Development of a protein-ligand extended connectivity (PLEC) fingerprint and its application for binding affinity predictions

HydraScreen: A Generalizable Structure-Based Deep Learning Approach to Drug Discovery

We propose HydraScreen, a deep-learning framework for safe and robust accelerated drug discovery. HydraScreen utilizes a state-of-the-art 3D convolutional neural network designed for the effective representation of molecular structures and interactions in protein-ligand binding. We designed an end-to-end pipeline for high-throughput screening and lead optimization, targeting applications in structure-based drug design. We assessed our approach using established public benchmarks based on the CASF-2016 core set, achieving top-tier results in affinity and pose prediction (Pearson's r = 0.86, RMSE = 1.15, Top-1 = 0.95). We introduced a novel approach for interaction profiling, aimed at detecting potential biases within both the model and data sets. This approach not only enhanced interpretability but also reinforced the impartiality of our methodology. Finally, we demonstrated HydraScreen's ability to generalize effectively across novel proteins and ligands through a temporal split. We also provide insights into potential avenues for future development aimed at enhancing the robustness of machine learning scoring functions. HydraScreen (accessible at http://hydrascreen.ro5.ai/paper) provides a user-friendly GUI and a public API, facilitating the easy-access assessment of protein-ligand complexes.

Proximity Graph Networks: Predicting Ligand Affinity with Message Passing Neural Networks

Message passing neural networks (MPNNs) on molecular graphs generate continuous and differentiable encodings of small molecules with state-of-the-art performance on protein-ligand complex scoring tasks. Here, we describe the proximity graph network (PGN) package, an open-source toolkit that constructs ligand-receptor graphs based on atom proximity and allows users to rapidly apply and evaluate MPNN architectures for a broad range of tasks. We demonstrate the utility of PGN by introducing benchmarks for affinity and docking score prediction tasks. Graph networks generalize better than fingerprint-based models and perform strongly for the docking score prediction task. Overall, MPNNs with proximity graph data structures augment the prediction of ligand-receptor complex properties when ligand-receptor data are available.

Comprehensive machine learning boosts structure-based virtual screening for PARP1 inhibitors

Poly ADP-ribose polymerase 1 (PARP1) is an attractive therapeutic target for cancer treatment. Machine-learning scoring functions constitute a promising approach to discovering novel PARP1 inhibitors. Cutting-edge PARP1-specific machine-learning scoring functions were investigated using semi-synthetic training data from docking activity-labelled molecules: known PARP1 inhibitors, hard-to-discriminate decoys property-matched to them with generative graph neural networks and confirmed inactives. We further made test sets harder by including only molecules dissimilar to those in the training set. Comprehensive analysis of these datasets using five supervised learning algorithms, and protein-ligand fingerprints extracted from docking poses and ligand only features revealed one highly predictive scoring function. This is the PARP1-specific support vector machine-based regressor, when employing PLEC fingerprints, which achieved a high Normalized Enrichment Factor at the top 1% on the hardest test set (NEF1% = 0.588, median of 10 repetitions), and was more predictive than any other investigated scoring function, especially the classical scoring function employed as baseline.

An ensemble-based approach to estimate confidence of predicted protein-ligand binding affinity values

When designing a machine learning-based scoring function, we access a limited number of protein-ligand complexes with experimentally determined binding affinity values, representing only a fraction of all possible protein-ligand complexes. Consequently, it is crucial to report a measure of confidence and quantify the uncertainty in the model's predictions during test time. Here, we adopt the conformal prediction technique to evaluate the confidence of a prediction for each member of the core set of the CASF 2016 benchmark. The conformal prediction technique requires a diverse ensemble of predictors for uncertainty estimation. To this end, we introduce ENS-Score as an ensemble predictor, which includes 30 models with different protein-ligand representation approaches and achieves Pearson's correlation of 0.842 on the core set of the CASF 2016 benchmark. Also, we comprehensively investigate the residual error of each data point to assess the normality behavior of the distribution of the residual errors and their correlation to the structural features of the ligands, such as hydrophobic interactions and halogen bonding. In the end, we provide a local host web application to facilitate the usage of ENS-Score. All codes to repeat results are provided at https://github.com/miladrayka/ENS_Score.

Systematic analysis, aggregation and visualisation of interaction fingerprints for molecular dynamics simulation data

Computational methods such as molecular docking or molecular dynamics (MD) simulations have been developed to simulate and explore the interactions between biomolecules. However, the interactions obtained using these methods are difficult to analyse and evaluate. Interaction fingerprints (IFPs) have been proposed to derive interactions from static 3D coordinates and transform them into 1D bit vectors. More recently, the concept has been applied to derive IFPs from MD simulations, which adds a layer of complexity by adding the temporal motion and dynamics of a system. As a result, many IFPs are obtained from one MD simulation, resulting in a large number of individual IFPs that are difficult to analyse compared to IFPs derived from static 3D structures. Scientific contribution: We introduce a new method to systematically aggregate IFPs derived from MD simulation data. In addition, we propose visualisations to effectively analyse and compare IFPs derived from MD simulation data to account for the temporal evolution of interactions and to compare IFPs across different MD simulations. This has been implemented as a freely available Python library and can therefore be easily adopted by other researchers and to different MD simulation datasets.

Machine Learning for Sequence and Structure-Based Protein-Ligand Interaction Prediction

Developing new drugs is too expensive and time -consuming. Accurately predicting the interaction between drugs and targets will likely change how the drug is discovered. Machine learning-based protein-ligand interaction prediction has demonstrated significant potential. In this paper, computational methods, focusing on sequence and structure to study protein-ligand interactions, are examined. Therefore, this paper starts by presenting an overview of the data sets applied in this area, as well as the various approaches applied for representing proteins and ligands. Then, sequence-based and structure-based classification criteria are subsequently utilized to categorize and summarize both the classical machine learning models and deep learning models employed in protein-ligand interaction studies. Moreover, the evaluation methods and interpretability of these models are proposed. Furthermore, delving into the diverse applications of protein-ligand interaction models in drug research is presented. Lastly, the current challenges and future directions in this field are addressed.

TopoFormer: Multiscale Topology-enabled Structure-to-Sequence Transformer for Protein-Ligand Interaction Predictions

Pre-trained deep Transformers have had tremendous success in a wide variety of disciplines. However, in computational biology, essentially all Transformers are built upon the biological sequences, which ignores vital stereochemical information and may result in crucial errors in downstream predictions. On the other hand, three-dimensional (3D) molecular structures are incompatible with the sequential architecture of Transformer and natural language processing (NLP) models in general. This work addresses this foundational challenge by a topological Transformer (TopoFormer). TopoFormer is built by integrating NLP and a multiscale topology techniques, the persistent topological hyperdigraph Laplacian (PTHL), which systematically converts intricate 3D protein-ligand complexes at various spatial scales into a NLP-admissible sequence of topological invariants and homotopic shapes. Element-specific PTHLs are further developed to embed crucial physical, chemical, and biological interactions into topological sequences. TopoFormer surges ahead of conventional algorithms and recent deep learning variants and gives rise to exemplary scoring accuracy and superior performance in ranking, docking, and screening tasks in a number of benchmark datasets. The proposed topological sequences can be extracted from all kinds of structural data in data science to facilitate various NLP models, heralding a new era in AI-driven discovery.

