Geometric deep learning methods and applications in 3D structure-based drug design.

3D structure-based drug design (SBDD) is considered a challenging and rational way for innovative drug discovery. Geometric deep learning is a promising approach that solves the accurate model training of 3D SBDD through building neural network models to learn non-Euclidean data, such as 3D molecular graphs and manifold data. Here, we summarize geometric deep learning methods and applications that contain 3D molecular representations, equivariant graph neural networks (EGNNs), and six generative model methods [diffusion model, flow-based model, generative adversarial networks (GANs), variational autoencoder (VAE), autoregressive models, and energy-based models]. Our review provides insights into geometric deep learning methods and advanced applications of 3D SBDD that will be of relevance for the drug discovery community.

A dual diffusion model enables 3D molecule generation and lead optimization based on target pockets.

Structure-based generative chemistry is essential in computer-aided drug discovery by exploring a vast chemical space to design ligands with high binding affinity for targets. However, traditional in silico methods are limited by computational inefficiency, while machine learning approaches face bottlenecks due to auto-regressive sampling. To address these concerns, we have developed a conditional deep generative model, PMDM, for 3D molecule generation fitting specified targets. PMDM consists of a conditional equivariant diffusion model with both local and global molecular dynamics, enabling PMDM to consider the conditioned protein information to generate molecules efficiently. The comprehensive experiments indicate that PMDM outperforms baseline models across multiple evaluation metrics. To evaluate the applications of PMDM under real drug design scenarios, we conduct lead compound optimization for SARS-CoV-2 main protease (Mpro) and Cyclin-dependent Kinase 2 (CDK2), respectively. The selected lead optimization molecules are synthesized and evaluated for their in-vitro activities against CDK2, displaying improved CDK2 activity.

Structure-Aware DropEdge Toward Deep Graph Convolutional Networks.

It has been discovered that graph convolutional networks (GCNs) encounter a remarkable drop in performance when multiple layers are piled up. The main factor that accounts for why deep GCNs fail lies in oversmoothing, which isolates the network output from the input with the increase of network depth, weakening expressivity and trainability. In this article, we start by investigating refined measures upon DropEdge-an existing simple yet effective technique to relieve oversmoothing. We term our method as DropEdge ++ for its two structure-aware samplers in contrast to DropEdge: layer-dependent (LD) sampler and feature-dependent (FD) sampler. Regarding the LD sampler, we interestingly find that increasingly sampling edges from the bottom layer yields superior performance than the decreasing counterpart as well as DropEdge. We theoretically reveal this phenomenon with mean-edge-number (MEN), a metric closely related to oversmoothing. For the FD sampler, we associate the edge sampling probability with the feature similarity of node pairs and prove that it further correlates the convergence subspace of the output layer with the input features. Extensive experiments on several node classification benchmarks, including both full-and semi-supervised tasks, illustrate the efficacy of DropEdge ++ and its compatibility with a variety of backbones by achieving generally better performance over DropEdge and the no-drop version.

Accelerated Discovery of Macrocyclic CDK2 Inhibitor QR-6401 by Generative Models and Structure-Based Drug Design.

Selective CDK2 inhibitors have the potential to provide effective therapeutics for CDK2-dependent cancers and for combating drug resistance due to high cyclin E1 (CCNE1) expression intrinsically or CCNE1 amplification induced by treatment of CDK4/6 inhibitors. Generative models that take advantage of deep learning are being increasingly integrated into early drug discovery for hit identification and lead optimization. Here we report the discovery of a highly potent and selective macrocyclic CDK2 inhibitor QR-6401 (23) accelerated by the application of generative models and structure-based drug design (SBDD). QR-6401 (23) demonstrated robust antitumor efficacy in an OVCAR3 ovarian cancer xenograft model via oral administration.

Blood-brain barrier penetration prediction enhanced by uncertainty estimation.

Blood-brain barrier is a pivotal factor to be considered in the process of central nervous system (CNS) drug development, and it is of great significance to rapidly explore the blood-brain barrier permeability (BBBp) of compounds in silico in early drug discovery process. Here, we focus on whether and how uncertainty estimation methods improve in silico BBBp models. We briefly surveyed the current state of in silico BBBp prediction and uncertainty estimation methods of deep learning models, and curated an independent dataset to determine the reliability of the state-of-the-art algorithms. The results exhibit that, despite the comparable performance on BBBp prediction between graph neural networks-based deep learning models and conventional physicochemical-based machine learning models, the GROVER-BBBp model shows greatly improvement when using uncertainty estimations. In particular, the strategy combined Entropy and MC-dropout can increase the accuracy of distinguishing BBB + from BBB - to above 99% by extracting predictions with high confidence level (uncertainty score < 0.1). Case studies on preclinical/clinical drugs for Alzheimer' s disease and marketed antitumor drugs that verified by literature proved the application value of uncertainty estimation enhanced BBBp prediction model, that may facilitate the drug discovery in the field of CNS diseases and metastatic brain tumors.

