Reliable method for predicting the binding affinity of RNA-small molecule interactions using machine learning.

Ribonucleic acids (RNAs) play important roles in cellular regulation. Consequently, dysregulation of both coding and non-coding RNAs has been implicated in several disease conditions in the human body. In this regard, a growing interest has been observed to probe into the potential of RNAs to act as drug targets in disease conditions. To accelerate this search for disease-associated novel RNA targets and their small molecular inhibitors, machine learning models for binding affinity prediction were developed specific to six RNA subtypes namely, aptamers, miRNAs, repeats, ribosomal RNAs, riboswitches and viral RNAs. We found that differences in RNA sequence composition, flexibility and polar nature of RNA-binding ligands are important for predicting the binding affinity. Our method showed an average Pearson correlation (r) of 0.83 and a mean absolute error of 0.66 upon evaluation using the jack-knife test, indicating their reliability despite the low amount of data available for several RNA subtypes. Further, the models were validated with external blind test datasets, which outperform other existing quantitative structure-activity relationship (QSAR) models. We have developed a web server to host the models, RNA-Small molecule binding Affinity Predictor, which is freely available at: https://web.iitm.ac.in/bioinfo2/RSAPred/.

Modeling interactions between Heparan sulfate and proteins based on the Heparan sulfate microarray analysis.

Heparan sulfate (HS), a sulfated polysaccharide abundant in the extracellular matrix, plays pivotal roles in various physiological and pathological processes by interacting with proteins. Investigating the binding selectivity of HS oligosaccharides to target proteins is essential, but the exhaustive inclusion of all possible oligosaccharides in microarray experiments is impractical. To address this challenge, we present a hybrid pipeline that integrates microarray and in silico techniques to design oligosaccharides with desired protein affinity. Using fibroblast growth factor 2 (FGF2) as a model protein, we assembled an in-house dataset of HS oligosaccharides on microarrays and developed two structural representations: a standard representation with all atoms explicit and a simplified representation with disaccharide units as "quasi-atoms." Predictive Quantitative Structure-Activity Relationship (QSAR) models for FGF2 affinity were developed using the Random Forest (RF) algorithm. The resulting models, considering the applicability domain, demonstrated high predictivity, with a correct classification rate of 0.81-0.80 and improved positive predictive values (PPV) up to 0.95. Virtual screening of 40 new oligosaccharides using the simplified model identified 15 computational hits, 11 of which were experimentally validated for high FGF2 affinity. This hybrid approach marks a significant step toward the targeted design of oligosaccharides with desired protein interactions, providing a foundation for broader applications in glycobiology.

Predicting anti-trypanosome effect of carbazole-derived compounds by powerful SVM with novel kernel function and comprehensive learning PSO.

In order to predict the anti-trypanosome effect of carbazole-derived compounds by quantitative structure-activity relationship, five models were established by the linear method, random forest, radial basis kernel function support vector machine, linear combination mix-kernel function support vector machine, and nonlinear combination mix-kernel function support vector machine (NLMIX-SVM). The heuristic method and optimized CatBoost were used to select two different key descriptor sets for building linear and nonlinear models, respectively. Hyperparameters in all nonlinear models were optimized by comprehensive learning particle swarm optimization with low complexity and fast convergence. Furthermore, the models' robustness and reliability underwent rigorous assessment using fivefold and leave-one-out cross-validation, y-randomization, and statistics including concordance correlation coefficient (CCC), [Formula: see text] , [Formula: see text] , and [Formula: see text] . Among all the models, the NLMIX-SVM model, which was established by support vector regression using a nonlinear combination of radial basis kernel function, sigmoid kernel function, and linear kernel function as a new kernel function, demonstrated excellent learning and generalization abilities as well as robustness: [Formula: see text] = 0.9581, mean square error (MSE) = 0.0199 for the training set and [Formula: see text] = 0.9528, MSE = 0.0174 for the test set. [Formula: see text] , [Formula: see text] , CCC, [Formula: see text] , [Formula: see text], and [Formula: see text] are 0.9539, 0.8908, 0.9752, 0.9529, 0.9528, and 0.9633, respectively. The NLMIX-SVM method proved to be a promising way in quantitative structure-activity relationship research. In addition, molecular docking experiments were conducted to analyze the properties of new derivatives, and a new potential candidate drug molecule was ultimately found. In summary, this study will provide help for the design and screening of novel anti-trypanosome drugs.