Inactive-enriched machine-learning models exploiting patent data improve structure-based virtual screening for PDL1 dimerizers

INTRODUCTION

Small-molecule Programmable Cell Death Protein 1/Programmable Death-Ligand 1 (PD1/PDL1) inhibition via PDL1 dimerization has the potential to lead to inexpensive drugs with better cancer patient outcomes and milder side effects. However, this therapeutic approach has proven challenging, with only one PDL1 dimerizer reaching early clinical trials so far. There is hence a need for fast and accurate methods to develop alternative PDL1 dimerizers.

OBJECTIVES

We aim to show that structure-based virtual screening (SBVS) based on PDL1-specific machine-learning (ML) scoring functions (SFs) is a powerful drug design tool for detecting PD1/PDL1 inhibitors via PDL1 dimerization.

METHODS

By incorporating the latest MLSF advances, we generated and evaluated PDL1-specific MLSFs (classifiers and inactive-enriched regressors) on two demanding test sets.

RESULTS

60 PDL1-specific MLSFs (30 classifiers and 30 regressors) were generated. Our large-scale analysis provides highly predictive PDL1-specific MLSFs that benefitted from training with large volumes of docked inactives and enabling inactive-enriched regression.

CONCLUSION

PDL1-specific MLSFs strongly outperformed generic SFs of various types on this target and are released here without restrictions.

Fingerprinting Interactions between Proteins and Ligands for Facilitating Machine Learning in Drug Discovery

Molecular recognition is fundamental in biology, underpinning intricate processes through specific protein-ligand interactions. This understanding is pivotal in drug discovery, yet traditional experimental methods face limitations in exploring the vast chemical space. Computational approaches, notably quantitative structure-activity/property relationship analysis, have gained prominence. Molecular fingerprints encode molecular structures and serve as property profiles, which are essential in drug discovery. While two-dimensional (2D) fingerprints are commonly used, three-dimensional (3D) structural interaction fingerprints offer enhanced structural features specific to target proteins. Machine learning models trained on interaction fingerprints enable precise binding prediction. Recent focus has shifted to structure-based predictive modeling, with machine-learning scoring functions excelling due to feature engineering guided by key interactions. Notably, 3D interaction fingerprints are gaining ground due to their robustness. Various structural interaction fingerprints have been developed and used in drug discovery, each with unique capabilities. This review recapitulates the developed structural interaction fingerprints and provides two case studies to illustrate the power of interaction fingerprint-driven machine learning. The first elucidates structure-activity relationships in β2 adrenoceptor ligands, demonstrating the ability to differentiate agonists and antagonists. The second employs a retrosynthesis-based pre-trained molecular representation to predict protein-ligand dissociation rates, offering insights into binding kinetics. Despite remarkable progress, challenges persist in interpreting complex machine learning models built on 3D fingerprints, emphasizing the need for strategies to make predictions interpretable. Binding site plasticity and induced fit effects pose additional complexities. Interaction fingerprints are promising but require continued research to harness their full potential.

An overview of recent advances and challenges in predicting compound-protein interaction (CPI)

Compound-protein interactions (CPIs) are critical in drug discovery for identifying therapeutic targets, drug side effects, and repurposing existing drugs. Machine learning (ML) algorithms have emerged as powerful tools for CPI prediction, offering notable advantages in cost-effectiveness and efficiency. This review provides an overview of recent advances in both structure-based and non-structure-based CPI prediction ML models, highlighting their performance and achievements. It also offers insights into CPI prediction-related datasets and evaluation benchmarks. Lastly, the article presents a comprehensive assessment of the current landscape of CPI prediction, elucidating the challenges faced and outlining emerging trends to advance the field.

Predicting Affinity Through Homology (PATH): Interpretable Binding Affinity Prediction with Persistent Homology

Accurate binding affinity prediction is crucial to structure-based drug design. Recent work used computational topology to obtain an effective representation of protein-ligand interactions. Although persistent homology encodes geometric features, previous works on binding affinity prediction using persistent homology employed uninterpretable machine learning models and failed to explain the underlying geometric and topological features that drive accurate binding affinity prediction. In this work, we propose a novel, interpretable algorithm for protein-ligand binding affinity prediction. Our algorithm achieves interpretability through an effective embedding of distances across bipartite matchings of the protein and ligand atoms into real-valued functions by summing Gaussians centered at features constructed by persistent homology. We name these functions internuclear persistent contours (IPCs) . Next, we introduce persistence fingerprints , a vector with 10 components that sketches the distances of different bipartite matching between protein and ligand atoms, refined from IPCs. Let the number of protein atoms in the protein-ligand complex be n , number of ligand atoms be m , and ω ≈ 2.4 be the matrix multiplication exponent. We show that for any 0 < ε < 1, after an 𝒪 ( mn log( mn )) preprocessing procedure, we can compute an ε -accurate approximation to the persistence fingerprint in 𝒪 ( m log 6 ω ( m/" )) time, independent of protein size. This is an improvement in time complexity by a factor of 𝒪 (( m + n ) 3 ) over any previous binding affinity prediction that uses persistent homology. We show that the representational power of persistence fingerprint generalizes to protein-ligand binding datasets beyond the training dataset. Then, we introduce PATH , Predicting Affinity Through Homology, an interpretable, small ensemble of shallow regression trees for binding affinity prediction from persistence fingerprints. We show that despite using 1,400-fold fewer features, PATH has comparable performance to a previous state-of-the-art binding affinity prediction algorithm that uses persistent homology features. Moreover, PATH has the advantage of being interpretable. Finally, we visualize the features captured by persistence fingerprint for variant HIV-1 protease complexes and show that persistence fingerprint captures binding-relevant structural mutations. The source code for PATH is released open-source as part of the osprey protein design software package.

