Cross-taxon Coarse-Grained IDP Simulations Enable Architecture-Independent Generalization

Abstract

Intrinsically disordered proteins and regions are found across all kingdoms of life, yet the computational characterisation of their conformational ensembles has remained almost entirely confined to the human proteome. Whether the force fields used to generate them remain accurate for taxonomically distant organisms, and whether the sequence-ensemble relationships they reveal transfer across taxa well enough to improve prediction on phylogenetically held-out organisms, are open questions. Here we introduce BENDER, a dataset of 11,533 IDP sequences spanning 13 taxonomic groups, each simulated under CALVADOS-2 molecular dynamics and annotated with ensemble-level geometric and novel contact-network properties, together with per-sequence pi–pi and cation–pi contact frequencies linked to phase-separation propensity. We show that CALVADOS-2 ensembles agree strongly with an orthogonal structural reference across the full dataset, with both held-out taxa performing above the dataset median, and that direct comparison against a second independently parameterized force field reveals no systematic scaling-exponent bias. We find that cross-taxon training data improves out-of-distribution ensemble prediction in two independent architectures despite training on one-third the data. Positive degree assortativity is conserved across all taxonomic groups, suggesting that hub topology in disordered protein contact networks is a conserved physical feature of sequence-encoded disorder rather than an evolutionary contingency.