Deep learning for low-data drug discovery: Hurdles and opportunities.

Deep learning is becoming increasingly relevant in drug discovery, from de novo design to protein structure prediction and synthesis planning. However, it is often challenged by the small data regimes typical of certain drug discovery tasks. In such scenarios, deep learning approaches-which are notoriously 'data-hungry'-might fail to live up to their promise. Developing novel approaches to leverage the power of deep learning in low-data scenarios is sparking great attention, and future developments are expected to propel the field further. This mini-review provides an overview of recent low-data-learning approaches in drug discovery, analyzing their hurdles and advantages. Finally, we venture to provide a forecast of future research directions in low-data learning for drug discovery.

Effectiveness of molecular fingerprints for exploring the chemical space of natural products.

Natural products are a diverse class of compounds with promising biological properties, such as high potency and excellent selectivity. However, they have different structural motifs than typical drug-like compounds, e.g., a wider range of molecular weight, multiple stereocenters and higher fraction of sp3-hybridized carbons. This makes the encoding of natural products via molecular fingerprints difficult, thus restricting their use in cheminformatics studies. To tackle this issue, we explored over 30 years of research to systematically evaluate which molecular fingerprint provides the best performance on the natural product chemical space. We considered 20 molecular fingerprints from four different sources, which we then benchmarked on over 100,000 unique natural products from the COCONUT (COlleCtion of Open Natural prodUcTs) and CMNPD (Comprehensive Marine Natural Products Database) databases. Our analysis focused on the correlation between different fingerprints and their classification performance on 12 bioactivity prediction datasets. Our results show that different encodings can provide fundamentally different views of the natural product chemical space, leading to substantial differences in pairwise similarity and performance. While Extended Connectivity Fingerprints are the de-facto option to encoding drug-like compounds, other fingerprints resulted to match or outperform them for bioactivity prediction of natural products. These results highlight the need to evaluate multiple fingerprinting algorithms for optimal performance and suggest new areas of research. Finally, we provide an open-source Python package for computing all molecular fingerprints considered in the study, as well as data and scripts necessary to reproduce the results, at https://github.com/dahvida/NP_Fingerprints .

Baricitinib and tofacitinib off-target profile, with a focus on Alzheimer's disease.

INTRODUCTION: Janus kinase (JAK) inhibitors were recently identified as promising drug candidates for repurposing in Alzheimer's disease (AD) due to their capacity to suppress inflammation via modulation of JAK/STAT signaling pathways. Besides interaction with primary therapeutic targets, JAK inhibitor drugs frequently interact with unintended, often unknown, biological off-targets, leading to associated effects. Nevertheless, the relevance of JAK inhibitors' off-target interactions in the context of AD remains unclear. METHODS: Putative off-targets of baricitinib and tofacitinib were predicted using a machine learning (ML) approach. After screening scientific literature, off-targets were filtered based on their relevance to AD. Targets that had not been previously identified as off-targets of baricitinib or tofacitinib were subsequently tested using biochemical or cell-based assays. From those, active concentrations were compared to bioavailable concentrations in the brain predicted by physiologically based pharmacokinetic (PBPK) modeling. RESULTS: With the aid of ML and in vitro activity assays, we identified two enzymes previously unknown to be inhibited by baricitinib, namely casein kinase 2 subunit alpha 2 (CK2-α2) and dual leucine zipper kinase (MAP3K12), both with binding constant (K d) values of 5.8 µM. Predicted maximum concentrations of baricitinib in brain tissue using PBPK modeling range from 1.3 to 23 nM, which is two to three orders of magnitude below the corresponding binding constant. CONCLUSION: In this study, we extended the list of baricitinib off-targets that are potentially relevant for AD progression and predicted drug distribution in the brain. The results suggest a low likelihood of successful repurposing in AD due to low brain permeability, even at the maximum recommended daily dose. While additional research is needed to evaluate the potential impact of the off-target interaction on AD, the combined approach of ML-based target prediction, in vitro confirmation, and PBPK modeling may help prioritize drugs with a high likelihood of being effectively repurposed for AD. Highlights: This study explored JAK inhibitors' off-targets in AD using a multidisciplinary approach.We combined machine learning, in vitro tests, and PBPK modelling to predict and validate new off-target interactions of tofacitinib and baricitinib in AD.Previously unknown inhibition of two enzymes (CK2-a2 and MAP3K12) by baricitinib were confirmed using in vitro experiments.Our PBPK model indicates that baricitinib low brain permeability limits AD repurposing.The proposed multidisciplinary approach optimizes drug repurposing efforts in AD research.