A practical guide to machine-learning scoring for structure-based virtual screening

Structure-based virtual screening (SBVS) via docking has been used to discover active molecules for a range of therapeutic targets. Chemical and protein data sets that contain integrated bioactivity information have increased both in number and in size. Artificial intelligence and, more concretely, its machine-learning (ML) branch, including deep learning, have effectively exploited these data sets to build scoring functions (SFs) for SBVS against targets with an atomic-resolution 3D model (e.g., generated by X-ray crystallography or predicted by AlphaFold2). Often outperforming their generic and non-ML counterparts, target-specific ML-based SFs represent the state of the art for SBVS. Here, we present a comprehensive and user-friendly protocol to build and rigorously evaluate these new SFs for SBVS. This protocol is organized into four sections: (i) using a public benchmark of a given target to evaluate an existing generic SF; (ii) preparing experimental data for a target from public repositories; (iii) partitioning data into a training set and a test set for subsequent target-specific ML modeling; and (iv) generating and evaluating target-specific ML SFs by using the prepared training-test partitions. All necessary code and input/output data related to three example targets (acetylcholinesterase, HMG-CoA reductase, and peroxisome proliferator-activated receptor-α) are available at https://github.com/vktrannguyen/MLSF-protocol , can be run by using a single computer within 1 week and make use of easily accessible software/programs (e.g., Smina, CNN-Score, RF-Score-VS and DeepCoy) and web resources. Our aim is to provide practical guidance on how to augment training data to enhance SBVS performance, how to identify the most suitable supervised learning algorithm for a data set, and how to build an SF with the highest likelihood of discovering target-active molecules within a given compound library.

Multi-shelled ECIF: improved extended connectivity interaction features for accurate binding affinity prediction

Motivation

Extended connectivity interaction features (ECIF) is a method developed to predict protein-ligand binding affinity, allowing for detailed atomic representation. It performed very well in terms of Comparative Assessment of Scoring Functions 2016 (CASF-2016) scoring power. However, ECIF has the limitation of not being able to adequately account for interatomic distances.

Results

To investigate what kind of distance representation is effective for P-L binding affinity prediction, we have developed two algorithms that improved ECIF's feature extraction method to take distance into account. One is multi-shelled ECIF, which takes into account the distance between atoms by dividing the distance between atoms into multiple layers. The other is weighted ECIF, which weights the importance of interactions according to the distance between atoms. A comparison of these two methods shows that multi-shelled ECIF outperforms weighted ECIF and the original ECIF, achieving a CASF-2016 scoring power Pearson correlation coefficient of 0.877.

Availability and implementation

All the codes and data are available on GitHub (https://github.com/koji11235/MSECIFv2).

ML-PLIC: a web platform for characterizing protein-ligand interactions and developing machine learning-based scoring functions

Cracking the entangling code of protein-ligand interaction (PLI) is of great importance to structure-based drug design and discovery. Different physical and biochemical representations can be used to describe PLI such as energy terms and interaction fingerprints, which can be analyzed by machine learning (ML) algorithms to create ML-based scoring functions (MLSFs). Here, we propose the ML-based PLI capturer (ML-PLIC), a web platform that automatically characterizes PLI and generates MLSFs to identify the potential binders of a specific protein target through virtual screening (VS). ML-PLIC comprises five modules, including Docking for ligand docking, Descriptors for PLI generation, Modeling for MLSF training, Screening for VS and Pipeline for the integration of the aforementioned functions. We validated the MLSFs constructed by ML-PLIC in three benchmark datasets (Directory of Useful Decoys-Enhanced, Active as Decoys and TocoDecoy), demonstrating accuracy outperforming traditional docking tools and competitive performance to the deep learning-based SF, and provided a case study of the Serine/threonine-protein kinase WEE1 in which MLSFs were developed by using the ML-based VS pipeline in ML-PLIC. Underpinning the latest version of ML-PLIC is a powerful platform that incorporates physical and biological knowledge about PLI, leveraging PLI characterization and MLSF generation into the design of structure-based VS pipeline. The ML-PLIC web platform is now freely available at http://cadd.zju.edu.cn/plic/.

Exploring the ability of machine learning-based virtual screening models to identify the functional groups responsible for binding

Many recently proposed structure-based virtual screening models appear to be able to accurately distinguish high affinity binders from non-binders. However, several recent studies have shown that they often do so by exploiting ligand-specific biases in the dataset, rather than identifying favourable intermolecular interactions in the input protein-ligand complex. In this work we propose a novel approach for assessing the extent to which machine learning-based virtual screening models are able to identify the functional groups responsible for binding. To sidestep the difficulty in establishing the ground truth importance of each atom of a large scale set of protein-ligand complexes, we propose a protocol for generating synthetic data. Each ligand in the dataset is surrounded by a randomly sampled point cloud of pharmacophores, and the label assigned to the synthetic protein-ligand complex is determined by a 3-dimensional deterministic binding rule. This allows us to precisely quantify the ground truth importance of each atom and compare it to the model generated attributions. Using our generated datasets, we demonstrate that a recently proposed deep learning-based virtual screening model, PointVS, identified the most important functional groups with 39% more efficiency than a fingerprint-based random forest, suggesting that it would generalise more effectively to new examples. In addition, we found that ligand-specific biases, such as those present in widely used virtual screening datasets, substantially impaired the ability of all ML models to identify the most important functional groups. We have made our synthetic data generation framework available to facilitate the benchmarking of new virtual screening models. Code is available at https://github.com/tomhadfield95/synthVS .

Geometric graph learning with extended atom-types features for protein-ligand binding affinity prediction

Understanding and accurately predicting protein-ligand binding affinity are essential in the drug design and discovery process. At present, machine learning-based methodologies are gaining popularity as a means of predicting binding affinity due to their efficiency and accuracy, as well as the increasing availability of structural and binding affinity data for protein-ligand complexes. In biomolecular studies, graph theory has been widely applied since graphs can be used to model molecules or molecular complexes in a natural manner. In the present work, we upgrade the graph-based learners for the study of protein-ligand interactions by integrating extensive atom types such as SYBYL and extended connectivity interactive features (ECIF) into multiscale weighted colored graphs (MWCG). By pairing with the gradient boosting decision tree (GBDT) machine learning algorithm, our approach results in two different methods, namely sybylGGL-Score and ecifGGL-Score. Both of our models are extensively validated in their scoring power using three commonly used benchmark datasets in the drug design area, namely CASF-2007, CASF-2013, and CASF-2016. The performance of our best model sybylGGL-Score is compared with other state-of-the-art models in the binding affinity prediction for each benchmark. While both of our models achieve state-of-the-art results, the SYBYL atom-type model sybylGGL-Score outperforms other methods by a wide margin in all benchmarks. Finally, the best-performing SYBYL atom-type model is evaluated on two test sets that are independent of CASF benchmarks.