Drug repositioning in drug discovery of T2DM and repositioning potential of antidiabetic agents.

Repositioning or repurposing drugs account for a substantial part of entering approval pipeline drugs, which indicates that drug repositioning has huge market potential and value. Computational technologies such as machine learning methods have accelerated the process of drug repositioning in the last few decades years. The repositioning potential of type 2 diabetes mellitus (T2DM) drugs for various diseases such as cancer, neurodegenerative diseases, and cardiovascular diseases have been widely studied. Hence, the related summary about repurposing antidiabetic drugs is of great significance. In this review, we focus on the machine learning methods for the development of new T2DM drugs and give an overview of the repurposing potential of the existing antidiabetic agents.

Cross-dependent graph neural networks for molecular property prediction.

MOTIVATION: The crux of molecular property prediction is to generate meaningful representations of the molecules. One promising route is to exploit the molecular graph structure through graph neural networks (GNNs). Both atoms and bonds significantly affect the chemical properties of a molecule, so an expressive model ought to exploit both node (atom) and edge (bond) information simultaneously. Inspired by this observation, we explore the multi-view modeling with GNN (MVGNN) to form a novel paralleled framework, which considers both atoms and bonds equally important when learning molecular representations. In specific, one view is atom-central and the other view is bond-central, then the two views are circulated via specifically designed components to enable more accurate predictions. To further enhance the expressive power of MVGNN, we propose a cross-dependent message-passing scheme to enhance information communication of different views. The overall framework is termed as CD-MVGNN. RESULTS: We theoretically justify the expressiveness of the proposed model in terms of distinguishing non-isomorphism graphs. Extensive experiments demonstrate that CD-MVGNN achieves remarkably superior performance over the state-of-the-art models on various challenging benchmarks. Meanwhile, visualization results of the node importance are consistent with prior knowledge, which confirms the interpretability power of CD-MVGNN. AVAILABILITY AND IMPLEMENTATION: The code and data underlying this work are available in GitHub at https://github.com/uta-smile/CD-MVGNN. SUPPLEMENTARY INFORMATION: Supplementary data are available at Bioinformatics online.

Development and Validation of an Image-based Deep Learning Algorithm for Detection of Synchronous Peritoneal Carcinomatosis in Colorectal Cancer.

OBJECTIVE: The aim of this study was to build a SVM classifier using ResNet-3D algorithm by artificial intelligence for prediction of synchronous PC. BACKGROUND: Adequate detection and staging of PC from CRC remain difficult. METHODS: The primary tumors in synchronous PC were delineated on preoperative contrast-enhanced computed tomography (CT) images. The features of adjacent peritoneum were extracted to build a ResNet3D + SVM classifier. The performance of ResNet3D + SVM classifier was evaluated in the test set and was compared to routine CT which was evaluated by radiologists. RESULTS: The training set consisted of 19,814 images from 54 patients with PC and 76 patients without PC. The test set consisted of 7837 images from 40 test patients. The ResNet-3D spent only 34 seconds to analyze the test images. To increase the accuracy of PC detection, we have built a SVM classifier by integrating ResNet-3D features with twelve PC-specific features (P < 0.05). The ResNet3D + SVM classifier showed accuracy of 94.11% with AUC of 0.922 (0.912-0.944), sensitivity of 93.75%, specificity of 94.44%, positive predictive value (PPV) of 93.75%, and negative predictive value (NPV) of 94.44% in the test set. The performance was superior to routine contrast-enhanced CT (AUC: 0.791). CONCLUSIONS: The ResNet3D + SVM classifier based on deep learning algorithm using ResNet-3D framework has shown great potential in prediction of synchronous PC in CRC.

Discovery of Pyrazolo[3,4-d]pyridazinone Derivatives as Selective DDR1 Inhibitors via Deep Learning Based Design, Synthesis, and Biological Evaluation.

Alterations of discoidin domain receptor1 (DDR1) may lead to increased production of inflammatory cytokines, making DDR1 an attractive target for inflammatory bowel disease (IBD) therapy. A scaffold-based molecular design workflow was established and performed by integrating a deep generative model, kinase selectivity screening and molecular docking, leading to a novel DDR1 inhibitor compound 2, which showed potent DDR1 inhibition profile (IC50 = 10.6 ± 1.9 nM) and excellent selectivity against a panel of 430 kinases (S (10) = 0.002 at 0.1 µM). Compound 2 potently inhibited the expression of pro-inflammatory cytokines and DDR1 autophosphorylation in cells, and it also demonstrated promising oral therapeutic effect in a dextran sulfate sodium (DSS)-induced mouse colitis model.

