Artificial Intelligence and Machine Learning Methods to Evaluate Cardiotoxicity following the Adverse Outcome Pathway Frameworks

Cardiovascular disease is a leading global cause of mortality. The potential cardiotoxic effects of chemicals from different classes, such as environmental contaminants, pesticides, and drugs can significantly contribute to effects on health. The same chemical can induce cardiotoxicity in different ways, following various Adverse Outcome Pathways (AOPs). In addition, the potential synergistic effects between chemicals further complicate the issue. In silico methods have become essential for tackling the problem from different perspectives, reducing the need for traditional in vivo testing, and saving valuable resources in terms of time and money. Artificial intelligence (AI) and machine learning (ML) are among today's advanced approaches for evaluating chemical hazards. They can serve, for instance, as a first-tier component of Integrated Approaches to Testing and Assessment (IATA). This study employed ML and AI to assess interactions between chemicals and specific biological targets within the AOP networks for cardiotoxicity, starting with molecular initiating events (MIEs) and progressing through key events (KEs). We explored methods to encode chemical information in a suitable way for ML and AI. We started with commonly used approaches in Quantitative Structure-Activity Relationship (QSAR) methods, such as molecular descriptors and different types of fingerprint. We then increased the complexity of encoders, incorporating graph-based methods, auto-encoders, and character embeddings employed in neural language processing. We also developed a multimodal neural network architecture, capable of considering the complementary nature of different chemical representations simultaneously. The potential of this approach, compared to more conventional architectures designed to handle a single encoder, becomes apparent when the amount of data increases.

The VEGA Tool to Check the Applicability Domain Gives Greater Confidence in the Prediction of In Silico Models

A sound assessment of in silico models and their applicability domain can support the use of new approach methodologies (NAMs) in chemical risk assessment and requires increasing the users' confidence in this approach. Several approaches have been proposed to evaluate the applicability domain of such models, but their prediction power still needs a thorough assessment. In this context, the VEGA tool capable of assessing the applicability domain of in silico models is examined for a range of toxicological endpoints. The VEGA tool evaluates chemical structures and other features related to the predicted endpoints and is efficient in measuring applicability domain, enabling the user to identify less accurate predictions. This is demonstrated with many models addressing different endpoints, towards toxicity of relevance to human health, ecotoxicological endpoints, environmental fate, physicochemical and toxicokinetic properties, for both regression models and classifiers.

In silico Assessment of Pharmacological Profile of Low Molecular Weight Oligo-Hydroxyalkanoates

Polyhydroxyalkanoates (PHAs) are a large class of polyesters that are biosynthesized by microorganisms at large molecular weights (Mw > 80 kDa) and have a great potential for medical applications because of their recognized biocompatibility. Among PHAs, poly(3-hydroxybutyrate), poly(4-hydroxybutyrate), poly(3-hydroxyvalerate), poly(4-hydroxyvalerate), and their copolymers are proposed to be used in biomedicine, but only poly(4-hydroxybutyrate) has been certified for medical application. Along with the hydrolysis of these polymers, low molecular weight oligomers are released typically. In this study, we have used a computational approach to assess the absorption, distribution, metabolism, and excretion (ADME)-Tox profiles of low molecular weight oligomers (≤32 units) consisting of 3-hydroxybutyrate, 4-hydroxybutyrate, 3-hydroxyvalerate, 4-hydroxyvalerate, 3-hydroxybutyrate-co-3-hydroxyvalerate, and the hypothetical PHA consisting of 4-hydroxybutyrate-co-4-hydroxyvalerate. According to our simulations, these oligomers do not show cardiotoxicity, hepatotoxicity, carcinogenicity or mutagenicity, and are neither substrates nor inhibitors of the cytochromes involved in the xenobiotic's metabolism. They also do not affect the human organic cation transporter 2 (OCT2). However, they are considered to be inhibitors of the organic anion transporters OATP1B1, and OATP1B3. In addition, they may produce eye irritation, and corrosion, skin irritation and have a low antagonistic effect on the androgen receptor.

Solubility and ADMET profiles of short oligomers of lactic acid

Polylactic acid (PLA) is a polymer with an increased potential to be used in different medical applications, including tissue engineering and drug-carries. The use of PLA in medical applications implies the evaluation of the human organism's response to the polymer inserting and to its degradation products. Consequently, within this study, we have investigated the solubility and ADMET profiles of the short oligomers (having the molecular weight lower than 3000 Da) resulting in degradation products of PLA. There is a linear decrease of the molar solubility of investigated oligomers with molecular weight. The results that are obtained also reveal that short oligomers of PLA have promising pharmacological profiles and limited toxicological effects on humans. These oligomers are predicted as potential inhibitors of the organic anion transporting peptides OATP1B1 and OATP1B3, they present minor probability to affect the androgen and glucocorticoid receptors, have a weak potential of hepatotoxicity, and may produce eye injuries. These outcomes may be used to guide or to supplement in vitro and/or in vivo toxicity tests such as to enhance the biodegradation properties of the biopolymer.

Key characteristics of 86 agents known to cause cancer in humans

Since the inception of the International Agency for Research on Cancer (IARC) in the early 1970s, the IARC Monographs Programme has evaluated more than 1000 agents with respect to carcinogenic hazard; of these, up to and including Volume 119 of the IARC Monographs, 120 agents met the criteria for classification as carcinogenic to humans (Group 1). Volume 100 of the IARC Monographs provided a review and update of Group 1 carcinogens. These agents were divided into six broad categories: (I) pharmaceuticals; (II) biological agents; (III) arsenic, metals, fibers, and dusts; (IV) radiation; (V) personal habits and indoor combustions; and (VI) chemical agents and related occupations. Data on biological mechanisms of action (MOA) were extracted from the Monographs to assemble a database on the basis of ten key characteristics attributed to human carcinogens. After some grouping of similar agents, the characteristic profiles were examined for 86 Group 1 agents for which mechanistic information was available in the IARC Monographs up to and including Volume 106, based upon data derived from human in vivo, human in vitro, animal in vivo, and animal in vitro studies. The most prevalent key characteristic was "is genotoxic", followed by "alters cell proliferation, cell death, or nutrient supply" and "induces oxidative stress". Most agents exhibited several of the ten key characteristics, with an average of four characteristics per agent, a finding consistent with the notion that cancer development in humans involves multiple pathways. Information on the key characteristics was often available from multiple sources, with many agents demonstrating concordance between human and animal sources, particularly with respect to genotoxicity. Although a detailed comparison of the characteristics of different types of agents was not attempted here, the overall characteristic profiles for pharmaceutical agents and for chemical agents and related occupations appeared similar. Further in-depth analyses of this rich database of characteristics of human carcinogens are expected to provide additional insights into the MOA of human cancer development.