A Robust Machine Learning Framework Built Upon Molecular Representations Predicts CYP450 Inhibition: Toward Precision in Drug Repurposing

Human cytochrome P450 (CYP450) enzymes play a crucial role in drug metabolism and pharmacokinetics. CYP450 inhibition can lead to toxicity, in particular when drugs are co-administered with other drugs and xenobiotics or in the case of polypharmacy. Predicting CYP450 inhibition is also important for rational drug discovery and development, and precision in drug repurposing. In this overarching context, digital transformation of drug discovery and development, for example, using machine and deep learning approaches, offers prospects for prediction of CYP450 inhibition through computational models. We report here the development of a majority-voting machine learning framework to classify inhibitors and noninhibitors for seven major human liver CYP450 isoforms (CYP1A2, CYP2A6, CYP2B6, CYP2C9, CYP2C19, CYP2D6, and CYP3A4). For the machine learning models reported herein, we employed interaction fingerprints that were derived from molecular docking simulations, thus adding an additional layer of information for protein-ligand interactions. The proposed machine learning framework is based on the structure of the binding site of isoforms to produce predictions beyond previously reported approaches. Also, we carried out a comparative analysis so as to identify which representation of test compounds (molecular descriptors, molecular fingerprints, or protein-ligand interaction fingerprints) affects the predictive performance of the models. This work underlines the ways in which the structure of the enzyme catalytic site influences machine learning predictions and the need for robust frameworks toward better-informed predictions.

SVSBI: sequence-based virtual screening of biomolecular interactions

Virtual screening (VS) is a critical technique in understanding biomolecular interactions, particularly in drug design and discovery. However, the accuracy of current VS models heavily relies on three-dimensional (3D) structures obtained through molecular docking, which is often unreliable due to the low accuracy. To address this issue, we introduce a sequence-based virtual screening (SVS) as another generation of VS models that utilize advanced natural language processing (NLP) algorithms and optimized deep K-embedding strategies to encode biomolecular interactions without relying on 3D structure-based docking. We demonstrate that SVS outperforms state-of-the-art performance for four regression datasets involving protein-ligand binding, protein-protein, protein-nucleic acid binding, and ligand inhibition of protein-protein interactions and five classification datasets for protein-protein interactions in five biological species. SVS has the potential to transform current practices in drug discovery and protein engineering.

Beware of Simple Methods for Structure-Based Virtual Screening: The Critical Importance of Broader Comparisons

We discuss how data unbiasing and simple methods such as protein-ligand Interaction FingerPrint (IFP) can overestimate virtual screening performance. We also show that IFP is strongly outperformed by target-specific machine-learning scoring functions, which were not considered in a recent report concluding that simple methods were better than machine-learning scoring functions at virtual screening.

GB-score: Minimally designed machine learning scoring function based on distance-weighted interatomic contact features

In recent years, thanks to advances in computer hardware and dataset availability, data-driven approaches (like machine learning) have become one of the essential parts of the drug design framework to accelerate drug discovery procedures. Constructing a new scoring function, a function that can predict the binding score for a generated protein-ligand pose during docking procedure or a crystal complex, based on machine and deep learning has become an active research area in computer-aided drug design. GB-Score is a state-of-the-art machine learning-based scoring function that utilizes distance-weighted interatomic contact features, PDBbind-v2019 general set, and Gradient Boosting Trees algorithm to the binding affinity prediction. The distance-weighted interatomic contact featurization method used the distance between different ligand and protein atom types for numerical representation of the protein-ligand complex. GB-Score attains Pearson's correlation 0.862 and RMSE 1.190 on the CASF-2016 benchmark test in the scoring power metric. GB-Score's codes are freely available on the web at https://github.com/miladrayka/GB_Score.

Persistent Path-Spectral (PPS) Based Machine Learning for Protein-Ligand Binding Affinity Prediction

Molecular descriptors are essential to quantitative structure activity/property relationship (QSAR/QSPR) models and machine learning models. Here we propose persistent path-spectral (PPS), PPS-based molecular descriptors, and PPS-based machine learning model for the prediction of the protein-ligand binding affinity, for the first time. For the graph, simplicial complex, and hypergraph representation of molecular structures and interactions, the path-Laplacian can be constructed and the derived path-spectral naturally gives a quantitative description of molecules. Further, by introducing the filtration process of the representation, the persistent path-spectral can be derived, which gives a multiscale characterization of molecules. Molecular descriptors from the persistent path-spectral attributes then are combined with the machine learning model, in particular, the gradient boosting tree, to form our PPS-ML model. We test our model on three most commonly used data sets, i.e., PDBbind-v2007, PDBbind-v2013, and PDBbind-v2016, and our model can achieve competitive results.

Active Learning Guided Drug Design Lead Optimization Based on Relative Binding Free Energy Modeling

In silico identification of potent protein inhibitors commonly requires prediction of a ligand binding free energy (BFE). Thermodynamics integration (TI) based on molecular dynamics (MD) simulations is a BFE calculation method capable of acquiring accurate BFE, but it is computationally expensive and time-consuming. In this work, we have developed an efficient automated workflow for identifying compounds with the lowest BFE among thousands of congeneric ligands, which requires only hundreds of TI calculations. Automated machine learning (AutoML) orchestrated by active learning (AL) in an AL-AutoML workflow allows unbiased and efficient search for a small set of best-performing molecules. We have applied this workflow to select inhibitors of the SARS-CoV-2 papain-like protease and were able to find 133 compounds with improved binding affinity, including 16 compounds with better than 100-fold binding affinity improvement. We obtained a hit rate that outperforms that expected of traditional expert medicinal chemist-guided campaigns. Thus, we demonstrate that the combination of AL and AutoML with free energy simulations provides at least 20× speedup relative to the naïve brute force approaches.

Leveraging scaffold information to predict protein-ligand binding affinity with an empirical graph neural network

Protein-ligand binding affinity prediction is an important task in structural bioinformatics for drug discovery and design. Although various scoring functions (SFs) have been proposed, it remains challenging to accurately evaluate the binding affinity of a protein-ligand complex with the known bound structure because of the potential preference of scoring system. In recent years, deep learning (DL) techniques have been applied to SFs without sophisticated feature engineering. Nevertheless, existing methods cannot model the differential contribution of atoms in various regions of proteins, and the relationship between atom properties and intermolecular distance is also not fully explored. We propose a novel empirical graph neural network for accurate protein-ligand binding affinity prediction (EGNA). Graphs of protein, ligand and their interactions are constructed based on different regions of each bound complex. Proteins and ligands are effectively represented by graph convolutional layers, enabling the EGNA to capture interaction patterns precisely by simulating empirical SFs. The contributions of different factors on binding affinity can thus be transparently investigated. EGNA is compared with the state-of-the-art machine learning-based SFs on two widely used benchmark data sets. The results demonstrate the superiority of EGNA and its good generalization capability.

