yet defy commonly held sequence-structure-function relationships. In my talk I will describe how we can use molecular simulations with coarse-grained models and machine learning to study the relationship between sequence, conformational properties, and functions of IDRs.

I will first describe how we have used experimental data to learn a coarse-grained molecular energy function—that we call CALVADOS—to predict conformational properties of IDPs, multidomain proteins and RNA. I will describe the Bayesian formalism we developed to parameterize CALVADOS by targeting experimental data, and how this model enables us to study interactions within and between IDRs in biomolecular condensates.

I will then describe how we have generated conformational ensembles of all intrinsically disordered regions of the human proteome and used these to provide insight into sequence-ensemble relationships and evolutionary conservation of IDRs. I will also show how we can use the information encoded in CALVADOS to design disordered proteins with desired conformational properties and propensity to undergo phase separation. Finally, I will describe how we can use CALVADOS together with an active learning protocol to learn a quantitative model for the sequence rules governing the formation of biomolecular condensates.