Artificial intelligence for natural product drug discovery.

Developments in computational omics technologies have provided new means to access the hidden diversity of natural products, unearthing new potential for drug discovery. In parallel, artificial intelligence approaches such as machine learning have led to exciting developments in the computational drug design field, facilitating biological activity prediction and de novo drug design for molecular targets of interest. Here, we describe current and future synergies between these developments to effectively identify drug candidates from the plethora of molecules produced by nature. We also discuss how to address key challenges in realizing the potential of these synergies, such as the need for high-quality datasets to train deep learning algorithms and appropriate strategies for algorithm validation.

Practical guidelines for the use of gradient boosting for molecular property prediction.

Decision tree ensembles are among the most robust, high-performing and computationally efficient machine learning approaches for quantitative structure-activity relationship (QSAR) modeling. Among them, gradient boosting has recently garnered particular attention, for its performance in data science competitions, virtual screening campaigns, and bioactivity prediction. However, different variants of gradient boosting exist, the most popular being XGBoost, LightGBM and CatBoost. Our study provides the first comprehensive comparison of these approaches for QSAR. To this end, we trained 157,590 gradient boosting models, which were evaluated on 16 datasets and 94 endpoints, comprising 1.4 million compounds in total. Our results show that XGBoost generally achieves the best predictive performance, while LightGBM requires the least training time, especially for larger datasets. In terms of feature importance, the models surprisingly rank molecular features differently, reflecting differences in regularization techniques and decision tree structures. Thus, expert knowledge must always be employed when evaluating data-driven explanations of bioactivity. Furthermore, our results show that the relevance of each hyperparameter varies greatly across datasets and that it is crucial to optimize as many hyperparameters as possible to maximize the predictive performance. In conclusion, our study provides the first set of guidelines for cheminformatics practitioners to effectively train, optimize and evaluate gradient boosting models for virtual screening and QSAR applications.

De Novo Design of Nurr1 Agonists via Fragment-Augmented Generative Deep Learning in Low-Data Regime.

Generative neural networks trained on SMILES can design innovative bioactive molecules de novo. These so-called chemical language models (CLMs) have typically been trained on tens of template molecules for fine-tuning. However, it is challenging to apply CLM to orphan targets with few known ligands. We have fine-tuned a CLM with a single potent Nurr1 agonist as template in a fragment-augmented fashion and obtained novel Nurr1 agonists using sampling frequency for design prioritization. Nanomolar potency and binding affinity of the top-ranking design and its structural novelty compared to available Nurr1 ligands highlight its value as an early chemical tool and as a lead for Nurr1 agonist development, as well as the applicability of CLM in very low-data scenarios.

Engineering cytokine therapeutics.

Cytokines have pivotal roles in immunity, making them attractive as therapeutics for a variety of immune-related disorders. However, the widespread clinical use of cytokines has been limited by their short blood half-lives and severe side effects caused by low specificity and unfavourable biodistribution. Innovations in bioengineering have aided in advancing our knowledge of cytokine biology and yielded new technologies for cytokine engineering. In this Review, we discuss how the development of bioanalytical methods, such as sequencing and high-resolution imaging combined with genetic techniques, have facilitated a better understanding of cytokine biology. We then present an overview of therapeutics arising from cytokine re-engineering, targeting and delivery, mRNA therapeutics and cell therapy. We also highlight the application of these strategies to adjust the immunological imbalance in different immune-mediated disorders, including cancer, infection and autoimmune diseases. Finally, we look ahead to the hurdles that must be overcome before cytokine therapeutics can live up to their full potential.

Identification of fluorescently-barcoded nanoparticles using machine learning.