Multi-instance learning of graph neural networks for aqueous pKa prediction.

MOTIVATION: The acid dissociation constant (pKa) is a critical parameter to reflect the ionization ability of chemical compounds and is widely applied in a variety of industries. However, the experimental determination of pKa is intricate and time-consuming, especially for the exact determination of micro-pKa information at the atomic level. Hence, a fast and accurate prediction of pKa values of chemical compounds is of broad interest. RESULTS: Here, we compiled a large-scale pKa dataset containing 16 595 compounds with 17 489 pKa values. Based on this dataset, a novel pKa prediction model, named Graph-pKa, was established using graph neural networks. Graph-pKa performed well on the prediction of macro-pKa values, with a mean absolute error around 0.55 and a coefficient of determination around 0.92 on the test dataset. Furthermore, combining multi-instance learning, Graph-pKa was also able to automatically deconvolute the predicted macro-pKa into discrete micro-pKa values. AVAILABILITY AND IMPLEMENTATION: The Graph-pKa model is now freely accessible via a web-based interface (https://pka.simm.ac.cn/). SUPPLEMENTARY INFORMATION: Supplementary data are available at Bioinformatics online.

Recognizing Predictive Substructures with Subgraph Information Bottleneck.

The emergence of Graph Convolutional Network (GCN) has greatly boosted the progress of graph learning. However, two disturbing factors, noise and redundancy in graph data, and lack of interpretation for prediction results, impede further development of GCN. One solution is to recognize a predictive yet compressed subgraph to get rid of the noise and redundancy and obtain the interpretable part of the graph. This setting of subgraph is similar to the information bottleneck (IB) principle, which is less studied on graph-structured data and GCN. Inspired by the IB principle, we propose a novel subgraph information bottleneck (SIB) framework to recognize such subgraphs, named IB-subgraph. However, the intractability of mutual information and the discrete nature of graph data makes the objective of SIB notoriously hard to optimize. To this end, we introduce a bilevel optimization scheme coupled with a mutual information estimator for irregular graphs. Moreover, we propose a continuous relaxation for subgraph selection with a connectivity loss for stabilization. We further theoretically prove the error bound of our estimation scheme for mutual information and the noise-invariant nature of IB-subgraph. Extensive experiments on graph learning and large-scale point cloud tasks demonstrate the superior property of IB-subgraph.

A Novel Scalarized Scaffold Hopping Algorithm with Graph-Based Variational Autoencoder for Discovery of JAK1 Inhibitors.

We have developed a graph-based Variational Autoencoder with Gaussian Mixture hidden space (GraphGMVAE), a deep learning approach for controllable magnitude of scaffold hopping in generative chemistry. It can effectively and accurately generate molecules from a given reference compound, with excellent scaffold novelty against known molecules in the literature or patents (97.9% are novel scaffolds). Moreover, a pipeline for prioritizing the generated compounds was also proposed to narrow down our validation focus. In this work, GraphGMVAE was validated by rapidly hopping the scaffold from FDA-approved upadacitinib, which is an inhibitor of human Janus kinase 1 (JAK1), to generate more potent molecules with novel chemical scaffolds. Seven compounds were synthesized and tested to be active in biochemical assays. The most potent molecule has 5.0 nM activity against JAK1 kinase, which shows that the GraphGMVAE model can design molecules like how a human expert does but with high efficiency and accuracy.

Generative Models for De Novo Drug Design.

Artificial intelligence (AI) is booming. Among various AI approaches, generative models have received much attention in recent years. Inspired by these successes, researchers are now applying generative model techniques to de novo drug design, which has been considered as the "holy grail" of drug discovery. In this Perspective, we first focus on describing models such as recurrent neural network, autoencoder, generative adversarial network, transformer, and hybrid models with reinforcement learning. Next, we summarize the applications of generative models to drug design, including generating various compounds to expand the compound library and designing compounds with specific properties, and we also list a few publicly available molecular design tools based on generative models which can be used directly to generate molecules. In addition, we also introduce current benchmarks and metrics frequently used for generative models. Finally, we discuss the challenges and prospects of using generative models to aid drug design.

WADDAICA: A webserver for aiding protein drug design by artificial intelligence and classical algorithm.

