Search bioRxiv⌕ Search

Biology subjects

Sanchez Utges, J.

Publications and source records attributed to Sanchez Utges, J..

2 recordsLinked to original sources

Interrogating contrastive learning embeddings for structure-based virtual screening: a case study on DrugCLIP

Virtual screening has become central to early-stage drug discovery, and structure-based approaches have recently been reframed as a retrieval problem through contrastive learning methods such as DrugCLIP, which project protein pockets and ligands into a shared embedding space. However, what these abstract representations exactly encode, and how they relate to conventional notions of structural and chemical similarity, remains unclear. Here we present a systematic dissection of DrugCLIP's latent space. We show that its pocket embeddings, despite not being explicitly trained for the task, set a new state of the art in pocket similarity search while running over 100 times faster than existing structural descriptors, and that this embedding space is structurally coherent and robust to conformational variation. Ligand embeddings, by contrast, encode a pocket-aware notion of chemical similarity that only partially mirrors fingerprint-based measures. Using a rigorous de-leakage benchmark, we further show that DrugCLIP generalises to unseen proteins and chemistries rather than memorising training data, recovering the correct bound ligand within the top 1% of 50,000 candidates for 55-75% of novel targets. Performance nonetheless declines under increasingly realistic screening conditions, a drop attributable to sidechain reorientation across apo, holo and AlphaFold-derived structures, and to residue mismatch when using predicted pockets. These findings clarify the practical boundaries of DrugCLIP's applicability, identify pocket prediction accuracy as a key factor for improving performance, and offer a transferable framework for interpreting the latent spaces of related contrastive pocket-ligand encoders. Together, these results support the improvement of existing methods and the development of a new generation of contrastive screening approaches.

bioinformatics↗

Mutations within the predicted fragment-binding region of FAM83G/SACK1G abolish its interaction with the Ser/Thr kinase CK1α

SACK1G (aka FAM83G, PAWS1) plays a central role in activating canonical WNT signalling via interaction with the Ser/Thr kinase CK1. This loss of CK1 binding and WNT signalling underlies the pathogenesis of Palmoplantar Keratoderma (PPK) caused by several reported mutations in the SACK1G gene. We modelled the scaffold anchor of CK1 (SACK1) domain of SACK1G and used fragment-bound structures of the SACK1B (FAM83B) dimer to guide our analysis. This allowed us to computationally predict several key residues near the fragment-binding site in SACK1G that may be important for its function. We mutated these residues, introduced them into SACK1G-/- DLD-1 colorectal cancer cells and investigated their ability to bind endogenous CK1. We uncovered two SACK1G mutations, namely Y204A and I206A, that abolish interaction with CK1 similarly to the PPK pathogenic mutant A34E. Consistent with this loss of SACK1G-CK1 interaction, the molecular glue degrader of CK1, DEG-77, fails to co-degrade the Y204A and I206A mutants while it still co-degrades native SACK1G. Our findings demonstrate the utility of our computational methods to uncover functional residues on proteins based on fragment-binding sites.

cell biology↗