Neural-Network Scoring Functions Identify Structurally Novel Estrogen-Receptor Ligands

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.

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.

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.

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.

Structure-based virtual screening for PDL1 dimerizers: Evaluating generic scoring functions

The interaction between PD1 and its ligand PDL1 has been shown to render tumor cells resistant to apoptosis and promote tumor progression. An innovative mechanism to inhibit the PD1/PDL1 interaction is PDL1 dimerization induced by small-molecule PDL1 binders. Structure-based virtual screening is a promising approach to discovering such small-molecule PD1/PDL1 inhibitors. Here we investigate which type of generic scoring functions is most suitable to tackle this problem. We consider CNN-Score, an ensemble of convolutional neural networks, as the representative of machine-learning scoring functions. We also evaluate Smina, a commonly used classical scoring function, and IFP, a top structural fingerprint similarity scoring function. These three types of scoring functions were evaluated on two test sets sharing the same set of small-molecule PD1/PDL1 inhibitors, but using different types of inactives: either true inactives (molecules with no in vitro PD1/PDL1 inhibition activity) or assumed inactives (property-matched decoy molecules generated from each active). On both test sets, CNN-Score performed much better than Smina, which in turn strongly outperformed IFP. The fact that the latter was the case, despite precluding any possibility of exploiting decoy bias, demonstrates the predictive value of CNN-Score for PDL1. These results suggest that re-scoring Smina-docked molecules with CNN-Score is a promising structure-based virtual screening method to discover new small-molecule inhibitors of this therapeutic target.

The rise of deep learning and transformations in bioactivity prediction power of molecular modeling tools

The search and design for the better use of bioactive compounds are used in many experiments to best mimic compounds' functions in the human body. However, finding a cost-effective and timesaving approach is a top priority in different disciplines. Nowadays, artificial intelligence (AI) and particularly deep learning (DL) methods are widely applied to improve the precision and accuracy of models used in the drug discovery process. DL approaches have been used to provide more opportunities for a faster, efficient, cost-effective, and reliable computer-aided drug discovery. Moreover, the increasing biomedical data volume in areas, like genome sequences, medical images, protein structures, etc., has made data mining algorithms very important in finding novel compounds that could be drugs, uncovering or repurposing drugs and improving the area of genetic markers-based personalized medicine. Furthermore, deep neural networks (DNNs) have been demonstrated to outperform other techniques such as random forests and SVMs for QSAR studies and ligand-based virtual screening. Despite this, in QSAR studies, the quality of different data sources and potential experimental errors has greatly affected the accuracy of QSAR predictions. Therefore, further researches are still needed to improve the accuracy, selectivity, and sensitivity of the DL approach in building the best models of drug discovery.

Recent progress on the prospective application of machine learning to structure-based virtual screening

As more bioactivity and protein structure data become available, scoring functions (SFs) using machine learning (ML) to leverage these data sets continue to gain further accuracy and broader applicability. Advances in our understanding of the optimal ways to train and evaluate these ML-based SFs have introduced further improvements. One of these advances is how to select the most suitable decoys (molecules assumed inactive) to train or test an ML-based SF on a given target. We also review the latest applications of ML-based SFs for prospective structure-based virtual screening (SBVS), with a focus on the observed improvement over those using classical SFs. Finally, we provide recommendations for future prospective SBVS studies based on the findings of recent methodological studies.

Artificial intelligence in drug discovery: recent advances and future perspectives

Introduction: Artificial intelligence (AI) has inspired computer-aided drug discovery. The widespread adoption of machine learning, in particular deep learning, in multiple scientific disciplines, and the advances in computing hardware and software, among other factors, continue to fuel this development. Much of the initial skepticism regarding applications of AI in pharmaceutical discovery has started to vanish, consequently benefitting medicinal chemistry.Areas covered: The current status of AI in chemoinformatics is reviewed. The topics discussed herein include quantitative structure-activity/property relationship and structure-based modeling, de novo molecular design, and chemical synthesis prediction. Advantages and limitations of current deep learning applications are highlighted, together with a perspective on next-generation AI for drug discovery.Expert opinion: Deep learning-based approaches have only begun to address some fundamental problems in drug discovery. Certain methodological advances, such as message-passing models, spatial-symmetry-preserving networks, hybrid de novo design, and other innovative machine learning paradigms, will likely become commonplace and help address some of the most challenging questions. Open data sharing and model development will play a central role in the advancement of drug discovery with AI.