Reducing false positive rate of docking-based virtual screening by active learning

Machine learning-based scoring functions (MLSFs) have become a very favorable alternative to classical scoring functions because of their potential superior screening performance. However, the information of negative data used to construct MLSFs was rarely reported in the literature, and meanwhile the putative inactive molecules recorded in existing databases usually have obvious bias from active molecules. Here we proposed an easy-to-use method named AMLSF that combines active learning using negative molecular selection strategies with MLSF, which can iteratively improve the quality of inactive sets and thus reduce the false positive rate of virtual screening. We chose energy auxiliary terms learning as the MLSF and validated our method on eight targets in the diverse subset of DUD-E. For each target, we screened the IterBioScreen database by AMLSF and compared the screening results with those of the four control models. The results illustrate that the number of active molecules in the top 1000 molecules identified by AMLSF was significantly higher than those identified by the control models. In addition, the free energy calculation results for the top 10 molecules screened out by the AMLSF, null model and control models based on DUD-E also proved that more active molecules can be identified, and the false positive rate can be reduced by AMLSF.

Latent Biases in Machine Learning Models for Predicting Binding Affinities Using Popular Data Sets

Drug design involves the process of identifying and designing molecules that bind well to a given receptor. A vital computational component of this process is the protein-ligand interaction scoring functions that evaluate the binding ability of various molecules or ligands with a given protein receptor binding pocket reasonably accurately. With the publicly available protein-ligand binding affinity data sets in both sequential and structural forms, machine learning methods have gained traction as a top choice for developing such scoring functions. While the performance shown by these models is optimistic, there are several hidden biases present in these data sets themselves that affect the utility of such models for practical purposes such as virtual screening. In this work, we use published methods to systematically investigate several such factors or biases present in these data sets. In our analysis, we highlight the importance of considering sequence, protein-ligand interaction, and pocket structure similarity while constructing data splits and provide an explanation for good protein-only and ligand-only performances in some data sets. Through this study, we provide to the community several pointers for the design of binding affinity predictors and data sets for reliable applicability.

Neural networks prediction of the protein-ligand binding affinity with circular fingerprints

BACKGROUND

Protein-ligand binding affinity is of significant importance in structure-based drug design. Recently, the development of machine learning techniques has provided an efficient and accurate way to predict binding affinity. However, the prediction performance largely depends on how molecules are represented.

OBJECTIVE

Different molecular descriptors are designed to capture different features. The study aims to identify the optimal circular fingerprints for predicting protein-ligand binding affinity with matched neural network architectures.

METHODS

Extended-connectivity fingerprints (ECFP) and protein-ligand extended connectivity fingerprints (PLEC) encode circular atomic and bonding connectivity environments with the preference for intra- and inter-molecular features, respectively. Densely-connected neural networks are employed to map the circular fingerprints of protein-ligand complexes to binding affinitiesRESULTS:The performance of neural networks is sensitive to the parameters used for ECFP and PLEC fingerprints. The R2_score of the evaluated ECFP and PLEC fingerprints reaches 0.52 and 0.49, higher than that of the improperly set ECFP and PLEC fingerprints with R2_score of 0.45 and 0.38, respectively. Additionally, compared to the predictions from the standalone fingerprints, the ECFP+PLEC conjoint ones slightly improve the prediction accuracy with R2_score of approximately 0.55.

CONCLUSION

Both intra- and inter-molecular structural features encoded in the circular fingerprints contribute to the protein-ligand binding affinity. Optimizing the parameters of ECFP and PLEC can enhance performance. The conjoint fingerprint scheme can be generally extended to other molecular descriptors for enhanced feature engineering and improved predictive performance.

Beyond sequence: Structure-based machine learning

Recent breakthroughs in protein structure prediction demarcate the start of a new era in structural bioinformatics. Combined with various advances in experimental structure determination and the uninterrupted pace at which new structures are published, this promises an age in which protein structure information is as prevalent and ubiquitous as sequence. Machine learning in protein bioinformatics has been dominated by sequence-based methods, but this is now changing to make use of the deluge of rich structural information as input. Machine learning methods making use of structures are scattered across literature and cover a number of different applications and scopes; while some try to address questions and tasks within a single protein family, others aim to capture characteristics across all available proteins. In this review, we look at the variety of structure-based machine learning approaches, how structures can be used as input, and typical applications of these approaches in protein biology. We also discuss current challenges and opportunities in this all-important and increasingly popular field.

Learning from Docked Ligands: Ligand-Based Features Rescue Structure-Based Scoring Functions When Trained on Docked Poses

Machine learning scoring functions for protein-ligand binding affinity have been found to consistently outperform classical scoring functions when trained and tested on crystal structures of bound protein-ligand complexes. However, it is less clear how these methods perform when applied to docked poses of complexes. We explore how the use of docked rather than crystallographic poses for both training and testing affects the performance of machine learning scoring functions. Using the PDBbind Core Sets as benchmarks, we show that the performance of a structure-based machine learning scoring function trained and tested on docked poses is lower than that of the same scoring function trained and tested on crystallographic poses. We construct a hybrid scoring function by combining both structure-based and ligand-based features, and show that its ability to predict binding affinity using docked poses is comparable to that of purely structure-based scoring functions trained and tested on crystal poses. We also present a new, freely available validation set─the Updated DUD-E Diverse Subset─for binding affinity prediction using data from DUD-E and ChEMBL. Despite strong performance on docked poses of the PDBbind Core Sets, we find that our hybrid scoring function sometimes generalizes poorly to a protein target not represented in the training set, demonstrating the need for improved scoring functions and additional validation benchmarks.

A multi-reference poly-conformational method for in silico design, optimization, and repositioning of pharmaceutical compounds illustrated for selected SARS-CoV-2 ligands

Background

This work presents a novel computational multi-reference poly-conformational algorithm for design, optimization, and repositioning of pharmaceutical compounds.

Methods

The algorithm searches for candidates by comparing similarities between conformers of the same compound and identifies target compounds, whose conformers are collectively close to the conformers of each compound in the reference set. Reference compounds may possess highly variable MoAs, which directly, and simultaneously, shape the properties of target candidate compounds.

Results

The algorithm functionality has been case study validated in silico, by scoring ChEMBL drugs against FDA-approved reference compounds that either have the highest predicted binding affinity to our chosen SARS-CoV-2 targets or are confirmed to be inhibiting such targets in-vivo. All our top scoring ChEMBL compounds also turned out to be either high-affinity ligands to the chosen targets (as confirmed in separate studies) or show significant efficacy, in-vivo, against those selected targets. In addition to method case study validation, in silico search for new compounds within two virtual libraries from the Enamine database is presented. The library's virtual compounds have been compared to the same set of reference drugs that we used for case study validation: Olaparib, Tadalafil, Ergotamine and Remdesivir. The large reference set of four potential SARS-CoV-2 compounds has been selected, since no drug has been identified to be 100% effective against the virus so far, possibly because each candidate drug was targeting only one, particular MoA. The goal here was to introduce a new methodology for identifying potential candidate(s) that cover multiple MoA-s presented within a set of reference compounds.

