A multi-task FP-GNN framework enables accurate prediction of selective PARP inhibitors

Highly Accurate and Explainable Predictions of Small-Molecule Antioxidants for Eight In Vitro Assays Simultaneously through an Alternating Multitask Learning Strategy

Small molecule antioxidants can inhibit or retard oxidation reactions and protect against free radical damage to cells, thus playing a key role in food, cosmetics, pharmaceuticals, the environment, as well as materials. Experimentally driven antioxidant discovery is a major paradigm, and computationally assisted antioxidants are rarely reported. In this study, a functional-group-based alternating multitask self-supervised molecular representation learning method is proposed to simultaneously predict the antioxidant activities of small molecules for eight commonly used in vitro antioxidant assays. Extensive evaluation results reveal that compared with the baseline models, the multitask FG-BERT model achieves the best overall predictive performance, with the highest average F1, BA, ROC-AUC, and PRC-AUC values of 0.860, 0.880, 0.954, and 0.937 for the test sets, respectively. The Y-scrambling testing results further demonstrate that such a deep learning model was not constructed by accident and that it has reliable predictive capabilities. Additionally, the excellent interpretability of the multitask FG-BERT model makes it easy to identify key structural fragments/groups that contribute significantly to the antioxidant effect of a given molecule. Finally, an online antioxidant activity prediction platform called AOP (freely available at https://aop.idruglab.cn/) and its local version were developed based on the high-quality multitask FG-BERT model for experts and nonexperts in the field. We anticipate that it will contribute to the discovery of novel small-molecule antioxidants.

MVGNet: Prediction of PI3K Inhibitors Using Multitask Learning and Multiview Frameworks

PI3K (phosphatidylinositol 3-kinase) is an intracellular phosphatidylinositol kinase composed of a regulatory subunit, p85, and a catalytic subunit, p110. Based on the different structures of the p110 catalytic subunit, PI3K can be divided into four isoforms: PI3Kα, PI3Kβ, PI3Kγ, and PI3Kδ. As molecularly targeted drugs, PI3K inhibitors have demonstrated antiproliferative effects on tumor cells and can also induce cancer cell death. In this study, a multiview deep learning framework (MVGNet) is proposed, which integrates fragment-based pharmacophore information and utilizes multitask learning to capture correlation information between subtasks. This framework predicts the inhibitory activity of molecules against the four PI3K isoforms (PI3Kα, PI3Kβ, PI3Kγ, and PI3Kδ). Compared to baseline prediction models based on three traditional machine learning methods (RF, SVM, and XGBoost) and four deep learning algorithms (GAT, D-MPNN, CMPNN, and KANO), our model demonstrates superior performance. The evaluation results show that our model achieves the highest average AUC-ROC and AUC-PR values on the test set, which are 0.927 ± 0.006 and 0.980 ± 0.002, respectively. This study provides a reference for exploring the structure-activity relationship of PI3K inhibitors.

MalariaFlow: A comprehensive deep learning platform for multistage phenotypic antimalarial drug discovery