Barcoding of nano- and micro-particles allows distinguishing multiple targets at the same time within a complex mixture and is emerging as a powerful tool to increase the throughput of many assays. Fluorescent barcoding is one of the most used strategies, where microparticles are labeled with dyes and classified based on fluorescence color, intensity, or other features. Microparticles are ideal targets due to their relative ease of detection, manufacturing, and higher homogeneity. Barcoding is considerably more challenging in the case of nanoparticles (NPs), where their small size results in a lower signal and greater heterogeneity. This is a significant limitation since many bioassays require the use of nano-sized carriers. In this study, we introduce a machine-learning-assisted workflow to write, read, and classify barcoded PLGA-PEG NPs at a single-particle level. This procedure is based on the encapsulation of fluorescent markers without modifying their physicochemical properties (writing), the optimization of their confocal imaging (reading), and the implementation of a machine learning-based barcode reader (classification). We found nanoparticle heterogeneity as one of the main factors that challenges barcode separation, and that information extracted from the dyes' nanoscale confinement effects (such as Förster Resonance Energy Transfer, FRET) can aid barcode identification. Moreover, we provide a guide to reaching the optimal trade-off between the number of simultaneous barcodes and classification accuracy supporting the use of this workflow for a variety of bioassays.

Chemical language models for de novo drug design: Challenges and opportunities.

Generative deep learning is accelerating de novo drug design, by allowing the generation of molecules with desired properties on demand. Chemical language models - which generate new molecules in the form of strings using deep learning - have been particularly successful in this endeavour. Thanks to advances in natural language processing methods and interdisciplinary collaborations, chemical language models are expected to become increasingly relevant in drug discovery. This minireview provides an overview of the current state-of-the-art of chemical language models for de novo design, and analyses current limitations, challenges, and advantages. Finally, a perspective on future opportunities is provided.

Leveraging molecular structure and bioactivity with chemical language models for de novo drug design.

Generative chemical language models (CLMs) can be used for de novo molecular structure generation by learning from a textual representation of molecules. Here, we show that hybrid CLMs can additionally leverage the bioactivity information available for the training compounds. To computationally design ligands of phosphoinositide 3-kinase gamma (PI3Kγ), a collection of virtual molecules was created with a generative CLM. This virtual compound library was refined using a CLM-based classifier for bioactivity prediction. This second hybrid CLM was pretrained with patented molecular structures and fine-tuned with known PI3Kγ ligands. Several of the computer-generated molecular designs were commercially available, enabling fast prescreening and preliminary experimental validation. A new PI3Kγ ligand with sub-micromolar activity was identified, highlighting the method's scaffold-hopping potential. Chemical synthesis and biochemical testing of two of the top-ranked de novo designed molecules and their derivatives corroborated the model's ability to generate PI3Kγ ligands with medium to low nanomolar activity for hit-to-lead expansion. The most potent compounds led to pronounced inhibition of PI3K-dependent Akt phosphorylation in a medulloblastoma cell model, demonstrating efficacy of PI3Kγ ligands in PI3K/Akt pathway repression in human tumor cells. The results positively advocate hybrid CLMs for virtual compound screening and activity-focused molecular design.

Exposing the Limitations of Molecular Machine Learning with Activity Cliffs.

Machine learning has become a crucial tool in drug discovery and chemistry at large, e.g., to predict molecular properties, such as bioactivity, with high accuracy. However, activity cliffs─pairs of molecules that are highly similar in their structure but exhibit large differences in potency─have received limited attention for their effect on model performance. Not only are these edge cases informative for molecule discovery and optimization but also models that are well equipped to accurately predict the potency of activity cliffs have increased potential for prospective applications. Our work aims to fill the current knowledge gap on best-practice machine learning methods in the presence of activity cliffs. We benchmarked a total of 24 machine and deep learning approaches on curated bioactivity data from 30 macromolecular targets for their performance on activity cliff compounds. While all methods struggled in the presence of activity cliffs, machine learning approaches based on molecular descriptors outperformed more complex deep learning methods. Our findings highlight large case-by-case differences in performance, advocating for (a) the inclusion of dedicated "activity-cliff-centered" metrics during model development and evaluation and (b) the development of novel algorithms to better predict the properties of activity cliffs. To this end, the methods, metrics, and results of this study have been encapsulated into an open-access benchmarking platform named MoleculeACE (Activity Cliff Estimation, available on GitHub at: https://github.com/molML/MoleculeACE). MoleculeACE is designed to steer the community toward addressing the pressing but overlooked limitation of molecular machine learning models posed by activity cliffs.

Identification of novel off targets of baricitinib and tofacitinib by machine learning with a focus on thrombosis and viral infection.

