Biology subjects

Karelina, M.

Publications and source records attributed to Karelina, M..

3 recordsLinked to original sources

bioRxiv2025

MedSAGE: Bridging Generative AI and Medicinal Chemistry for Structure-Based Design of Small Molecule Drugs

While generative AI is transforming the de novo design of proteins, its effectiveness for structure-based design of small molecules remains limited. Current methods, including diffusion models, often produce small molecules with difficult-to-synthesize structures, poor medicinal chemistry properties, and limited target selectivity. To address these limitations, we introduce MedSAGE, a novel generative AI framework that adapts diffusion models specifically for de novo small-molecule design. Rather than using atoms or strings, we develop a novel representation for generating molecules using fragments relevant for medicinal chemistry. Chemical and geometric information characterizing these fragments is embedded in a smooth and interpretable latent space. We also develop an algorithm to optimize the connectivity between generated fragments while preserving chemical validity and synthesizability. In a benchmark of multiple methods across 25 therapeutically relevant protein targets, MedSAGE achieved state-of-the-art performance, producing synthesizable, drug-like molecules with predicted affinity and selectivity closely matching known drugs and drug candidates. Compared to large-scale virtual screening, MedSAGE produced molecules with high predicted affinity over 100 times more efficiently. Our results demonstrate that MedSAGE is already practically useful and paves the way for next-generation tools in structure-guided drug design.

biophysics↗

bioRxiv2023

GPR161 structure uncovers the redundant role of sterol-regulated ciliary cAMP signaling in the Hedgehog pathway

The orphan G protein-coupled receptor (GPCR) GPR161 is enriched in primary cilia, where it plays a central role in suppressing Hedgehog signaling1. GPR161 mutations lead to developmental defects and cancers2,3,4. The fundamental basis of how GPR161 is activated, including potential endogenous activators and pathway-relevant signal transducers, remains unclear. To elucidate GPR161 function, we determined a cryogenic-electron microscopy structure of active GPR161 bound to the heterotrimeric G protein complex Gs. This structure revealed an extracellular loop 2 that occupies the canonical GPCR orthosteric ligand pocket. Furthermore, we identify a sterol that binds to a conserved extrahelical site adjacent to transmembrane helices 6 and 7 and stabilizes a GPR161 conformation required for Gs coupling. Mutations that prevent sterol binding to GPR161 suppress cAMP pathway activation. Surprisingly, these mutants retain the ability to suppress GLI2 transcription factor accumulation in cilia, a key function of ciliary GPR161 in Hedgehog pathway suppression. By contrast, a protein kinase A-binding site in the GPR161 C-terminus is critical in suppressing GLI2 ciliary accumulation. Our work highlights how unique structural features of GPR161 interface with the Hedgehog pathway and sets a foundation to understand the broader role of GPR161 function in other signaling pathways.

biochemistry↗

bioRxiv2023

How accurately can one predict drug binding modes using AlphaFold models?

Computational prediction of protein structure has been pursued intensely for decades, motivated largely by the goal of using structural models for drug discovery. Recently developed machine learning methods such as AlphaFold 2 (AF2) have dramatically improved protein structure prediction, with reported accuracy approaching that of experimentally determined structures. To what extent do these advances translate to an ability to predict more accurately how drugs and drug candidates bind to their target proteins? Here we carefully examine the utility of AF2 protein structure models for predicting binding poses of drug-like molecules at the largest class of drug targets, the G protein-coupled receptors. We find that AF2 models capture binding pocket structures much more accurately than traditional homology models, with errors nearly as small as differences between structures of the same protein determined experimentally with different ligands bound. Strikingly, however, the accuracy of ligand binding poses predicted by computational docking to AF2 models is not significantly higher than when docking to traditional homology models and is much lower than when docking to structures determined experimentally without these ligands bound. These results have important implications for all those who might use predicted protein structures for drug discovery.

biophysics↗