On Some Novel Similarity-Based Functions Used in the ML-Based q-RASAR Approach for Efficient Quantitative Predictions of Selected Toxicity End Points

Abstract

The novel quantitative read-across structure−activity relationship (q-RASAR) approach uses read-across-derived similarity functions in the quantitative structure−activity relationship (QSAR) modeling framework in a unique way for supervised model generation. The aim of this study is to explore how this workflow enhances the external (test set) prediction quality of conventional QSAR models by the incorporation of some novel similaritybased functions as additional descriptors using the same level of chemical information. To establish this, five different toxicity data sets, for which QSAR models were reported previously, have been considered in the q-RASAR modeling exercise, which uses chemical similarity-derived measures. The identical sets of chemical features along with the same compositions of training and test sets as reported previously were used in the present analysis for ease of comparison. The RASAR descriptors were calculated based on a chosen similarity measure with the default setting of relevant hyperparameter(s) and were then clubbed with the original structural and physicochemical descriptors, and the number of selected features was further optimized by employing a grid search technique applied on the respective training sets. These features were then used to develop multiple linear regression (MLR) q-RASAR models that show enhanced predictivity as compared to the QSAR models developed previously. Moreover, various other ML algorithms like support vector machine (SVM), linear SVM, random forest, partial least squares, and ridge regression were also employed using the same feature combinations as used in the MLR models to compare the prediction qualities. The q-RASAR models for five different data sets possess at least one of the RASAR descriptors, RA f unction, g m , and average similarity, suggesting that these are important determinants of similarities that contribute to the development of predictive q-RASAR models, as also evident from the SHAP analysis of the models.

Introduction to the Special Issue: AI Meets Toxicology

“…The methods described in this special issue cover a wide range of AI methods ranging from expert systems, ,, over similarity measures including read-across methods, − to classical machine learning such as random forests (RF), support vector machines (SVM), and artificial neural networks (ANN) − ,, to deep learning (DL) methods ,,− , including equivariant neural networks, deep generative models, and even large language models . In addition to models relying purely on the chemical structure, there is a notable trend of bringing in additional modalities to improve or inform predictive models. , In the following, we provide an overview of the AI approaches used in the publications contained in the SI.…”

“…17,18 Aside from the chemical structure, there is a growing tendency to incorporate biological characterizations and read-outs, for example, via cell morphology 20,21 or transcriptomics. 22 The utilization of diverse representations, ranging from molecular structures to biological features, enhances the predictive models showcased in this section and could improve the comprehensive understanding of toxicological properties.…”

“…The ensemble model, featuring five random forest classifiers guided by features learned by convolutional neural networks, achieved up to 90% accuracy, demonstrating potential as a quick and cost-effective screening tool for preclinical applications and toxicological studies. Lejal et al 22 utilized cell painting and transcriptomic data on compounds to predict their potential for DILI. They associated cell morphological signatures with DILI chemical data sets, analyzed the mechanisms of action, identified signaling pathways related to liver toxicity, and established a hypothetical link between morphological perturbations and gene deregulation.…”

“…The authors also presented an evaluation framework that could efficiently assess many substances, with the potential for further improvements using AI approaches. Banerjee and Roy have recently introduced the quantitative read-across structure–activity relationship (q-RASAR) approach, which incorporates novel similarity-based functions as additional descriptors, resulting in enhanced predictive models compared to conventional QSAR models across multiple toxicity data sets. Lester et al proposed an approach that incorporates biological and toxicological features quantitatively, using fingerprint keys, to compute a similarity score for read-across predictions in toxicological safety assessment, enhancing transparency and consistency among experts and increasing regulatory acceptance.…”

Predictive Quantitative Read-Across Structure–Property Relationship Modeling of the Retention Time (LogtR) of Pesticide Residues Present in Foods and Vegetables

“…The software tools used in this study are easy to use and fast, and the majority of the tools are freely available, which makes our workflow quite economical compared to experimentation. The q-RASPR is a more efficient technique as a result of its better external predictivity, interpretability, and transferability; 42 therefore, it has the potential to be used as a good alternative approach of retention time prediction and toxicity identification.…”