Antibody-drug conjugates (ADCs) have revolutionized the field of cancer treatment in the era of precision medicine due to their ability to precisely target cancer cells and release highly effective drugs. Nevertheless, the rational design and discovery of ADCs remain challenging because the relationship between their quintuple structures and activities is difficult to explore and understand. To address this issue, we first introduce a unified deep learning framework called ADCNet to explore such relationship and help design potential ADCs. The ADCNet highly integrates the protein representation learning language model ESM-2 and small-molecule representation learning language model functional group-based bidirectional encoder representations from transformers to achieve activity prediction through learning meaningful features from antigen and antibody protein sequences of ADC, SMILES strings of linker and payload, and drug-antibody ratio (DAR) value. Based on a carefully designed and manually tailored ADC data set, extensive evaluation results reveal that ADCNet performs best on the test set compared to baseline machine learning models across all evaluation metrics. For example, it achieves an average prediction accuracy of 87.12%, a balanced accuracy of 0.8689, and an area under receiver operating characteristic curve of 0.9293 on the test set. In addition, cross-validation, ablation experiments, and external independent testing results further prove the stability, advancement, and robustness of the ADCNet architecture. For the convenience of the community, we develop the first online platform (https://ADCNet.idruglab.cn) for the prediction of ADCs activity based on the optimal ADCNet model, and the source code is publicly available at https://github.com/idrugLab/ADCNet.

The recent progress of deep generative models in modeling complex real-world data distributions has enabled the generation of novel compounds with potential therapeutic applications for various diseases. However, most studies fail to optimize the properties of generated molecules from the perspective of the intrinsic nature of chemical reactions. In this work, we propose a novel molecule generation model to overcome the limitation by deep reinforcement learning, in which an agent learns to optimize the properties of molecules initialized with a chemically inspired contrastive pretrained model. We finally assess the generation model by evaluating its ability to generate inhibitors against two prominent therapeutic targets in cancer treatment. Experimental results show that our model could generate 100% valid and novel structures and also exhibits superior performance in generating molecules with fewer structural alerts against several baselines. More importantly, the molecules generated by our proposed model show potent biological activities against ataxia telangiectasia and Rad3-related (ATR) and cyclin-dependent kinase 9 (CDK9) targets in wet-lab experiments.

Artificial intelligence-based molecular property prediction plays a key role in molecular design such as bioactive molecules and functional materials. In this study, we propose a self-supervised pretraining deep learning (DL) framework, called functional group bidirectional encoder representations from transformers (FG-BERT), pertained based on ~1.45 million unlabeled drug-like molecules, to learn meaningful representation of molecules from function groups. The pretrained FG-BERT framework can be fine-tuned to predict molecular properties. Compared to state-of-the-art (SOTA) machine learning and DL methods, we demonstrate the high performance of FG-BERT in evaluating molecular properties in tasks involving physical chemistry, biophysics and physiology across 44 benchmark datasets. In addition, FG-BERT utilizes attention mechanisms to focus on FG features that are critical to the target properties, thereby providing excellent interpretability for downstream training tasks. Collectively, FG-BERT does not require any artificially crafted features as input and has excellent interpretability, providing an out-of-the-box framework for developing SOTA models for a variety of molecule (especially for drug) discovery tasks.

Accurate prediction of molecular properties, such as physicochemical and bioactive properties, as well as ADME/T (absorption, distribution, metabolism, excretion and toxicity) properties, remains a fundamental challenge for molecular design, especially for drug design and discovery. In this study, we advanced a novel deep learning architecture, termed FP-GNN (fingerprints and graph neural networks), which combined and simultaneously learned information from molecular graphs and fingerprints for molecular property prediction. To evaluate the FP-GNN model, we conducted experiments on 13 public datasets, an unbiased LIT-PCBA dataset and 14 phenotypic screening datasets for breast cell lines. Extensive evaluation results showed that compared to advanced deep learning and conventional machine learning algorithms, the FP-GNN algorithm achieved state-of-the-art performance on these datasets. In addition, we analyzed the influence of different molecular fingerprints, and the effects of molecular graphs and molecular fingerprints on the performance of the FP-GNN model. Analysis of the anti-noise ability and interpretation ability also indicated that FP-GNN was competitive in real-world situations. Collectively, FP-GNN algorithm can assist chemists, biologists and pharmacists in predicting and discovering better molecules with desired functions or properties.