Toward Nano-Specific In Silico NAMs: How to Adjust Nano-QSAR to the Recent Advancements of Nanotoxicology?

The rapid development of engineered nanomaterials (ENMs) causes humans to become increasingly exposed to them. Therefore, a better understanding of the health impact of ENMs is highly demanded. Considering the 3Rs (Replacement, Reduction, and Refinement) principle, in vitro and computational methods are excellent alternatives for testing on animals. Among computational methods, nano-quantitative structure-activity relationship (nano-QSAR), which links the physicochemical and structural properties of EMNs with biological activities, is one of the leading method. The nature of toxicological experiments has evolved over the last decades; currently, one experiment can provide thousands of measurements of the organism's functioning at the molecular level. At the same time, the capacity of the in vitro systems to mimic the human organism is also improving significantly. Hence, the authors would like to discuss whether the nano-QSAR approach follows modern toxicological studies and takes full advantage of the opportunities offered by modern toxicological platforms. Challenges and possibilities for improving data integration are underlined narratively, including the need for a consensus built between the in vitro and the QSAR domains.

deepGraphh: AI-driven web service for graph-based quantitative structure-activity relationship analysis.

Artificial intelligence (AI)-based computational techniques allow rapid exploration of the chemical space. However, representation of the compounds into computational-compatible and detailed features is one of the crucial steps for quantitative structure-activity relationship (QSAR) analysis. Recently, graph-based methods are emerging as a powerful alternative to chemistry-restricted fingerprints or descriptors for modeling. Although graph-based modeling offers multiple advantages, its implementation demands in-depth domain knowledge and programming skills. Here we introduce deepGraphh, an end-to-end web service featuring a conglomerate of established graph-based methods for model generation for classification or regression tasks. The graphical user interface of deepGraphh supports highly configurable parameter support for model parameter tuning, model generation, cross-validation and testing of the user-supplied query molecules. deepGraphh supports four widely adopted methods for QSAR analysis, namely, graph convolution network, graph attention network, directed acyclic graph and Attentive FP. Comparative analysis revealed that deepGraphh supported methods are comparable to the descriptors-based machine learning techniques. Finally, we used deepGraphh models to predict the blood-brain barrier permeability of human and microbiome-generated metabolites. In summary, deepGraphh offers a one-stop web service for graph-based methods for chemoinformatics.

Using molecular embeddings in QSAR modeling: does it make a difference?

With the consolidation of deep learning in drug discovery, several novel algorithms for learning molecular representations have been proposed. Despite the interest of the community in developing new methods for learning molecular embeddings and their theoretical benefits, comparing molecular embeddings with each other and with traditional representations is not straightforward, which in turn hinders the process of choosing a suitable representation for Quantitative Structure-Activity Relationship (QSAR) modeling. A reason behind this issue is the difficulty of conducting a fair and thorough comparison of the different existing embedding approaches, which requires numerous experiments on various datasets and training scenarios. To close this gap, we reviewed the literature on methods for molecular embeddings and reproduced three unsupervised and two supervised molecular embedding techniques recently proposed in the literature. We compared these five methods concerning their performance in QSAR scenarios using different classification and regression datasets. We also compared these representations to traditional molecular representations, namely molecular descriptors and fingerprints. As opposed to the expected outcome, our experimental setup consisting of over $25 000$ trained models and statistical tests revealed that the predictive performance using molecular embeddings did not significantly surpass that of traditional representations. Although supervised embeddings yielded competitive results compared with those using traditional molecular representations, unsupervised embeddings tended to perform worse than traditional representations. Our results highlight the need for conducting a careful comparison and analysis of the different embedding techniques prior to using them in drug design tasks and motivate a discussion about the potential of molecular embeddings in computer-aided drug design.