Selecting machine-learning scoring functions for structure-based virtual screening

Interest in docking technologies has grown parallel to the ever increasing number and diversity of 3D models for macromolecular therapeutic targets. Structure-Based Virtual Screening (SBVS) aims at leveraging these experimental structures to discover the necessary starting points for the drug discovery process. It is now established that Machine Learning (ML) can strongly enhance the predictive accuracy of scoring functions for SBVS by exploiting large datasets from targets, molecules and their associations. However, with greater choice, the question of which ML-based scoring function is the most suitable for prospective use on a given target has gained importance. Here we analyse two approaches to select an existing scoring function for the target along with a third approach consisting in generating a scoring function tailored to the target. These analyses required discussing the limitations of popular SBVS benchmarks, the alternatives to benchmark scoring functions for SBVS and how to generate them or use them using freely-available software.

Application of Various Molecular Modelling Methods in the Study of Estrogens and Xenoestrogens

In this review, applications of various molecular modelling methods in the study of estrogens and xenoestrogens are summarized. Selected biomolecules that are the most commonly chosen as molecular modelling objects in this field are presented. In most of the reviewed works, ligand docking using solely force field methods was performed, employing various molecular targets involved in metabolism and action of estrogens. Other molecular modelling methods such as molecular dynamics and combined quantum mechanics with molecular mechanics have also been successfully used to predict the properties of estrogens and xenoestrogens. Among published works, a great number also focused on the application of different types of quantitative structure-activity relationship (QSAR) analyses to examine estrogen's structures and activities. Although the interactions between estrogens and xenoestrogens with various proteins are the most commonly studied, other aspects such as penetration of estrogens through lipid bilayers or their ability to adsorb on different materials are also explored using theoretical calculations. Apart from molecular mechanics and statistical methods, quantum mechanics calculations are also employed in the studies of estrogens and xenoestrogens. Their applications include computation of spectroscopic properties, both vibrational and Nuclear Magnetic Resonance (NMR), and also in quantum molecular dynamics simulations and crystal structure prediction. The main aim of this review is to present the great potential and versatility of various molecular modelling methods in the studies on estrogens and xenoestrogens.

Machine learning classification can reduce false positives in structure-based virtual screening

With the recent explosion in the size of libraries available for screening, virtual screening is positioned to assume a more prominent role in early drug discovery's search for active chemical matter. In typical virtual screens, however, only about 12% of the top-scoring compounds actually show activity when tested in biochemical assays. We argue that most scoring functions used for this task have been developed with insufficient thoughtfulness into the datasets on which they are trained and tested, leading to overly simplistic models and/or overtraining. These problems are compounded in the literature because studies reporting new scoring methods have not validated their models prospectively within the same study. Here, we report a strategy for building a training dataset (D-COID) that aims to generate highly compelling decoy complexes that are individually matched to available active complexes. Using this dataset, we train a general-purpose classifier for virtual screening (vScreenML) that is built on the XGBoost framework. In retrospective benchmarks, our classifier shows outstanding performance relative to other scoring functions. In a prospective context, nearly all candidate inhibitors from a screen against acetylcholinesterase show detectable activity; beyond this, 10 of 23 compounds have IC₅₀ better than 50 μM. Without any medicinal chemistry optimization, the most potent hit has IC₅₀ 280 nM, corresponding to K_i of 173 nM. These results support using the D-COID strategy for training classifiers in other computational biology tasks, and for vScreenML in virtual screening campaigns against other protein targets. Both D-COID and vScreenML are freely distributed to facilitate such efforts.