Artificial intelligence can train the related known drug data into deep learning models for drug design, while classical algorithms can design drugs through established and predefined procedures. Both deep learning and classical algorithms have their merits for drug design. Here, the webserver WADDAICA is built to employ the advantage of deep learning model and classical algorithms for drug design. The WADDAICA mainly contains two modules. In the first module, WADDAICA provides deep learning models for scaffold hopping of compounds to modify or design new novel drugs. The deep learning model which is used in WADDAICA shows a good scoring power based on the PDBbind database. In the second module, WADDAICA supplies functions for modifying or designing new novel drugs by classical algorithms. WADDAICA shows better Pearson and Spearman correlations of binding affinity than Autodock Vina that is considered to have the best scoring power. Besides, WADDAICA supplies a friendly and convenient web interface for users to submit drug design jobs. We believe that WADDAICA is a useful and effective tool to help researchers to modify or design novel drugs by deep learning models and classical algorithms. WADDAICA is free and accessible at https://bqflab.github.io or https://heisenberg.ucam.edu:5000.

Drug repurposing against breast cancer by integrating drug-exposure expression profiles and drug-drug links based on graph neural network.

MOTIVATION: Breast cancer is one of the leading causes of cancer deaths among women worldwide. It is necessary to develop new breast cancer drugs because of the shortcomings of existing therapies. The traditional discovery process is time-consuming and expensive. Repositioning of clinically approved drugs has emerged as a novel approach for breast cancer therapy. However, serendipitous or experiential repurposing cannot be used as a routine method. RESULTS: In this study, we proposed a graph neural network model GraphRepur based on GraphSAGE for drug repurposing against breast cancer. GraphRepur integrated two major classes of computational methods, drug network-based and drug signature-based. The differentially expressed genes of disease, drug-exposure gene expression data and the drug-drug links information were collected. By extracting the drug signatures and topological structure information contained in the drug relationships, GraphRepur can predict new drugs for breast cancer, outperforming previous state-of-the-art approaches and some classic machine learning methods. The high-ranked drugs have indeed been reported as new uses for breast cancer treatment recently. AVAILABILITYAND IMPLEMENTATION: The source code of our model and datasets are available at: https://github.com/cckamy/GraphRepur and https://figshare.com/articles/software/GraphRepur_Breast_Cancer_Drug_Repurposing/14220050. SUPPLEMENTARY INFORMATION: Supplementary data are available at Bioinformatics online.

MolAICal: a soft tool for 3D drug design of protein targets by artificial intelligence and classical algorithm.

Deep learning is an important branch of artificial intelligence that has been successfully applied into medicine and two-dimensional ligand design. The three-dimensional (3D) ligand generation in the 3D pocket of protein target is an interesting and challenging issue for drug design by deep learning. Here, the MolAICal software is introduced to supply a way for generating 3D drugs in the 3D pocket of protein targets by combining with merits of deep learning model and classical algorithm. The MolAICal software mainly contains two modules for 3D drug design. In the first module of MolAICal, it employs the genetic algorithm, deep learning model trained by FDA-approved drug fragments and Vinardo score fitting on the basis of PDBbind database for drug design. In the second module, it uses deep learning generative model trained by drug-like molecules of ZINC database and molecular docking invoked by Autodock Vina automatically. Besides, the Lipinski's rule of five, Pan-assay interference compounds (PAINS), synthetic accessibility (SA) and other user-defined rules are introduced for filtering out unwanted ligands in MolAICal. To show the drug design modules of MolAICal, the membrane protein glucagon receptor and non-membrane protein SARS-CoV-2 main protease are chosen as the investigative drug targets. The results show MolAICal can generate the various and novel ligands with good binding scores and appropriate XLOGP values. We believe that MolAICal can use the advantages of deep learning model and classical programming for designing 3D drugs in protein pocket. MolAICal is freely for any nonprofit purpose and accessible at https://molaical.github.io.

Hybrid-DCA: A double asynchronous approach for stochastic dual coordinate ascent.

In prior works, stochastic dual coordinate ascent (SDCA) has been parallelized in a multi-core environment where the cores communicate through shared memory, or in a multi-processor distributed memory environment where the processors communicate through message passing. In this paper, we propose a hybrid SDCA framework for multi-core clusters, the most common high performance computing environment that consists of multiple nodes each having multiple cores and its own shared memory. We distribute data across nodes where each node solves a local problem in an asynchronous parallel fashion on its cores, and then the local updates are aggregated via an asynchronous across-node update scheme. The proposed double asynchronous method converges to a global solution for L-Lipschitz continuous loss functions, and at a linear convergence rate if a smooth convex loss function is used. Extensive empirical comparison has shown that our algorithm scales better than the best known shared-memory methods and runs faster than previous distributed-memory methods. Big datasets, such as one of 280 GB from the LIBSVM repository, cannot be accommodated on a single node and hence cannot be solved by a parallel algorithm. For such a dataset, our hybrid algorithm takes less than 30 seconds to achieve a duality gap of 10-5 on 16 nodes each using 12 cores, which is significantly faster than the best known distributed algorithms, such as CoCoA+, that take more than 160 seconds on 16 nodes.