Malaria remains a significant global health challenge due to the growing drug resistance of Plasmodium parasites and the failure to block transmission within human host. While machine learning (ML) and deep learning (DL) methods have shown promise in accelerating antimalarial drug discovery, the performance of deep learning models based on molecular graph and other co-representation approaches warrants further exploration. Current research has overlooked mutant strains of the malaria parasite with varying degrees of sensitivity or resistance, and has not covered the prediction of inhibitory activities across the three major life cycle stages (liver, asexual blood, and gametocyte) within the human host, which is crucial for both treatment and transmission blocking. In this study, we manually curated a benchmark antimalarial activity dataset comprising 407,404 unique compounds and 410,654 bioactivity data points across ten Plasmodium phenotypes and three stages. The performance was systematically compared among two fingerprint-based ML models (RF::Morgan and XGBoost:Morgan), four graph-based DL models (GCN, GAT, MPNN, and Attentive FP), and three co-representations DL models (FP-GNN, HiGNN, and FG-BERT), which reveal that: 1) The FP-GNN model achieved the best predictive performance, outperforming the other methods in distinguishing active and inactive compounds across balanced, more positive, and more negative datasets, with an overall AUROC of 0.900; 2) Fingerprint-based ML models outperformed graph-based DL models on large datasets (>1000 compounds), but the three co-representations DL models were able to incorporate domain-specific chemical knowledge to bridge this gap, achieving better predictive performance. These findings provide valuable guidance for selecting appropriate ML and DL methods for antimalarial activity prediction tasks. The interpretability analysis of the FP-GNN model revealed its ability to accurately capture the key structural features responsible for the liver- and blood-stage activities of the known antimalarial drug atovaquone. Finally, we developed a web server, MalariaFlow, incorporating these high-quality models for antimalarial activity prediction, virtual screening, and similarity search, successfully predicting novel triple-stage antimalarial hits validated through experimental testing, demonstrating its effectiveness and value in discovering potential multistage antimalarial drug candidates.

Prediction of Cytochrome P450 Substrates Using the Explainable Multitask Deep Learning Models

Cytochromes P450 (P450s or CYPs) are the most important phase I metabolic enzymes in the human body and are responsible for metabolizing ∼75% of the clinically used drugs. P450-mediated metabolism is also closely associated with the formation of toxic metabolites and drug-drug interactions. Therefore, it is of high importance to predict if a compound is the substrate of a given P450 in the early stage of drug development. In this study, we built the multitask learning models to simultaneously predict the substrates of five major drug-metabolizing P450 enzymes, namely, CYP3A4, 2C9, 2C19, 2D6, and 1A2, based on the collected substrate data sets. Compared to the single-task model and conventional machine learning models, the multitask fingerprints and graph neural networks model achieved superior performance with the average AUC values of 90.8% on the test set. Notably, the multitask model demonstrated its good performance on the small amount of substrate data sets such as CYP1A2, 2C9, and 2C19. In addition, the Shapley additive explanation and the attention mechanism were used to reveal specific substructures associated with P450 substrates, which were further confirmed and complemented by the substructure mining tool and the literature.

GraphEGFR: Multi-task and transfer learning based on molecular graph attention mechanism and fingerprints improving inhibitor bioactivity prediction for EGFR family proteins on data scarcity

The proteins within the human epidermal growth factor receptor (EGFR) family, members of the tyrosine kinase receptor family, play a pivotal role in the molecular mechanisms driving the development of various tumors. Tyrosine kinase inhibitors, key compounds in targeted therapy, encounter challenges in cancer treatment due to emerging drug resistance mutations. Consequently, machine learning has undergone significant evolution to address the challenges of cancer drug discovery related to EGFR family proteins. However, the application of deep learning in this area is hindered by inherent difficulties associated with small-scale data, particularly the risk of overfitting. Moreover, the design of a model architecture that facilitates learning through multi-task and transfer learning, coupled with appropriate molecular representation, poses substantial challenges. In this study, we introduce GraphEGFR, a deep learning regression model designed to enhance molecular representation and model architecture for predicting the bioactivity of inhibitors against both wild-type and mutant EGFR family proteins. GraphEGFR integrates a graph attention mechanism for molecular graphs with deep and convolutional neural networks for molecular fingerprints. We observed that GraphEGFR models employing multi-task and transfer learning strategies generally achieve predictive performance comparable to existing competitive methods. The integration of molecular graphs and fingerprints adeptly captures relationships between atoms and enables both global and local pattern recognition. We further validated potential multi-targeted inhibitors for wild-type and mutant HER1 kinases, exploring key amino acid residues through molecular dynamics simulations to understand molecular interactions. This predictive model offers a robust strategy that could significantly contribute to overcoming the challenges of developing deep learning models for drug discovery with limited data and exploring new frontiers in multi-targeted kinase drug discovery for EGFR family proteins.