As there are no clear on-target mechanisms that explain the increased risk for thrombosis and viral infection or reactivation associated with JAK inhibitors, the observed elevated risk may be a result of an off-target effect. Computational approaches combined with in vitro studies can be used to predict and validate the potential for an approved drug to interact with additional (often unwanted) targets and identify potential safety-related concerns. Potential off-targets of the JAK inhibitors baricitinib and tofacitinib were identified using two established machine learning approaches based on ligand similarity. The identified targets related to thrombosis or viral infection/reactivation were subsequently validated using in vitro assays. Inhibitory activity was identified for four drug-target pairs (PDE10A [baricitinib], TRPM6 [tofacitinib], PKN2 [baricitinib, tofacitinib]). Previously unknown off-target interactions of the two JAK inhibitors were identified. As the proposed pharmacological effects of these interactions include attenuation of pulmonary vascular remodeling, modulation of HCV response, and hypomagnesemia, the newly identified off-target interactions cannot explain an increased risk of thrombosis or viral infection/reactivation. While further evidence is required to explain both the elevated thrombosis and viral infection/reactivation risk, our results add to the evidence that these JAK inhibitors are promiscuous binders and highlight the potential for repurposing.

Perplexity-Based Molecule Ranking and Bias Estimation of Chemical Language Models.

Chemical language models (CLMs) can be employed to design molecules with desired properties. CLMs generate new chemical structures in the form of textual representations, such as the simplified molecular input line entry system (SMILES) strings. However, the quality of these de novo generated molecules is difficult to assess a priori. In this study, we apply the perplexity metric to determine the degree to which the molecules generated by a CLM match the desired design objectives. This model-intrinsic score allows identifying and ranking the most promising molecular designs based on the probabilities learned by the CLM. Using perplexity to compare "greedy" (beam search) with "explorative" (multinomial sampling) methods for SMILES generation, certain advantages of multinomial sampling become apparent. Additionally, perplexity scoring is performed to identify undesired model biases introduced during model training and allows the development of a new ranking system to remove those undesired biases.

De Novo Molecular Design with Chemical Language Models.

Artificial intelligence (AI) offers new possibilities for hit and lead finding in medicinal chemistry. Several instances of AI have been used for prospective de novo drug design. Among these, chemical language models have been shown to perform well in various experimental scenarios. In this study, we provide a hands-on introduction to chemical language modeling. A technique based on recurrent neural networks is discussed in detail, together with a step-by-step guide to applying this AI method for focused compound library design. The program code is freely available at URL: github.com/ETHmodlab/de_novo_design_RNN .

Beam Search for Automated Design and Scoring of Novel ROR Ligands with Machine Intelligence*.

Chemical language models enable de novo drug design without the requirement for explicit molecular construction rules. While such models have been applied to generate novel compounds with desired bioactivity, the actual prioritization and selection of the most promising computational designs remains challenging. Herein, we leveraged the probabilities learnt by chemical language models with the beam search algorithm as a model-intrinsic technique for automated molecule design and scoring. Prospective application of this method yielded novel inverse agonists of retinoic acid receptor-related orphan receptors (RORs). Each design was synthesizable in three reaction steps and presented low-micromolar to nanomolar potency towards RORγ. This model-intrinsic sampling technique eliminates the strict need for external compound scoring functions, thereby further extending the applicability of generative artificial intelligence to data-driven drug discovery.

Combining generative artificial intelligence and on-chip synthesis for de novo drug design.

Automating the molecular design-make-test-analyze cycle accelerates hit and lead finding for drug discovery. Using deep learning for molecular design and a microfluidics platform for on-chip chemical synthesis, liver X receptor (LXR) agonists were generated from scratch. The computational pipeline was tuned to explore the chemical space of known LXRα agonists and generate novel molecular candidates. To ensure compatibility with automated on-chip synthesis, the chemical space was confined to the virtual products obtainable from 17 one-step reactions. Twenty-five de novo designs were successfully synthesized in flow. In vitro screening of the crude reaction products revealed 17 (68%) hits, with up to 60-fold LXR activation. The batch resynthesis, purification, and retesting of 14 of these compounds confirmed that 12 of them were potent LXR agonists. These results support the suitability of the proposed design-make-test-analyze framework as a blueprint for automated drug design with artificial intelligence and miniaturized bench-top synthesis.