Three-dimensional (3D) molecular similarity, one major ligand-based virtual screening (VS) method, has been widely used in the drug discovery process. A variety of 3D molecular similarity tools have been developed in recent decades. In this study, we assessed a panel of 15 3D molecular similarity programs against the DUD-E and LIT-PCBA datasets, including commercial ROCS and Phase, in terms of screening power and scaffold-hopping power. The results revealed that (1) SHAFTS, LS-align, Phase Shape_Pharm and LIGSIFT showed the best VS capability in terms of screening power. Some 3D similarity tools available to academia can yield relatively better VS performance than commercial ROCS and Phase software. (2) Current 3D similarity VS tools exhibit a considerable ability to capture actives with new chemotypes in terms of scaffold hopping. (3) Multiple conformers relative to single conformations will generally improve VS performance for most 3D similarity tools, with marginal improvement observed in area under the receiving operator characteristic curve values, enrichment factor in the top 1% and hit rate in the top 1% values showed larger improvement. Moreover, redundancy and complementarity analyses of hit lists from different query seeds and different 3D similarity VS tools showed that the combination of different query seeds and/or different 3D similarity tools in VS campaigns retrieved more (and more diverse) active molecules. These findings provide useful information for guiding choices of the optimal 3D molecular similarity tools for VS practices and designing possible combination strategies to discover more diverse active compounds.

SUMMARY: Molecular-similarity searches based on two-dimensional (2D) fingerprint and three-dimensional (3D) shape represent two widely used ligand-based virtual screening (VS) methods in computer-aided drug design. 2D fingerprint-based VS utilizes the binary fragment information on a known ligand, whereas 3D shape-based VS takes advantage of geometric information for predefined features from a 3D conformation. Given their different advantages, it would be desirable to hybridize 2D fingerprint and 3D shape molecular-similarity approaches in drug discovery. Here, we presented a general hybrid molecular-similarity protocol, referred to as HybridSim, obtained by combining the 2D fingerprint- and 3D shape-based similarity search methods and evaluated its performance on 595,036 actives and decoys for 40 pharmaceutically relevant targets available in the Directory of Useful Decoys Enhanced (DUD-E). Our results showed that HybridSim significantly improved the overall performance in 40 VS projects as compared with using only 2D fingerprint and 3D shape methods. Furthermore, HybridSim-VS, the first online platform using the proposed HybridSim method coupled with 17,839,945 screenable and purchasable compounds, was developed to provide large-scale and proficient VS capabilities to experts and nonexperts in the field. AVAILABILITY AND IMPLEMENTATION: HybridSim-VS web server is freely available at http://www.rcidm.org/HybridSim-VS/. CONTACT: [email protected]. SUPPLEMENTARY INFORMATION: Supplementary data are available at Bioinformatics online.

SUMMARY: Evaluation of the capacity for separating actives from challenging decoys is a crucial metric of performance related to molecular docking or a virtual screening workflow. The Directory of Useful Decoys (DUD) and its enhanced version (DUD-E) provide a benchmark for molecular docking, although they only contain a limited set of decoys for limited targets. DecoyFinder was released to compensate the limitations of DUD or DUD-E for building target-specific decoy sets. However, desirable query template design, generation of multiple decoy sets of similar quality, and computational speed remain bottlenecks, particularly when the numbers of queried actives and retrieved decoys increases to hundreds or more. Here, we developed a program suite called RApid DEcoy Retriever (RADER) to facilitate the decoy-based assessment of virtual screening. This program adopts a novel database-management regime that supports rapid and large-scale retrieval of decoys, enables high portability of databases, and provides multifaceted options for designing initial query templates from a large number of active ligands and generating subtle decoy sets. RADER provides two operational modes: as a command-line tool and on a web server. Validation of the performance and efficiency of RADER was also conducted and is described. AVAILABILITY AND IMPLEMENTATION: RADER web server and a local version are freely available at http://rcidm.org/rader/ . CONTACT: [email protected] or [email protected] . SUPPLEMENTARY INFORMATION: Supplementary data are available at Bioinformatics online.