#34 · Structural bio & protein design

Simulation and screening method choices are being framed as reportable design decisions

Across these items, the common thread is method design treated as an explicit element to report alongside results. An evolved, streamlined collective-variable strategy for enhanced-sampling simulations of class A GPCR activation free energies is tested on ADRB1 and framed as transferable and robust. An automated RAMD plus infrequent metadynamics protocol predicts absolute drug residence time, benchmarked on 29 complexes across five targets. COFFEE-PRESC, a fragment-pair pre-screening method, retrieves compounds by pairwise positional relationships of representative fragments before docking and reports a 32-fold speedup. The items do not share a common head-to-head benchmark, so the shared element is the reportability of choices such as collective variables, protocol components, and pre-screening design.