The impact of compound library size on the performance of scoring functions for structure-based virtual screening

Larger training datasets have been shown to improve the accuracy of machine learning (ML)-based scoring functions (SFs) for structure-based virtual screening (SBVS). In addition, massive test sets for SBVS, known as ultra-large compound libraries, have been demonstrated to enable the fast discovery of selective drug leads with low-nanomolar potency. This proof-of-concept was carried out on two targets using a single docking tool along with its SF. It is thus unclear whether this high level of performance would generalise to other targets, docking tools and SFs. We found that screening a larger compound library results in more potent actives being identified in all six additional targets using a different docking tool along with its classical SF. Furthermore, we established that a way to improve the potency of the retrieved molecules further is to rank them with more accurate ML-based SFs (we found this to be true in four of the six targets; the difference was not significant in the remaining two targets). A 3-fold increase in average hit rate across targets was also achieved by the ML-based SFs. Lastly, we observed that classical and ML-based SFs often find different actives, which supports using both types of SFs on those targets.

Computer-Aided Ligand Discovery for Estrogen Receptor Alpha

Breast cancer (BCa) is one of the most predominantly diagnosed cancers in women. Notably, 70% of BCa diagnoses are Estrogen Receptor α positive (ERα+) making it a critical therapeutic target. With that, the two subtypes of ER, ERα and ERβ, have contrasting effects on BCa cells. While ERα promotes cancerous activities, ERβ isoform exhibits inhibitory effects on the same. ER-directed small molecule drug discovery for BCa has provided the FDA approved drugs tamoxifen, toremifene, raloxifene and fulvestrant that all bind to the estrogen binding site of the receptor. These ER-directed inhibitors are non-selective in nature and may eventually induce resistance in BCa cells as well as increase the risk of endometrial cancer development. Thus, there is an urgent need to develop novel drugs with alternative ERα targeting mechanisms that can overcome the limitations of conventional anti-ERα therapies. Several functional sites on ERα, such as Activation Function-2 (AF2), DNA binding domain (DBD), and F-domain, have been recently considered as potential targets in the context of drug research and discovery. In this review, we summarize methods of computer-aided drug design (CADD) that have been employed to analyze and explore potential targetable sites on ERα, discuss recent advancement of ERα inhibitor development, and highlight the potential opportunities and challenges of future ERα-directed drug discovery.

Beware of the generic machine learning-based scoring functions in structure-based virtual screening

Machine learning-based scoring functions (MLSFs) have attracted extensive attention recently and are expected to be potential rescoring tools for structure-based virtual screening (SBVS). However, a major concern nowadays is whether MLSFs trained for generic uses rather than a given target can consistently be applicable for VS. In this study, a systematic assessment was carried out to re-evaluate the effectiveness of 14 reported MLSFs in VS. Overall, most of these MLSFs could hardly achieve satisfactory results for any dataset, and they could even not outperform the baseline of classical SFs such as Glide SP. An exception was observed for RFscore-VS trained on the Directory of Useful Decoys-Enhanced dataset, which showed its superiority for most targets. However, in most cases, it clearly illustrated rather limited performance on the targets that were dissimilar to the proteins in the corresponding training sets. We also used the top three docking poses rather than the top one for rescoring and retrained the models with the updated versions of the training set, but only minor improvements were observed. Taken together, generic MLSFs may have poor generalization capabilities to be applicable for the real VS campaigns. Therefore, it should be quite cautious to use this type of methods for VS.

Key Topics in Molecular Docking for Drug Design

Molecular docking has been widely employed as a fast and inexpensive technique in the past decades, both in academic and industrial settings. Although this discipline has now had enough time to consolidate, many aspects remain challenging and there is still not a straightforward and accurate route to readily pinpoint true ligands among a set of molecules, nor to identify with precision the correct ligand conformation within the binding pocket of a given target molecule. Nevertheless, new approaches continue to be developed and the volume of published works grows at a rapid pace. In this review, we present an overview of the method and attempt to summarise recent developments regarding four main aspects of molecular docking approaches: (i) the available benchmarking sets, highlighting their advantages and caveats, (ii) the advances in consensus methods, (iii) recent algorithms and applications using fragment-based approaches, and (iv) the use of machine learning algorithms in molecular docking. These recent developments incrementally contribute to an increase in accuracy and are expected, given time, and together with advances in computing power and hardware capability, to eventually accomplish the full potential of this area.