3MTox: A motif-level graph-based multi-view chemical language model for toxicity identification with deep interpretation

Toxicity identification plays a key role in maintaining human health, as it can alert humans to the potential hazards caused by long-term exposure to a wide variety of chemical compounds. Experimental methods for determining toxicity are time-consuming, and costly, while computational methods offer an alternative for the early identification of toxicity. For example, some classical ML and DL methods, which demonstrate excellent performance in toxicity prediction. However, these methods also have some defects, such as over-reliance on artificial features and easy overfitting, etc. Proposing novel models with superior prediction performance is still an urgent task. In this study, we propose a motifs-level graph-based multi-view pretraining language model, called 3MTox, for toxicity identification. The 3MTox model uses Bidirectional Encoder Representations from Transformers (BERT) as the backbone framework, and a motif graph as input. The results of extensive experiments showed that our 3MTox model achieved state-of-the-art performance on toxicity benchmark datasets and outperformed the baseline models considered. In addition, the interpretability of the model ensures that the it can quickly and accurately identify toxicity sites in a given molecule, thereby contributing to the determination of the status of toxicity and associated analyses. We think that the 3MTox model is among the most promising tools that are currently available for toxicity identification.

Improved QSAR models for PARP-1 inhibition using data balancing, interpretable machine learning, and matched molecular pair analysis

The poly (ADP-ribose) polymerase-1 (PARP-1) enzyme is an important target in the treatment of breast cancer. Currently, treatment options include the drugs Olaparib, Niraparib, Rucaparib, and Talazoparib; however, these drugs can cause severe side effects including hematological toxicity and cardiotoxicity. Although in silico models for the prediction of PARP-1 activity have been developed, the drawbacks of these models include low specificity, a narrow applicability domain, and a lack of interpretability. To address these issues, a comprehensive machine learning (ML)-based quantitative structure-activity relationship (QSAR) approach for the informed prediction of PARP-1 activity is presented. Classification models built using the Synthetic Minority Oversampling Technique (SMOTE) for data balancing gave robust and predictive models based on the K-nearest neighbor algorithm (accuracy 0.86, sensitivity 0.88, specificity 0.80). Regression models were built on structurally congeneric datasets, with the models for the phthalazinone class and fused cyclic compounds giving the best performance. In accordance with the Organization for Economic Cooperation and Development (OECD) guidelines, a mechanistic interpretation is proposed using the Shapley Additive Explanations (SHAP) to identify the important topological features to differentiate between PARP-1 actives and inactives. Moreover, an analysis of the PARP-1 dataset revealed the prevalence of activity cliffs, which possibly negatively impacts the model's predictive performance. Finally, a set of chemical transformation rules were extracted using the matched molecular pair analysis (MMPA) which provided mechanistic insights and can guide medicinal chemists in the design of novel PARP-1 inhibitors.

Meta Learning with Attention Based FP-GNNs for Few-Shot Molecular Property Prediction

Molecular property prediction holds significant importance in drug discovery, enabling the identification of biologically active compounds with favorable drug-like properties. However, the low data problem, arising from the scarcity of labeled data in drug discovery, poses a substantial obstacle for accurate predictions. To address this challenge, we introduce a novel architecture, AttFPGNN-MAML, for few-shot molecular property prediction. The proposed approach incorporates a hybrid feature representation to enrich molecular representations and model intermolecular relationships specific to the task. By leveraging ProtoMAML, a meta-learning strategy, our model is trained and adapted to new tasks. Evaluation on two few-shot data sets, MoleculeNet and FS-Mol, demonstrates our method's superior performance in three out of four tasks and across various support set sizes. These results convincingly validate the effectiveness of our method in the realm of few-shot molecular property prediction. The source code is publicly available at https://github.com/sanomics-lab/AttFPGNN-MAML.

FG-BERT: a generalized and self-supervised functional group-based molecular representation learning framework for properties prediction

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.

