Adding Stochastic Negative Examples into Machine Learning Improves Molecular Bioactivity Prediction

Developing muscarinic receptor M1 classification models utilizing transfer learning and generative AI techniques

Muscarinic receptor subtype 1 (M1) is a G protein-coupled receptor (GPCR) and a key pharmacological target for peripheral neuropathy, chronic obstructive pulmonary disease, nerve agent exposures, and cognitive disorders. Screening and identifying compounds with potential to interact with M1 will aid in rational drug design for these disorders. In this work, we developed machine learning-based M1 classification models utilizing publicly available bioactivity data. As inactive compounds are rarely reported in the literature, we encountered the problem of imbalanced datasets. We investigated two strategies to overcome this bottleneck: 1) transfer learning and 2) using generative models to oversample the inactive class. Our analysis shows that these approaches reduced misclassification of the inactive class not only for M1 but also for other GPCR targets. Overall, we have developed classification models for M1 receptor that will enable rapid screening of large chemical databases and advance drug discovery.

Progress of machine learning in the application of small molecule druggability prediction

Machine learning (ML) has become an important tool for predicting the pharmaceutical properties of small molecules. Recent advancements in ML algorithms enable the rapid and accurate evaluation of solubility, activity, toxicity, pharmacokinetics, and other molecular properties through ML-based models. By conducting virtual screening of drug targets and elucidating drug-target protein interactions, researchers can conduct preliminary evaluations of the activity and safety of compounds from the ultra-large drug compound libraries, thereby accelerating the screening process for lead compounds. Moreover, ML leverages existing experimental data to train and generate new datasets, addressing the challenge of limited compounds and protein target data. This review provided a concise overview of ML applications in predicting small molecule properties, focusing on model construction principles, molecular feature selection, and other essential aspects. It also discussed the potential applications of ML in the screening of pharmaceutical small molecules.

One size does not fit all: revising traditional paradigms for assessing accuracy of QSAR models used for virtual screening

Traditional best practices for quantitative structure activity relationship (QSAR) modeling recommend dataset balancing and balanced accuracy (BA) as the key desired objective of model development. This study explores the value of the conventional norms in the context of using QSAR models for virtual screening of modern large and ultra-large chemical libraries. For this increasingly common task, we now recommend the use of models with the highest positive predictive value (PPV) built on imbalanced training sets as preferred virtual screening tools. This recommendation stems from practical considerations of how the results of virtual screening are used in experimental laboratories where only a small fraction of virtually screened molecules can be tested using standard well plates. As a proof of concept, we have developed QSAR models for five expansive datasets with different ratios of active and inactive molecules and compared model performance in virtual screening using BA, PPV, and other metrics. We show that training on imbalanced datasets achieves a hit rate at least 30% higher than using balanced datasets, and that the PPV metric captured this difference of performance with no parameter tuning. Importantly, hit rates were estimated for top scoring compounds organized in batches of the size of plates (for instance, 128 molecules) used in the experimental high throughput screening. Based on the results of our studies, we posit that QSAR models trained on imbalanced datasets with the highest PPV should be relied upon to identify and test hit compounds in early drug discovery studies.

Advancing Bioactivity Prediction Through Molecular Docking and Self-Attention

Bioactivity refers to the ability of a substance to induce biological effects within living systems, often describing the influence of molecules, drugs, or chemicals on organisms. In drug discovery, predicting bioactivity streamlines early-stage candidate screening by swiftly identifying potential active molecules. The popular deep learning methods in bioactivity prediction primarily model the ligand structure-bioactivity relationship under the premise of Quantitative Structure-Activity Relationship (QSAR). However, bioactivity is determined by multiple factors, including not only the ligand structure but also drug-target interactions, signaling pathways, reaction environments, pharmacokinetic properties, and species differences. Our study first integrates drug-target interactions into bioactivity prediction using protein-ligand complex data from molecular docking. We devise a Drug-Target Interaction Graph Neural Network (DTIGN), infusing interatomic forces into intermolecular graphs. DTIGN employs multi-head self-attention to identify native-like binding pockets and poses within molecular docking results. To validate the fidelity of the self-attention mechanism, we gather ground truth data from crystal structure databases. Subsequently, we employ these limited native structures to refine bioactivity prediction via semi-supervised learning. For this study, we establish a unique benchmark dataset for evaluating bioactivity prediction models in the context of protein-ligand complexes, showcasing the superior performance of our method (with an average improvement of 27.03%) through comparison with 9 leading deep learning-based bioactivity prediction methods.