Concepts and applications of chemical fingerprint for hit and lead screening

Molecular fingerprints are used to represent chemical (structural, physicochemical, etc.) properties of large-scale chemical sets in a low computational cost way. They have a prominent role in transforming chemical data sets into consistent input formats (bit strings or numeric values) suitable for in silico approaches. In this review, we summarize and classify common and state-of-the-art fingerprints into eight different types (dictionary based, circular, topological, pharmacophore, protein-ligand interaction, shape based, reinforced, and multi). We also highlight applications of fingerprints in early drug research and development (R&D). Thus, this review provides a guide for the selection of appropriate fingerprints of compounds (or ligand-protein complexes) for use in drug R&D.

DHU-Pred: accurate prediction of dihydrouridine sites using position and composition variant features on diverse classifiers

Background

Dihydrouridine (D) is a modified transfer RNA post-transcriptional modification (PTM) that occurs abundantly in bacteria, eukaryotes, and archaea. The D modification assists in the stability and conformational flexibility of tRNA. The D modification is also responsible for pulmonary carcinogenesis in humans.

Objective

For the detection of D sites, mass spectrometry and site-directed mutagenesis have been developed. However, both are labor-intensive and time-consuming methods. The availability of sequence data has provided the opportunity to build computational models for enhancing the identification of D sites. Based on the sequence data, the DHU-Pred model was proposed in this study to find possible D sites.

Methodology

The model was built by employing comprehensive machine learning and feature extraction approaches. It was then validated using in-demand evaluation metrics and rigorous experimentation and testing approaches.

Results

The DHU-Pred revealed an accuracy score of 96.9%, which was considerably higher compared to the existing D site predictors.

Availability and Implementation

A user-friendly web server for the proposed model was also developed and is freely available for the researchers.

Deep generative molecular design reshapes drug discovery

Recent advances and accomplishments of artificial intelligence (AI) and deep generative models have established their usefulness in medicinal applications, especially in drug discovery and development. To correctly apply AI, the developer and user face questions such as which protocols to consider, which factors to scrutinize, and how the deep generative models can integrate the relevant disciplines. This review summarizes classical and newly developed AI approaches, providing an updated and accessible guide to the broad computational drug discovery and development community. We introduce deep generative models from different standpoints and describe the theoretical frameworks for representing chemical and biological structures and their applications. We discuss the data and technical challenges and highlight future directions of multimodal deep generative models for accelerating drug discovery.

Biomolecular Topology: Modelling and Analysis

With the great advancement of experimental tools, a tremendous amount of biomolecular data has been generated and accumulated in various databases. The high dimensionality, structural complexity, the nonlinearity, and entanglements of biomolecular data, ranging from DNA knots, RNA secondary structures, protein folding configurations, chromosomes, DNA origami, molecular assembly, to others at the macromolecular level, pose a severe challenge in their analysis and characterization. In the past few decades, mathematical concepts, models, algorithms, and tools from algebraic topology, combinatorial topology, computational topology, and topological data analysis, have demonstrated great power and begun to play an essential role in tackling the biomolecular data challenge. In this work, we introduce biomolecular topology, which concerns the topological problems and models originated from the biomolecular systems. More specifically, the biomolecular topology encompasses topological structures, properties and relations that are emerged from biomolecular structures, dynamics, interactions, and functions. We discuss the various types of biomolecular topology from structures (of proteins, DNAs, and RNAs), protein folding, and protein assembly. A brief discussion of databanks (and databases), theoretical models, and computational algorithms, is presented. Further, we systematically review related topological models, including graphs, simplicial complexes, persistent homology, persistent Laplacians, de Rham-Hodge theory, Yau-Hausdorff distance, and the topology-based machine learning models.

Harnessing systematic protein-ligand interaction fingerprints for drug discovery

Determining protein-ligand interaction characteristics and mechanisms is crucial to the drug discovery process. Here, we review recent progress and successful applications of a systematic protein-ligand interaction fingerprint (IFP) approach for investigating proteome-wide protein-ligand interactions for drug development. Specifically, we review the use of this IFP approach for revealing polypharmacology across the kinome, predicting promising targets from which to design allosteric inhibitors and covalent kinase inhibitors, uncovering the binding mechanisms of drugs of interest, and demonstrating resistant mechanisms of specific drugs. Together, we demonstrate that the IFP strategy is efficient and practical for drug design research for protein kinases as targets and is extensible to other protein families.

Chemical Space Exploration with Active Learning and Alchemical Free Energies

Drug discovery can be thought of as a search for a needle in a haystack: searching through a large chemical space for the most active compounds. Computational techniques can narrow the search space for experimental follow up, but even they become unaffordable when evaluating large numbers of molecules. Therefore, machine learning (ML) strategies are being developed as computationally cheaper complementary techniques for navigating and triaging large chemical libraries. Here, we explore how an active learning protocol can be combined with first-principles based alchemical free energy calculations to identify high affinity phosphodiesterase 2 (PDE2) inhibitors. We first calibrate the procedure using a set of experimentally characterized PDE2 binders. The optimized protocol is then used prospectively on a large chemical library to navigate toward potent inhibitors. In the active learning cycle, at every iteration a small fraction of compounds is probed by alchemical calculations and the obtained affinities are used to train ML models. With successive rounds, high affinity binders are identified by explicitly evaluating only a small subset of compounds in a large chemical library, thus providing an efficient protocol that robustly identifies a large fraction of true positives.

Systematic Improvement of the Performance of Machine Learning Scoring Functions by Incorporating Features of Protein-Bound Water Molecules

Water molecules at the ligand-protein interfaces play crucial roles in the binding of the ligands, but the behavior of protein-bound water is largely ignored in many currently used machine learning (ML)-based scoring functions (SFs). In an attempt to improve the prediction performance of existing ML-based SFs, we estimated the water distribution with a HydraMap (HM) method and then incorporated the features extracted from protein-bound waters obtained in this way into three ML-based SFs: RF-Score, ECIF, and PLEC. It was found that a combination of HM-based features can consistently improve the performance of all three SFs, including their scoring, ranking, and docking power. HydraMap-based features show consistently good performance with both crystal structures and docked structures, demonstrating their robustness for SFs. Overall, HM-based features, which are a statistical representation of hydration sites at protein-ligand interfaces, are expected to improve the prediction performance for diverse SFs.