Chemical Predictive Modelling and Natural Product-based Divergent Synthesis - Design of Type B PPAPs with Nanomolar Activities against MRSA.

In response to the pressing global challenge of antibiotic resistance, time efficient design and synthesis of novel antibiotics are of immense need. Polycyclic polyprenylated acylphloroglucinols (PPAP) were previously reported to effectively combat a range of gram-positive bacteria. Although the exact mode of action is still not clear, we conceptualized a late-stage divergent synthesis approach to expand our natural product-based PPAP library by 30 additional entities to perform SAR studies against methicillin-resistant Staphylococcus aureus (MRSA). Although at this point only data from cellular assays are available and understanding of molecular drug-target interactions are lacking, the experimental data were used to generate 3D-QSAR models via an artificial intelligence training and to identify a common pharmacophore model. The experimentally validated QSAR model enabled the estimation of anti-MRSA activities of a virtual compound library consisting of more than 100,000 in-silico generated B PPAPs, out of which the 20 most promising candidates were synthesized. These novel PPAPs revealed significantly improved cellular activities against MRSA with growth inhibition down to concentrations less than 1 µm.

In Silico Prediction of Skin Sensitization for Compounds via Flexible Evidence Combination Based on Machine Learning and Dempster-Shafer Theory.

Skin sensitization is increasingly becoming a significant concern in the development of drugs and cosmetics due to consumer safety and occupational health problems. In silico methods have emerged as alternatives to traditional in vivo animal testing due to ethical and economic considerations. In this study, machine learning methods were used to build quantitative structure-activity relationship (QSAR) models on five skin sensitization data sets (GPMT, LLNA, DPRA, KeratinoSens, and h-CLAT), achieving effective predictive accuracies (correct classification rates of 0.688-0.764 on test sets). To address the complex mechanisms of human skin sensitization, the Dempster-Shafer theory was applied to merge multiple QSAR models, resulting in an evidence-based integrated decision model. Various evidence combinations and combination rules were explored, with the self-defined Q3 rule showing superior balance. The combination of evidence such as GPMT and KeratinoSens and h-CLAT achieved a correct classification rate (CCR) of 0.880 and coverage of 0.893 while maintaining the competitiveness of other combinations. Additionally, the Shapley additive explanations (SHAP) method was used to interpret important features and substructures related to skin sensitization. A comparative analysis of an external human test set demonstrated the superior performance of the proposed method. Finally, to enhance accessibility, the workflow was implemented into a user-friendly software named HSkinSensDS.

Enhancing hERG Risk Assessment with Interpretable Classificatory and Regression Models.

The human Ether-à-go-go-Related Gene (hERG) is a transmembrane protein that regulates cardiac action potential, and its inhibition can induce a potentially deadly cardiac syndrome. In vitro tests help identify hERG blockers at early stages; however, the high cost motivates searching for alternative, cost-effective methods. The primary goal of this study was to enhance the Pred-hERG tool for predicting hERG blockage. To achieve this, we developed new QSAR models that incorporated additional data, updated existing classificatory and multiclassificatory models, and introduced new regression models. Notably, we integrated SHAP (SHapley Additive exPlanations) values to offer a visual interpretation of these models. Utilizing the latest data from ChEMBL v30, encompassing over 14,364 compounds with hERG data, our binary and multiclassification models outperformed both the previous iteration of Pred-hERG and all publicly available models. Notably, the new version of our tool introduces a regression model for predicting hERG activity (pIC50). The optimal model demonstrated an R2 of 0.61 and an RMSE of 0.48, surpassing the only available regression model in the literature. Pred-hERG 5.0 now offers users a swift, reliable, and user-friendly platform for the early assessment of chemically induced cardiotoxicity through hERG blockage. The tool provides versatile outcomes, including (i) classificatory predictions of hERG blockage with prediction reliability, (ii) multiclassificatory predictions of hERG blockage with reliability, (iii) regression predictions with estimated pIC50 values, and (iv) probability maps illustrating the contribution of chemical fragments for each prediction. Furthermore, we implemented explainable AI analysis (XAI) to visualize SHAP values, providing insights into the contribution of each feature to binary classification predictions. A consensus prediction calculated based on the predictions of the three developed models is also present to assist the user's decision-making process. Pred-hERG 5.0 has been designed to be user-friendly, making it accessible to users without computational or programming expertise. The tool is freely available at http://predherg.labmol.com.br.