Unleashing the power of generative AI in drug discovery

Artificial intelligence (AI) is revolutionizing drug discovery by enhancing precision, reducing timelines and costs, and enabling AI-driven computer-aided drug design. This review focuses on recent advancements in deep generative models (DGMs) for de novo drug design, exploring diverse algorithms and their profound impact. It critically analyses the challenges that are intricately interwoven into these technologies, proposing strategies to unlock their full potential. It features case studies of both successes and failures in advancing drugs to clinical trials with AI assistance. Last, it outlines a forward-looking plan for optimizing DGMs in de novo drug design, thereby fostering faster and more cost-effective drug development.

BigBind: Learning from Nonstructural Data for Structure-Based Virtual Screening

Deep learning methods that predict protein-ligand binding have recently been used for structure-based virtual screening. Many such models have been trained using protein-ligand complexes with known crystal structures and activities from the PDBBind data set. However, because PDBbind only includes 20K complexes, models typically fail to generalize to new targets, and model performance is on par with models trained with only ligand information. Conversely, the ChEMBL database contains a wealth of chemical activity information but includes no information about binding poses. We introduce BigBind, a data set that maps ChEMBL activity data to proteins from the CrossDocked data set. BigBind comprises 583 K ligand activities and includes 3D structures of the protein binding pockets. Additionally, we augmented the data by adding an equal number of putative inactives for each target. Using this data, we developed Banana (basic neural network for binding affinity), a neural network-based model to classify active from inactive compounds, defined by a 10 μM cutoff. Our model achieved an AUC of 0.72 on BigBind's test set, while a ligand-only model achieved an AUC of 0.59. Furthermore, Banana achieved competitive performance on the LIT-PCBA benchmark (median EF1% 1.81) while running 16,000 times faster than molecular docking with Gnina. We suggest that Banana, as well as other models trained on this data set, will significantly improve the outcomes of prospective virtual screening tasks.

Advances and applications of machine learning and deep learning in environmental ecology and health

Machine learning (ML) and deep learning (DL) possess excellent advantages in data analysis (e.g., feature extraction, clustering, classification, regression, image recognition and prediction) and risk assessment and management in environmental ecology and health (EEH). Considering the rapid growth and increasing complexity of data in EEH, it is of significance to summarize recent advances and applications of ML and DL in EEH. This review summarized the basic processes and fundamental algorithms of the ML and DL modeling, and indicated the urgent needs of ML and DL in EEH. Recent research hotspots such as environmental ecology and restoration, environmental fate of new pollutants, chemical exposures and risks, chemical hazard identification and control were highlighted. Various applications of ML and DL in EEH demonstrate their versatility and technological revolution, and present some challenges. The perspective of ML and DL in EEH were further outlined to promote the innovative analysis and cultivation of the ML-driven research paradigm.

BiComp-DTA: Drug-target binding affinity prediction through complementary biological-related and compression-based featurization approach

Drug-target binding affinity prediction plays a key role in the early stage of drug discovery. Numerous experimental and data-driven approaches have been developed for predicting drug-target binding affinity. However, experimental methods highly rely on the limited structural-related information from drug-target pairs, domain knowledge, and time-consuming assays. On the other hand, learning-based methods have shown an acceptable prediction performance. However, most of them utilize several simple and complex types of proteins and drug compounds data, ranging from the protein sequences to the topology of a graph representation of drug compounds, employing multiple deep neural networks for encoding and feature extraction, and so, leads to the computational overheads. In this study, we propose a unified measure for protein sequence encoding, named BiComp, which provides compression-based and evolutionary-related features from the protein sequences. Specifically, we employ Normalized Compression Distance and Smith-Waterman measures for capturing complementary information from the algorithmic information theory and biological domains, respectively. We utilize the proposed measure to encode the input proteins feeding a new deep neural network-based method for drug-target binding affinity prediction, named BiComp-DTA. BiComp-DTA is evaluated utilizing four benchmark datasets for drug-target binding affinity prediction. Compared to the state-of-the-art methods, which employ complex models for protein encoding and feature extraction, BiComp-DTA provides superior efficiency in terms of accuracy, runtime, and the number of trainable parameters. The latter achievement facilitates execution of BiComp-DTA on a normal desktop computer in a fast fashion. As a comparative study, we evaluate BiComp's efficiency against its components for drug-target binding affinity prediction. The results have shown superior accuracy of BiComp due to the orthogonality and complementary nature of Smith-Waterman and Normalized Compression Distance measures for protein sequences. Such a protein sequence encoding provides efficient representation with no need for multiple sources of information, deep domain knowledge, and complex neural networks.