PLAS-5k: Dataset of Protein-Ligand Affinities from Molecular Dynamics for Machine Learning Applications

Computational methods and recently modern machine learning methods have played a key role in structure-based drug design. Though several benchmarking datasets are available for machine learning applications in virtual screening, accurate prediction of binding affinity for a protein-ligand complex remains a major challenge. New datasets that allow for the development of models for predicting binding affinities better than the state-of-the-art scoring functions are important. For the first time, we have developed a dataset, PLAS-5k comprised of 5000 protein-ligand complexes chosen from PDB database. The dataset consists of binding affinities along with energy components like electrostatic, van der Waals, polar and non-polar solvation energy calculated from molecular dynamics simulations using MMPBSA (Molecular Mechanics Poisson-Boltzmann Surface Area) method. The calculated binding affinities outperformed docking scores and showed a good correlation with the available experimental values. The availability of energy components may enable optimization of desired components during machine learning-based drug design. Further, OnionNet model has been retrained on PLAS-5k dataset and is provided as a baseline for the prediction of binding affinities.

Benchmarking AlphaFold-enabled molecular docking predictions for antibiotic discovery

Efficient identification of drug mechanisms of action remains a challenge. Computational docking approaches have been widely used to predict drug binding targets; yet, such approaches depend on existing protein structures, and accurate structural predictions have only recently become available from AlphaFold2. Here, we combine AlphaFold2 with molecular docking simulations to predict protein‐ligand interactions between 296 proteins spanning Escherichia coli's essential proteome, and 218 active antibacterial compounds and 100 inactive compounds, respectively, pointing to widespread compound and protein promiscuity. We benchmark model performance by measuring enzymatic activity for 12 essential proteins treated with each antibacterial compound. We confirm extensive promiscuity, but find that the average area under the receiver operating characteristic curve (auROC) is 0.48, indicating weak model performance. We demonstrate that rescoring of docking poses using machine learning‐based approaches improves model performance, resulting in average auROCs as large as 0.63, and that ensembles of rescoring functions improve prediction accuracy and the ratio of true‐positive rate to false‐positive rate. This work indicates that advances in modeling protein‐ligand interactions, particularly using machine learning‐based approaches, are needed to better harness AlphaFold2 for drug discovery.

Scoring Functions for Protein-Ligand Binding Affinity Prediction using Structure-Based Deep Learning: A Review

The rapid and accurate in silico prediction of protein-ligand binding free energies or binding affinities has the potential to transform drug discovery. In recent years, there has been a rapid growth of interest in deep learning methods for the prediction of protein-ligand binding affinities based on the structural information of protein-ligand complexes. These structure-based scoring functions often obtain better results than classical scoring functions when applied within their applicability domain. Here we review structure-based scoring functions for binding affinity prediction based on deep learning, focussing on different types of architectures, featurization strategies, data sets, methods for training and evaluation, and the role of explainable artificial intelligence in building useful models for real drug-discovery applications.

fingeRNAt—A novel tool for high-throughput analysis of nucleic acid-ligand interactions

Computational methods play a pivotal role in drug discovery and are widely applied in virtual screening, structure optimization, and compound activity profiling. Over the last decades, almost all the attention in medicinal chemistry has been directed to protein-ligand binding, and computational tools have been created with this target in mind. With novel discoveries of functional RNAs and their possible applications, RNAs have gained considerable attention as potential drug targets. However, the availability of bioinformatics tools for nucleic acids is limited. Here, we introduce fingeRNAt—a software tool for detecting non-covalent interactions formed in complexes of nucleic acids with ligands. The program detects nine types of interactions: (i) hydrogen and (ii) halogen bonds, (iii) cation-anion, (iv) pi-cation, (v) pi-anion, (vi) pi-stacking, (vii) inorganic ion-mediated, (viii) water-mediated, and (ix) lipophilic interactions. However, the scope of detected interactions can be easily expanded using a simple plugin system. In addition, detected interactions can be visualized using the associated PyMOL plugin, which facilitates the analysis of medium-throughput molecular complexes. Interactions are also encoded and stored as a bioinformatics-friendly Structural Interaction Fingerprint (SIFt)—a binary string where the respective bit in the fingerprint is set to 1 if a particular interaction is present and to 0 otherwise. This output format, in turn, enables high-throughput analysis of interaction data using data analysis techniques. We present applications of fingeRNAt-generated interaction fingerprints for visual and computational analysis of RNA-ligand complexes, including analysis of interactions formed in experimentally determined RNA-small molecule ligand complexes deposited in the Protein Data Bank. We propose interaction fingerprint-based similarity as an alternative measure to RMSD to recapitulate complexes with similar interactions but different folding. We present an application of interaction fingerprints for the clustering of molecular complexes. This approach can be used to group ligands that form similar binding networks and thus have similar biological properties. The fingeRNAt software is freely available at https://github.com/n-szulc/fingeRNAt.

Machine learning approaches for biomolecular, biophysical, and biomaterials research

A fluent conversation with a virtual assistant, person-tailored news feeds, and deep-fake images created within seconds-all those things that have been unthinkable for a long time are now a part of our everyday lives. What these examples have in common is that they are realized by different means of machine learning (ML), a technology that has fundamentally changed many aspects of the modern world. The possibility to process enormous amount of data in multi-hierarchical, digital constructs has paved the way not only for creating intelligent systems but also for obtaining surprising new insight into many scientific problems. However, in the different areas of biosciences, which typically rely heavily on the collection of time-consuming experimental data, applying ML methods is a bit more challenging: Here, difficulties can arise from small datasets and the inherent, broad variability, and complexity associated with studying biological objects and phenomena. In this Review, we give an overview of commonly used ML algorithms (which are often referred to as "machines") and learning strategies as well as their applications in different bio-disciplines such as molecular biology, drug development, biophysics, and biomaterials science. We highlight how selected research questions from those fields were successfully translated into machine readable formats, discuss typical problems that can arise in this context, and provide an overview of how to resolve those encountered difficulties.

A simple spatial extension to the extended connectivity interaction features for binding affinity prediction

The representation of the protein-ligand complexes used in building machine learning models play an important role in the accuracy of binding affinity prediction. The Extended Connectivity Interaction Features (ECIF) is one such representation. We report that (i) including the discretized distances between protein-ligand atom pairs in the ECIF scheme improves predictive accuracy, and (ii) in an evaluation using gradient boosted trees, we found that the resampling method used in selecting the best hyperparameters has a strong effect on predictive performance, especially for benchmarking purposes.

Modeling of the Crystallization Conditions for Organic Synthesis Product Purification Using Deep Learning

Crystallization is an important purification technique for solid products in a chemical laboratory. However, the correct selection of a solvent is important for the success of the procedure. In order to accelerate the solvent or solvent mixture search process, we offer an in silico alternative, i.e., a never previously demonstrated approach that can model the reaction mixture crystallization conditions which are invariant to the reaction type. The offered deep learning-based method is trained to directly predict the solvent labels used in the crystallization steps of the synthetic procedure. Our solvent label prediction task is a multi-label multi-class classification task during which the method must correctly choose one or several solvents from 13 possible examples. During the experimental investigation, we tested two multi-label classifiers (i.e., Feed-Forward and Long Short-Term Memory neural networks) applied on top of vectors. For the vectorization, we used two methods (i.e., extended-connectivity fingerprints and autoencoders) with various parameters. Our optimized technique was able to reach the accuracy of 0.870 ± 0.004 (which is 0.693 above the baseline) on the testing dataset. This allows us to assume that the proposed approach can help to accelerate manual R&D processes in chemical laboratories.