Hybrid 2D/3D-quantitative structure-activity relationship studies on the bioactivities and molecular mechanism of antibacterial peptides.

Antimicrobial peptide (AMP) is the polypeptide, which protects the organism avoiding attack from pathogenic bacteria. Studies have shown that there were some antimicrobial peptides with molecular action mechanism involved in crossing the cell membrane without inducing severe membrane collapse, then interacting with cytoplasmic target-nucleic acid, and exerting antibacterial activity by interfacing the transmission of genetic information of pathogenic microorganisms. However, the relationship between the antibacterial activities and peptide structures was still unclear. Therefore, in the present work, a series of AMPs with a sequence of 20 amino acids was extracted from DBAASP database, then, quantitative structure-activity relationship (QSAR) methods were conducted on these peptides. In addition, novel antimicrobial peptides with  stronger antimicrobial activities were designed according to the information originated from the constructed models. Hence, the outcome of this study would lay a solid foundation for the in-silico design and exploration of novel antibacterial peptides with improved activity activities.

Adapting Deep Learning QSPR Models to Specific Drug Discovery Projects.

Medicinal chemistry and drug design efforts can be assisted by machine learning (ML) models that relate the molecular structure to compound properties. Such quantitative structure-property relationship models are generally trained on large data sets that include diverse chemical series (global models). In the pharmaceutical industry, these ML global models are available across discovery projects as an "out-of-the-box" solution to assist in drug design, synthesis prioritization, and experiment selection. However, drug discovery projects typically focus on confined parts of the chemical space (e.g., chemical series), where global models might not be applicable. Local ML models are sometimes generated to focus on specific projects or series. Herein, ML-based global models, local models, and hybrid global-local strategies were benchmarked. Analyses were done for more than 300 drug discovery projects at Novartis and ten absorption, distribution, metabolism, and excretion (ADME) assays. In this work, hybrid global-local strategies based on transfer learning approaches were proposed to leverage both historical ADME data (global) and project-specific data (local) to adapt model predictions. Fine-tuning a pretrained global ML model (used for weights' initialization, WI) was the top-performing method. Average improvements of mean absolute errors across all assays were 16% and 27% compared with global and local models, respectively. Interestingly, when the effect of training set size was analyzed, WI fine-tuning was found to be successful even in low-data scenarios (e.g., ∼10 molecules per project). Taken together, this work highlights the potential of domain adaptation in the field of molecular property predictions to refine existing pretrained models on a new compound data distribution.

Deciphering the Lexicon of Protein Targets: A Review on Multifaceted Drug Discovery in the Era of Artificial Intelligence.

Understanding protein sequence and structure is essential for understanding protein-protein interactions (PPIs), which are essential for many biological processes and diseases. Targeting protein binding hot spots, which regulate signaling and growth, with rational drug design is promising. Rational drug design uses structural data and computational tools to study protein binding sites and protein interfaces to design inhibitors that can change these interactions, thereby potentially leading to therapeutic approaches. Artificial intelligence (AI), such as machine learning (ML) and deep learning (DL), has advanced drug discovery and design by providing computational resources and methods. Quantum chemistry is essential for drug reactivity, toxicology, drug screening, and quantitative structure-activity relationship (QSAR) properties. This review discusses the methodologies and challenges of identifying and characterizing hot spots and binding sites. It also explores the strategies and applications of artificial-intelligence-based rational drug design technologies that target proteins and protein-protein interaction (PPI) binding hot spots. It provides valuable insights for drug design with therapeutic implications. We have also demonstrated the pathological conditions of heat shock protein 27 (HSP27) and matrix metallopoproteinases (MMP2 and MMP9) and designed inhibitors of these proteins using the drug discovery paradigm in a case study on the discovery of drug molecules for cancer treatment. Additionally, the implications of benzothiazole derivatives for anticancer drug design and discovery are deliberated.

