Unimodal vs Multimodal Learning: A Systematic Evaluation of Fusion Strategies and Model Design for Molecular Property Prediction and Uncertainty Quantification

Self-Supervised Learning for Molecular Property Prediction: Methods, Multimodal Insights, and Benchmark Comparisons.

This review provides a systematic overview of recent advances in SSL-based molecular property prediction and analyzes how multimodal molecular representation learning by integrating sequence, graph, three-dimensional structure, and textual information can improve the quality and expressiveness of molecular representations.

A multimodal representation learning platform for accurate molecular ADMET prediction

Trimole-Hybrid is presented, a task-wise multimodal framework that addresses ADMET heterogeneity by selecting or combining predictors built from complementary molecular representations, and shows sensitivity to changes in essential functional motifs, suggesting its ability to capture ADMET-relevant molecular substructures.

Bayesian Uncertainty-Guided Fidelity Fusion for Bioactivity Prediction.

Accurate prediction of molecular bioactivity is a fundamental goal in rational drug design but remains challenging due to data scarcity and label imbalance. To address these limitations, we propose a unified Bayesian framework that integrates classification-to-regression knowledge fusion, uncertainty quantification, and active learning for data-efficient molecular property prediction. Specifically, we develop the Bayesian Class-Attentive Transformer Network (BCATNet). This model learns activity patterns from abundant classification data and incorporates the predicted probabilities as informative priors to guide the subsequent Bayesian regression task. Structurally, BCATNet employs a cross-token attention mechanism to model nonlinear interactions between class-derived semantics and molecular structural features. Comparative experiments against conventional machine learning models, graph neural networks, pretrained molecular models, and classification-guided baselines further demonstrated that explicit classification-to-regression knowledge fusion can provide a competitive and data-efficient alternative to generic molecular pretraining. Under reduced regression supervision, BCATNet maintained lower prediction errors and stronger robustness than competing models, supporting its utility in label-scarce settings. Beyond accuracy, the Bayesian formulation generated uncertainty estimates that were informative for reliability assessment: high-uncertainty predictions showed larger regression errors, and uncertainty-based risk stratification separated low-, medium-, and high-risk molecular predictions. Finally, BCATNet uncertainty served as an effective acquisition signal in active learning, with uncertainty-driven strategies achieving the best final performance in most benchmark tasks. Overall, BCATNet establishes a generalizable paradigm for uncertainty-aware molecular modeling by bridging classification and regression tasks within a Bayesian framework, offering a principled route toward reliable, interpretable, and resource-efficient drug discovery.

Multimodal contrastive learning for integrating molecular representations and cellular phenotypes in drug-target interaction prediction

Abstract Motivation Accurate prediction of drug-target interactions (DTIs) is fundamental to drug discovery and mechanistic understanding. While deep learning has advanced computational DTI prediction, most existing methods rely primarily on molecular structural representations, including drug structures and protein sequences, while overlooking cellular phenotypes that reflect downstream biological effects. Cell Painting enables high-content morphological profiling that captures systems-level responses to chemical and genetic perturbations but remains underutilized in DTI modeling. Integrating molecular information with cellular phenotypes offers an opportunity to improve both predictive performance and biological interpretability. Results We propose a two-stage contrastive learning framework integrating drug structures, protein sequences, and Cell Painting morphological profiles into a unified embedding space. Stage 1 learns modality-specific representations independently from structure-based and image-based data; Stage 2 aligns these via multi-positive contrastive learning to bridge molecular structural information with cellular phenotypes. Cross-modal retrieval achieves median Recall@10 values of 0.77 (random split) and 0.33 (scaffold split), outperforming bilinear and random baselines. In external DTI prediction on the BIOSNAP dataset, our model achieves an AUC of 0.92 with image-based representations and 0.90 under structure-only settings, surpassing existing methods. Model interpretation via integrated gradients reveals pathway-specific morphological signatures associated with drug targets, providing biologically interpretable insights into drug mechanisms. Availability https://github.com/YJRubyLai/Unified-DTI

LMF-CP: An Interpretable Multimodal Late-Fusion Framework for Compound Carcinogenicity Prediction

Accurately predicting the carcinogenicity of compounds is of great significance for drug discovery, clinical drug safety, and chemical risk assessment. Traditional methods for assessing carcinogenicity rely on animal testing, which suffers from limitations such as time-consuming processes, high costs, significant interspecies differences, and low predictive throughput. In recent years, computational modeling-based prediction methods (such as Quantitative Structure–Activity Relationships, QSAR) have made some progress, but they still face challenges such as insufficient molecular feature information and poor model interpretability. To overcome these barriers, the multimodal deep learning framework LMF-CP (Late Multimodal Fusion of Carcinogenicity Prediction) is proposed to enhance the performance and interpretability of compound carcinogenicity prediction. First, to comprehensively characterize the structural and physicochemical properties of compounds, a multimodal representation system based on four molecular modalities is constructed, namely SMILES sequences, molecular fingerprints, molecular images, and molecular graph structures. Specifically, Text Convolutional Neural Network (TextCNN), Multi-Layer Perceptron (MLP), Visual Geometry Group Network (VGGNet), as well as Molecular Graph Attention Network (MGAT) are employed to process this information, respectively. Second, to integrate information from different molecular representations, a late-stage fusion strategy based on Lasso stacking is employed. On the test set, LMF-CP achieves an area under curve (AUC) of 0.828, an accuracy (ACC) of 0.782, an F1 score of 0.786, a sensitivity (SEN) of 0.786, and a specificity (SPE) of 0.779. In addition, this paper combines Shapley Additive Explanations (SHAP) analysis with Bemis–Murcko scaffold analysis to interpret the model results from two perspectives. Finally, a visual online platform for predicting the carcinogenicity of compounds is designed, providing a convenient tool for the rapid assessment of compound carcinogenicity and structural interpretation.

Multimodal deep learning with a joint uncertainty quantification scheme for drug-target interaction prediction.