DeepCancerMap: A versatile deep learning platform for target- and cell-based anticancer drug discovery

Discovering new anticancer drugs has been widely concerned and remains an open challenge. Target- and phenotypic-based experimental screening represent two mainstream anticancer drug discovery methods, which suffer from time-consuming, labor-intensive, and high experimental costs. In this study, we collected 485,900 compounds involving in 3,919,974 bioactivity records against 426 anticancer targets and 346 cancer cell lines from academic literature, as well as 60 tumor cell lines from NCI-60 panel. A total of 832 classification models (426 target- and 406 cell-based predictive models) were then constructed to predict the inhibitory activity of compounds against targets and tumor cell lines using FP-GNN deep learning method. Compared to the classical machine learning and deep learning methods, the FP-GNN models achieve considerable overall predictive performance, with the highest AUC values of 0.91, 0.88, 0.91 for the test sets of targets, academia-sourced and NCI-60 cancer cell lines, respectively. A user-friendly webserver called DeepCancerMap and its local version were developed based on these high-quality models, enabling users to perform anticancer drug discovery-related tasks including large-scale virtual screening, profiling prediction of anticancer agents, target fishing, and drug repositioning. We anticipate this platform to accelerate the discovery of anticancer drugs in the field. DeepCancerMap is freely available at https://deepcancermap.idruglab.cn.

HiGNN: A Hierarchical Informative Graph Neural Network for Molecular Property Prediction Equipped with Feature-Wise Attention

Elucidating and accurately predicting the druggability and bioactivities of molecules plays a pivotal role in drug design and discovery and remains an open challenge. Recently, graph neural networks (GNNs) have made remarkable advancements in graph-based molecular property prediction. However, current graph-based deep learning methods neglect the hierarchical information of molecules and the relationships between feature channels. In this study, we propose a well-designed hierarchical informative graph neural network (termed HiGNN) framework for predicting molecular property by utilizing a corepresentation learning of molecular graphs and chemically synthesizable breaking of retrosynthetically interesting chemical substructure (BRICS) fragments. Furthermore, a plug-and-play feature-wise attention block is first designed in HiGNN architecture to adaptively recalibrate atomic features after the message passing phase. Extensive experiments demonstrate that HiGNN achieves state-of-the-art predictive performance on many challenging drug discovery-associated benchmark data sets. In addition, we devise a molecule-fragment similarity mechanism to comprehensively investigate the interpretability of the HiGNN model at the subgraph level, indicating that HiGNN as a powerful deep learning tool can help chemists and pharmacists identify the key components of molecules for designing better molecules with desired properties or functions. The source code is publicly available at https://github.com/idruglab/hignn.

DEEPCYPs: A deep learning platform for enhanced cytochrome P450 activity prediction

Cytochrome P450 (CYP) is a superfamily of heme-containing oxidizing enzymes involved in the metabolism of a wide range of medicines, xenobiotics, and endogenous compounds. Five of the CYPs (1A2, 2C9, 2C19, 2D6, and 3A4) are responsible for metabolizing the vast majority of approved drugs. Adverse drug-drug interactions, many of which are mediated by CYPs, are one of the important causes for the premature termination of drug development and drug withdrawal from the market. In this work, we reported in silicon classification models to predict the inhibitory activity of molecules against these five CYP isoforms using our recently developed FP-GNN deep learning method. The evaluation results showed that, to the best of our knowledge, the multi-task FP-GNN model achieved the best predictive performance with the highest average AUC (0.905), F1 (0.779), BA (0.819), and MCC (0.647) values for the test sets, even compared to advanced machine learning, deep learning, and existing models. Y-scrambling testing confirmed that the results of the multi-task FP-GNN model were not attributed to chance correlation. Furthermore, the interpretability of the multi-task FP-GNN model enables the discovery of critical structural fragments associated with CYPs inhibition. Finally, an online webserver called DEEPCYPs and its local version software were created based on the optimal multi-task FP-GNN model to detect whether compounds bear potential inhibitory activity against CYPs, thereby promoting the prediction of drug-drug interactions in clinical practice and could be used to rule out inappropriate compounds in the early stages of drug discovery and/or identify new CYPs inhibitors.