Comparative Analysis of Chemical Descriptors by Machine Learning Reveals Atomistic Insights into Solute-Lipid Interactions.

This study explores the research area of drug solubility in lipid excipients, an area persistently complex despite recent advancements in understanding and predicting solubility based on molecular structure. To this end, this research investigated novel descriptor sets, employing machine learning techniques to understand the determinants governing interactions between solutes and medium-chain triglycerides (MCTs). Quantitative structure-property relationships (QSPR) were constructed on an extended solubility data set comprising 182 experimental values of structurally diverse drug molecules, including both development and marketed drugs to extract meaningful property relationships. Four classes of molecular descriptors, ranging from traditional representations to complex geometrical descriptions, were assessed and compared in terms of their predictive accuracy and interpretability. These include two-dimensional (2D) and three-dimensional (3D) descriptors, Abraham solvation parameters, extended connectivity fingerprints (ECFPs), and the smooth overlap of atomic position (SOAP) descriptor. Through testing three distinct regularized regression algorithms alongside various preprocessing schemes, the SOAP descriptor enabled the construction of a superior performing model in terms of interpretability and accuracy. Its atom-centered characteristics allowed contributions to be estimated at the atomic level, thereby enabling the ranking of prevalent molecular motifs and their influence on drug solubility in MCTs. The performance on a separate test set demonstrated high predictive accuracy (RMSE = 0.50) for 2D and 3D, SOAP, and Abraham Solvation descriptors. The model trained on ECFP4 descriptors resulted in inferior predictive accuracy. Lastly, uncertainty estimations for each model were introduced to assess their applicability domains and provide information on where the models may extrapolate in chemical space and, thus, where more data may be necessary to refine a data-driven approach to predict solubility in MCTs. Overall, the presented approaches further enable computationally informed formulation development by introducing a novel in silico approach for rational drug development and prediction of dose loading in lipids.

Design of potential anti-cancer agents as COX-2 inhibitors, using 3D-QSAR modeling, molecular docking, oral bioavailability proprieties, and molecular dynamics simulation.

Modeling the structural properties of novel morpholine-bearing 1, 5-diaryl-diazole derivatives as potent COX-2 inhibitor, two proposed models based on CoMFA and CoMSIA were evaluated by external and internal validation methods. Partial least squares analysis produced statistically significant models with Q 2 values of 0.668 and 0.652 for CoMFA and CoMSIA, respectively, and also a significant non-validated correlation coefficient R² with values of 0.882 and 0.878 for CoMFA and CoMSIA, respectively. Both models met the requirements of Golbraikh and Tropsha, which means that both models are consistent with all validation techniques. Analysis of the CoMFA and CoMSIA contribution maps and molecular docking revealed that the R1 substituent has a very significant effect on their biological activity. The most active molecules were evaluated for their thermodynamic stability by performing MD simulations for 100 ns; it was revealed that the designed macromolecular ligand complex with 3LN1 protein exhibits a high degree of structural and conformational stability. Based on these results, we predicted newly designed compounds, which have acceptable oral bioavailability properties and would have high synthetic accessibility.

Monte Carlo Optimization Method Based QSAR Modeling of Placental Barrier Permeability.