Dowker complex based machine learning (DCML) models for protein-ligand binding affinity prediction

With the great advancements in experimental data, computational power and learning algorithms, artificial intelligence (AI) based drug design has begun to gain momentum recently. AI-based drug design has great promise to revolutionize pharmaceutical industries by significantly reducing the time and cost in drug discovery processes. However, a major issue remains for all AI-based learning model that is efficient molecular representations. Here we propose Dowker complex (DC) based molecular interaction representations and Riemann Zeta function based molecular featurization, for the first time. Molecular interactions between proteins and ligands (or others) are modeled as Dowker complexes. A multiscale representation is generated by using a filtration process, during which a series of DCs are generated at different scales. Combinatorial (Hodge) Laplacian matrices are constructed from these DCs, and the Riemann zeta functions from their spectral information can be used as molecular descriptors. To validate our models, we consider protein-ligand binding affinity prediction. Our DC-based machine learning (DCML) models, in particular, DC-based gradient boosting tree (DC-GBT), are tested on three most-commonly used datasets, i.e., including PDBbind-2007, PDBbind-2013 and PDBbind-2016, and extensively compared with other existing state-of-the-art models. It has been found that our DC-based descriptors can achieve the state-of-the-art results and have better performance than all machine learning models with traditional molecular descriptors. Our Dowker complex based machine learning models can be used in other tasks in AI-based drug design and molecular data analysis.

With the ever-increasing accumulation of chemical and biomolecular data, data-driven artificial intelligence (AI) models will usher in an era of faster, cheaper and more-efficient drug design and drug discovery. However, unlike image, text, video, audio data, molecular data from chemistry and biology, have much complicated three-dimensional structures, as well as physical and chemical properties. Efficient molecular representations and descriptors are key to the success of machine learning models in drug design. Here, we propose Dowker complex based molecular representation and Riemann Zeta function based molecular featurization, for the first time. To characterize the complicated molecular structures and interactions at the atomic level, Dowker complexes are constructed. Based on them, intrinsic mathematical invariants are derived and used as molecular descriptors, which can be further combined with machine learning and deep learning models. Our model has achieved state-of-the-art results in protein-ligand binding affinity prediction, demonstrating its great potential for other drug design and discovery problems.

On the Value of Using 3D Shape and Electrostatic Similarities in Deep Generative Methods

Multiparameter optimization, the heart of drug design, is still an open challenge. Thus, improved methods for automated compound design with multiple controlled properties are desired. Here, we present a significant extension to our previously described fragment-based reinforcement learning method (DeepFMPO) for the generation of novel molecules with optimal properties. As before, the generative process outputs optimized molecules similar to the input structures, now with the improved feature of replacing parts of these molecules with fragments of similar three-dimensional (3D) shape and electrostatics. We developed and benchmarked a new python package, ESP-Sim, for the comparison of the electrostatic potential and the molecular shape, allowing the calculation of high-quality partial charges (e.g., RESP with B3LYP/6-31G**) obtained using the quantum chemistry program Psi4. By performing comparisons of 3D fragments, we can simulate 3D properties while overcoming the notoriously difficult step of accurately describing bioactive conformations. The new improved generative (DeepFMPO v3D) method is demonstrated with a scaffold-hopping exercise identifying CDK2 bioisosteres. The code is open-source and freely available.

Molecular persistent spectral image (Mol-PSI) representation for machine learning models in drug design

Artificial intelligence (AI)-based drug design has great promise to fundamentally change the landscape of the pharmaceutical industry. Even though there are great progress from handcrafted feature-based machine learning models, 3D convolutional neural networks (CNNs) and graph neural networks, effective and efficient representations that characterize the structural, physical, chemical and biological properties of molecular structures and interactions remain to be a great challenge. Here, we propose an equal-sized molecular 2D image representation, known as the molecular persistent spectral image (Mol-PSI), and combine it with CNN model for AI-based drug design. Mol-PSI provides a unique one-to-one image representation for molecular structures and interactions. In general, deep models are empowered to achieve better performance with systematically organized representations in image format. A well-designed parallel CNN architecture for adapting Mol-PSIs is developed for protein-ligand binding affinity prediction. Our results, for the three most commonly used databases, including PDBbind-v2007, PDBbind-v2013 and PDBbind-v2016, are better than all traditional machine learning models, as far as we know. Our Mol-PSI model provides a powerful molecular representation that can be widely used in AI-based drug design and molecular data analysis.

Protein-ligand binding affinity prediction based on profiles of intermolecular contacts

As a key element in structure-based drug design, binding affinity prediction (BAP) for putative protein-ligand complexes can be efficiently achieved by the incorporation of structural descriptors and machine-learning models. However, developing concise descriptors that will lead to accurate and interpretable BAP remains a difficult problem in this field. Herein, we introduce the profiles of intermolecular contacts (IMCPs) as descriptors for machine-learning-based BAP. IMCPs describe each group of protein-ligand contacts by the count and average distance of the group members, and collaborate closely with classical machine-learning models. Performed on multiple validation sets, IMCP-based models often result in better BAP accuracy than those originating from other similar descriptors. Additionally, IMCPs are simple and concise, and easy to interpret in model training. These descriptors highly conclude the structural information of protein-ligand complexes and can be easily updated with personalized profile features. IMCPs have been implemented in the BAP Toolkit on github ( https://github.com/debbydanwang/BAP).

InteractionGraphNet: A Novel and Efficient Deep Graph Representation Learning Framework for Accurate Protein-Ligand Interaction Predictions

Accurate quantification of protein-ligand interactions remains a key challenge to structure-based drug design. However, traditional machine learning (ML)-based methods based on handcrafted descriptors, one-dimensional protein sequences, and/or two-dimensional graph representations limit their capability to learn the generalized molecular interactions in 3D space. Here, we proposed a novel deep graph representation learning framework named InteractionGraphNet (IGN) to learn the protein-ligand interactions from the 3D structures of protein-ligand complexes. In IGN, two independent graph convolution modules were stacked to sequentially learn the intramolecular and intermolecular interactions, and the learned intermolecular interactions can be efficiently used for subsequent tasks. Extensive binding affinity prediction, large-scale structure-based virtual screening, and pose prediction experiments demonstrated that IGN achieved better or competitive performance against other state-of-the-art ML-based baselines and docking programs. More importantly, such state-of-the-art performance was proven from the successful learning of the key features in protein-ligand interactions instead of just memorizing certain biased patterns from data.