Crystal structure of 1-benzyl-3-cyano-6-phenyl-1,2-dihydropyridine, C19H16N2

C19H16N2, orthorhombic, P212121 (no. 19), a = 7.5177(10) Å, b = 12.4761(16) Å, c = 16.118(2) Å, V = 1511.7(3) Å3, Z = 4, R gt(F) = 0.0424, wR ref(F 2) = 0.0967, T = 293(2) K.

Classical scoring functions for docking are unable to exploit large volumes of structural and interaction data

Motivation

Studies have shown that the accuracy of random forest (RF)-based scoring functions (SFs), such as RF-Score-v3, increases with more training samples, whereas that of classical SFs, such as X-Score, does not. Nevertheless, the impact of the similarity between training and test samples on this matter has not been studied in a systematic manner. It is therefore unclear how these SFs would perform when only trained on protein-ligand complexes that are highly dissimilar or highly similar to the test set. It is also unclear whether SFs based on machine learning algorithms other than RF can also improve accuracy with increasing training set size and to what extent they learn from dissimilar or similar training complexes.

Results

We present a systematic study to investigate how the accuracy of classical and machine-learning SFs varies with protein-ligand complex similarities between training and test sets. We considered three types of similarity metrics, based on the comparison of either protein structures, protein sequences or ligand structures. Regardless of the similarity metric, we found that incorporating a larger proportion of similar complexes to the training set did not make classical SFs more accurate. In contrast, RF-Score-v3 was able to outperform X-Score even when trained on just 32% of the most dissimilar complexes, showing that its superior performance owes considerably to learning from dissimilar training complexes to those in the test set. In addition, we generated the first SF employing Extreme Gradient Boosting (XGBoost), XGB-Score, and observed that it also improves with training set size while outperforming the rest of SFs. Given the continuous growth of training datasets, the development of machine-learning SFs has become very appealing.

Availability and implementation

https://github.com/HongjianLi/MLSF

Supplementary information

Supplementary data are available at Bioinformatics online.

Improving small molecule virtual screening strategies for the next generation of therapeutics

The new generation of post-genomic targets, such as protein-protein interactions (PPIs), often require new chemotypes not well represented in current compound libraries. This is one reason for why traditional high throughput screening (HTS) approaches are not more successful in delivering medicinal chemistry starting points for PPIs. In silico screening methods of an expanded chemical space are then potential alternatives for developing novel chemical probes to modulate PPIs. In this review, we report on the state-of-the-art pipelines for virtual screening, emphasizing prospectively validated methods capable of addressing the challenge of drugging difficult targets in the human interactome. Collectively, we show that optimal strategies for structure based virtual screening vary depending on receptor structure and degree of flexibility.

Protein-Ligand Scoring with Convolutional Neural Networks

Computational approaches to drug discovery can reduce the time and cost associated with experimental assays and enable the screening of novel chemotypes. Structure-based drug design methods rely on scoring functions to rank and predict binding affinities and poses. The ever-expanding amount of protein-ligand binding and structural data enables the use of deep machine learning techniques for protein-ligand scoring. We describe convolutional neural network (CNN) scoring functions that take as input a comprehensive three-dimensional (3D) representation of a protein-ligand interaction. A CNN scoring function automatically learns the key features of protein-ligand interactions that correlate with binding. We train and optimize our CNN scoring functions to discriminate between correct and incorrect binding poses and known binders and nonbinders. We find that our CNN scoring function outperforms the AutoDock Vina scoring function when ranking poses both for pose prediction and virtual screening.