PURPOSE: In order to ensure that drug administration is safe during pregnancy, it is crucial to have the possibility to predict the placental permeability of drugs in humans. The experimental method which is most widely used for the said purpose is in vitro human placental perfusion, though the approach is highly expensive and time consuming. Quantitative structure-activity relationship (QSAR) modeling represents a powerful tool for the assessment of the drug placental transfer, and can be successfully employed to be an alternative in in vitro experiments. METHODS: The conformation-independent QSAR models covered in the present study were developed through the use of the SMILES notation descriptors and local molecular graph invariants. What is more, the Monte Carlo optimization method, was used in the test sets and the training sets as the model developer with three independent molecular splits. RESULTS: A range of different statistical parameters was used to validate the developed QSAR model, including the standard error of estimation, mean absolute error, root-mean-square error (RMSE), correlation coefficient, cross-validated correlation coefficient, Fisher ratio, MAE-based metrics and the correlation ideality index. Once the mentioned statistical methods were employed, an excellent predictive potential and robustness of the developed QSAR model was demonstrated. In addition, the molecular fragments, which are derived from the SMILES notation descriptors accounting for the decrease or increase in the investigated activity, were revealed. CONCLUSION: The presented QSAR modeling can be an invaluable tool for the high-throughput screening of the placental permeability of drugs.

Evaluation of Various Approaches to Estimate Transplacental Clearance of Vancomycin for Predicting Fetal Concentrations using a Maternal-Fetal Physiologically Based Pharmacokinetic Model.

BACKGROUND: Evaluating drug transplacental clearance is vital for forecasting fetal drug exposure. Ex vivo human placenta perfusion experiments are the most suitable approach for this assessment. Various in silico methods are also proposed. This study aims to compare these prediction methods for drug transplacental clearance, focusing on the large molecular weight drug vancomycin (1449.3 g/mol), using maternal-fetal physiologically based pharmacokinetic (m-f PBPK) modeling. METHODS: Ex vivo human placenta perfusion experiments, in silico approaches using intestinal permeability as a substitute (quantitative structure property relationship (QSPR) model and Caco-2 permeability in vitro-in vivo correlation model) and midazolam calibration model with Caco-2 scaling were assessed for determining the transplacental clearance (CLPD) of vancomycin. The m-f PBPK model was developed stepwise using Simcyp, incorporating the determined CLPD values as a crucial input parameter for transplacental kinetics. RESULTS: The developed PBPK model of vancomycin for non-pregnant adults demonstrated excellent predictive performance. By incorporating the CLPD parameterization derived from ex vivo human placenta perfusion experiments, the extrapolated m-f PBPK model consistently predicted maternal and fetal concentrations of vancomycin across diverse doses and distinct gestational ages. However, when the CLPD parameter was derived from alternative prediction methods, none of the extrapolated maternal-fetal PBPK models produced fetal predictions in line with the observed data. CONCLUSION: Our study showcased that combination of ex vivo human placenta perfusion experiments and m-f PBPK model has the capability to predict fetal exposure for the large molecular weight drug vancomycin, whereas other in silico approaches failed to achieve the same level of accuracy.

Machine learning-based QSAR and LB-PaCS-MD guided design of SARS-CoV-2 main protease inhibitors.

The global outbreak of the COVID-19 pandemic caused by the SARS-CoV-2 virus had led to profound respiratory health implications. This study focused on designing organoselenium-based inhibitors targeting the SARS-CoV-2 main protease (Mpro). The ligand-binding pathway sampling method based on parallel cascade selection molecular dynamics (LB-PaCS-MD) simulations was employed to elucidate plausible paths and conformations of ebselen, a synthetic organoselenium drug, within the Mpro catalytic site. Ebselen effectively engaged the active site, adopting proximity to H41 and interacting through the benzoisoselenazole ring in a π-π T-shaped arrangement, with an additional π-sulfur interaction with C145. In addition, the ligand-based drug design using the QSAR with GFA-MLR, RF, and ANN models were employed for biological activity prediction. The QSAR-ANN model showed robust statistical performance, with an r2training exceeding 0.98 and an RMSEtest of 0.21, indicating its suitability for predicting biological activities. Integration the ANN model with the LB-PaCS-MD insights enabled the rational design of novel compounds anchored in the ebselen core structure, identifying promising candidates with favorable predicted IC50 values. The designed compounds exhibited suitable drug-like characteristics and adopted an active conformation similar to ebselen, inhibiting Mpro function. These findings represent a synergistic approach merging ligand and structure-based drug design; with the potential to guide experimental synthesis and enzyme assay testing.

