Drug discovery relies on cellular assays to determine whether a candidate produces a desired response, reveals its mechanism, or causes toxicity, but only a small fraction of compound-dose-time-context combinations can be measured experimentally. Recent perturbation atlases, single-cell and imaging technologies, and intervention-conditioned AI models now make prediction of unmeasured cellular responses a testable objective. AI virtual cells could therefore complement conventional discovery by prioritizing compounds, contexts, and follow-up experiments rather than replacing laboratory assays. Here, we review the resources and modeling advances supporting this capability, assess their value for mechanism-of-action analysis, efficacy, safety, resistance, and combination studies, and define evidence requirements for pharmacological use. Data coverage makes oncology, including immuno-oncology, plausible early proving grounds, with safety assessment as a crosscutting use case; broader deployment requires perturbation-specific, mechanistic, and decision-level validation.
Shi-Hang Wang, Yang Zhang, Dong Wang et al.· TIPS - Trends in Pharmacolog...· 0 citations
Structure-sensitive properties (SSPs), including activity cliffs and chirality-dependent properties, challenge molecular machine learning because small structural perturbations can cause abrupt property changes and invalidate smooth structure–property assumptions. Here, we present CAMF (Chirality- and Activity-cliff-aware Multimodal Framework), a task-adaptive framework that models SSPs through selective integration of complementary molecular evidence. To systematically evaluate this problem, we construct SSPBench, a benchmark spanning 77 conventional ADMET and physicochemical tasks together with activity-cliff and chirality-sensitive benchmarks. CAMF integrates molecular embeddings and expert-defined descriptors using random-forest-based feature selection and adaptive fusion, enabling property-specific prioritization of informative signals while reducing multimodal redundancy. Across ten baselines, CAMF achieves the best overall performance on SSP tasks, improving mean R2 by up to 29.5% on activity-cliff datasets and reducing MAE by up to 23.3% on 90 364 chiral molecules with TD-DFT-computed optical rotatory strengths. Ablation analyses show that these gains arise from task-adaptive multimodal integration rather than naive feature concatenation. More broadly, our results reveal that modality relevance is strongly task-dependent, with descriptors and 3D geometry becoming especially important in non-smooth property regimes. Case studies further support the interpretability and practical utility of CAMF in identifying activity-associated substructures and clinically relevant toxicity liabilities.
Shaolong Lin, Si-Long Zhai, Shi-Hang Wang et al.· Chemical Science· 0 citations
TPS-Flow is presented, a physics-guided flow-based generative framework for end point-conditioned conformational path sampling between predefined protein states (not equilibrium ensembles), thereby bridging atomistic simulation and deep generative modeling of protein transition paths.
Kai Xu, Likun Zhao, Yanan Tian et al.· Journal of Chemical Informat...· 0 citations
CoBind is presented, a multitask deep learning framework that jointly predicts RNA–compound interactions and nucleotide-level binding-site probabilities within a unified architecture and provides complementary nucleotide-level binding-site localization, supporting a site-aware view of RNA–ligand recognition under distribution shift.
Shihang Wang, Lin Wang, Wei Zhao et al.· Journal of Medicinal Chemist...· 0 citations
We use cookies to run the site and, with your consent, for analytics and to show ads.
See our Cookie Policy.
