# Structure-guided molecular design with contrastive 3D protein-ligand learning Source: [Structure-guided molecular design with contrastive 3D protein-ligand learning]() Collection index: [llms.txt]() Data updated: 2026-04-23T05:00:12.895Z Authors: Carles Navarro, Philipp Tholke, Gianni de Fabritiis Published: 2026-04-21T15:13:41.000Z ## Abstract Structure-based drug discovery faces the dual challenge of accurately capturing 3D protein-ligand interactions while navigating ultra-large chemical spaces to identify synthetically accessible candidates. In this work, we present a unified framework that addresses these challenges by combining contrastive 3D structure encoding with autoregressive molecular generation conditioned on commercial compound spaces. First, we introduce an SE(3)-equivariant transformer that encodes ligand and pocket structures into a shared embedding space via contrastive learning, achieving competitive results in zero-shot virtual screening. Second, we integrate these embeddings into a multimodal Chemical Language Model (MCLM). The model generates target-specific molecules conditioned on either pocket or ligand structures, with a learned dataset token that steers the output toward targeted chemical spaces, yielding candidates with favorable predicted binding properties across diverse targets. ## Recorded benchmark results A research result applies to its stated dataset, split, metric and evaluation configuration. Compare like-for-like settings, preserve the metric's unit and direction, and distinguish paper-reported results from independent reproduction. The best result recorded here is not a claim of exhaustive, current global coverage. - [Structure-Conditioned Molecular Design on LIT-PCBA 15 targets](): Structure-Conditioned Molecular Design; LIT-PCBA 15 targets; Affinity Probability: 68. - [Virtual Screening on LIT-PCBA (test)](): Virtual Screening; LIT-PCBA (test); AUROC: 53.76. ## Original sources and implementation [Original paper]() [PDF]()