DiffDec: Structure-Aware Scaffold Decoration with an End-to-End Diffusion Model.

In molecular optimization, one popular way is R-group decoration on molecular scaffolds, and many efforts have been made to generate R-groups based on deep generative models. However, these methods mostly use information on known binding ligands, without fully utilizing target structure information. In this study, we proposed a new method, DiffDec, to involve 3D pocket constraints by a modified diffusion technique for optimizing molecules through molecular scaffold decoration. For end-to-end generation of R-groups with different sizes, we designed a novel fake atom mechanism. DiffDec was shown to be able to generate structure-aware R-groups with realistic geometric substructures by the analysis of bond angles and dihedral angles and simultaneously generate multiple R-groups for one scaffold on different growth anchors. The growth anchors could be provided by users or automatically determined by our model. DiffDec achieved R-group recovery rates of 69.67% and 45.34% in the single and multiple R-group decoration tasks, respectively, and these values were significantly higher than competing methods (37.33% and 26.85%). According to the molecular docking study, our decorated molecules obtained a better average binding affinity than baseline methods. The docking pose analysis revealed that DiffDec could decorate scaffolds with R-groups that exhibited improved binding affinities and more favorable interactions with the pocket. These results demonstrated the potential and applicability of DiffDec in real-world scaffold decoration for molecular optimization.

MELLODDY: Cross-pharma Federated Learning at Unprecedented Scale Unlocks Benefits in QSAR without Compromising Proprietary Information.

Federated multipartner machine learning has been touted as an appealing and efficient method to increase the effective training data volume and thereby the predictivity of models, particularly when the generation of training data is resource-intensive. In the landmark MELLODDY project, indeed, each of ten pharmaceutical companies realized aggregated improvements on its own classification or regression models through federated learning. To this end, they leveraged a novel implementation extending multitask learning across partners, on a platform audited for privacy and security. The experiments involved an unprecedented cross-pharma data set of 2.6+ billion confidential experimental activity data points, documenting 21+ million physical small molecules and 40+ thousand assays in on-target and secondary pharmacodynamics and pharmacokinetics. Appropriate complementary metrics were developed to evaluate the predictive performance in the federated setting. In addition to predictive performance increases in labeled space, the results point toward an extended applicability domain in federated learning. Increases in collective training data volume, including by means of auxiliary data resulting from single concentration high-throughput and imaging assays, continued to boost predictive performance, albeit with a saturating return. Markedly higher improvements were observed for the pharmacokinetics and safety panel assay-based task subsets.

Quantifying the Benefits of Imputation over QSAR Methods in Toxicology Data Modeling.

Imputation machine learning (ML) surpasses traditional approaches in modeling toxicity data. The method was tested on an open-source data set comprising approximately 2500 ingredients with limited in vitro and in vivo data obtained from the OECD QSAR Toolbox. By leveraging the relationships between different toxicological end points, imputation extracts more valuable information from each data point compared to well-established single end point methods, such as ML-based Quantitative Structure Activity Relationship (QSAR) approaches, providing a final improvement of up to around 0.2 in the coefficient of determination. A significant aspect of this methodology is its resilience to the inclusion of extraneous chemical or experimental data. While additional data typically introduces a considerable level of noise and can hinder performance of single end point QSAR modeling, imputation models remain unaffected. This implies a reduction in the need for laborious manual preprocessing tasks such as feature selection, thereby making data preparation for ML analysis more efficient. This successful test, conducted on open-source data, validates the efficacy of imputation approaches in toxicity data analysis. This work opens the way for applying similar methods to other types of sparse toxicological data matrices, and so we discuss the development of regulatory authority guidelines to accept imputation models, a key aspect for the wider adoption of these methods.