Yes SIR! On the structure-inactivity relationships in drug discovery

In analogy with structure-activity relationships (SARs), which are at the core of medicinal chemistry, studying structure-inactivity relationships (SIRs) is essential to understanding and predicting biological activity. Current computational methods should predict or distinguish 'activity' and 'inactivity' with the same confidence because both concepts are complementary. However, the lack of inactivity data, in particular in the public domain, limits the development of predictive models and its broad application. In this review, we encourage the scientific community to disclose and analyze high-confidence activity data considering both the labeled 'active' and 'inactive' compounds.

Machine Learning for the Discovery, Design, and Engineering of Materials

Machine learning (ML) has become a part of the fabric of high-throughput screening and computational discovery of materials. Despite its increasingly central role, challenges remain in fully realizing the promise of ML. This is especially true for the practical acceleration of the engineering of robust materials and the development of design strategies that surpass trial and error or high-throughput screening alone. Depending on the quantity being predicted and the experimental data available, ML can either outperform physics-based modes, be used to accelerate such models, or be integrated with them to improve their performance. We cover recent advances in algorithms and in their application that are starting to make inroads toward (a) the discovery of new materials through large-scale enumerative screening, (b) the design of materials through identification of rules and principles that govern materials properties, and (c) the engineering of practical materials by satisfying multiple objectives. We conclude with opportunities for further advancement to realize ML as a widespread tool for practical computational materials design. Expected final online publication date for the Annual Review of Chemical and Biomolecular Engineering, Volume 13 is October 2022. Please see http://www.annualreviews.org/page/journal/pubdates for revised estimates.

Comparing classification models-a practical tutorial

While machine learning models have become a mainstay in Cheminformatics, the field has yet to agree on standards for model evaluation and comparison. In many cases, authors compare methods by performing multiple folds of cross-validation and reporting the mean value for an evaluation metric such as the area under the receiver operating characteristic. These comparisons of mean values often lack statistical rigor and can lead to inaccurate conclusions. In the interest of encouraging best practices, this tutorial provides an example of how multiple methods can be compared in a statistically rigorous fashion.

Computational Discovery of Transition-metal Complexes: From High-throughput Screening to Machine Learning

Transition-metal complexes are attractive targets for the design of catalysts and functional materials. The behavior of the metal-organic bond, while very tunable for achieving target properties, is challenging to predict and necessitates searching a wide and complex space to identify needles in haystacks for target applications. This review will focus on the techniques that make high-throughput search of transition-metal chemical space feasible for the discovery of complexes with desirable properties. The review will cover the development, promise, and limitations of "traditional" computational chemistry (i.e., force field, semiempirical, and density functional theory methods) as it pertains to data generation for inorganic molecular discovery. The review will also discuss the opportunities and limitations in leveraging experimental data sources. We will focus on how advances in statistical modeling, artificial intelligence, multiobjective optimization, and automation accelerate discovery of lead compounds and design rules. The overall objective of this review is to showcase how bringing together advances from diverse areas of computational chemistry and computer science have enabled the rapid uncovering of structure-property relationships in transition-metal chemistry. We aim to highlight how unique considerations in motifs of metal-organic bonding (e.g., variable spin and oxidation state, and bonding strength/nature) set them and their discovery apart from more commonly considered organic molecules. We will also highlight how uncertainty and relative data scarcity in transition-metal chemistry motivate specific developments in machine learning representations, model training, and in computational chemistry. Finally, we will conclude with an outlook of areas of opportunity for the accelerated discovery of transition-metal complexes.

Putting Density Functional Theory to the Test in Machine-Learning-Accelerated Materials Discovery

Accelerated discovery with machine learning (ML) has begun to provide the advances in efficiency needed to overcome the combinatorial challenge of computational materials design. Nevertheless, ML-accelerated discovery both inherits the biases of training data derived from density functional theory (DFT) and leads to many attempted calculations that are doomed to fail. Many compelling functional materials and catalytic processes involve strained chemical bonds, open-shell radicals and diradicals, or metal-organic bonds to open-shell transition-metal centers. Although promising targets, these materials present unique challenges for electronic structure methods and combinatorial challenges for their discovery. In this Perspective, we describe the advances needed in accuracy, efficiency, and approach beyond what is typical in conventional DFT-based ML workflows. These challenges have begun to be addressed through ML models trained to predict the results of multiple methods or the differences between them, enabling quantitative sensitivity analysis. For DFT to be trusted for a given data point in a high-throughput screen, it must pass a series of tests. ML models that predict the likelihood of calculation success and detect the presence of strong correlation will enable rapid diagnoses and adaptation strategies. These "decision engines" represent the first steps toward autonomous workflows that avoid the need for expert determination of the robustness of DFT-based materials discoveries.