HTS-Oracle X: AI-Guided Prospective Discovery of Small Molecule Immune Checkpoint Binders

Targeting immune checkpoint protein-protein interactions (PPIs) using small molecules remains limited by the shallow, featureless binding surfaces of co-stimulatory and co-inhibitory receptors and the characteristically low hit rates of conventional high-throughput screening against these interfaces. Here we report HTS-Oracle X, a multimodal deep learning platform that integrates bidirectional cross-attention fusion of ChemBERTa SMILES embeddings with extended RDKit descriptors, trains on continuous biophysical binding signals rather than binary labels, and employs Monte Carlo Dropout uncertainty quantification for uncertainty-adjusted compound selection. Trained on 45,760 Dianthus TRIC-screened compounds per target under scaffold-aware cross-validation, HTS-Oracle X was applied prospectively to a 100,160-compound Enamine library against CD28, TIM-3, and VISTA. From 150 model-selected compounds, 45 dose-response confirmed binders were identified (30.0% overall hit rate), yielding enrichment factors of 234-408x over experimentally established random prospective baselines and 16 sub-micromolar hits. The top hits, HX-CD28-1 (KD = 233 nM), HX-TIM3-1 (KD = 249 nM), and HX-VISTA-1 (KD = 345 nM), demonstrated on-target functional activity in immune cell and tumor co-culture assays. HTSOracle X represents a scalable AI-guided framework for small molecule discovery against non-enzymatic immune checkpoint targets. Table of Contents artwork O_FIG O_LINKSMALLFIG WIDTH=200 HEIGHT=115 SRC="FIGDIR/small/732853v1_ufig1.gif" ALT="Figure 1"> View larger version (47K): org.highwire.dtl.DTLVardef@187fac7org.highwire.dtl.DTLVardef@6150d7org.highwire.dtl.DTLVardef@10266f0org.highwire.dtl.DTLVardef@b436e1_HPS_FORMAT_FIGEXP M_FIG C_FIG