Abstract:Increasing the predictive potential of early diagnostics and correction of negative outcomes is becoming especially relevant for preventing and reducing personalized and population health risks, including those caused by environmental exposures. The purpose of the study was to develop scientific and methodological grounds for iterative numerical prediction of risk and harm to human health under chemical environmental exposures. The study design was based on the consistent implementation of an algorithm for system analysis of the development of negative effects under a chemical environmental exposure, from protein targets to systemic metabolic disorders. The in-depth examination covered more than 1 million people living under real long-term combined inhalation exposure at doses up to 5–10 RfC. About 350 digital multifactor models, including about 5.5 thousand parameters, were evaluated. Structural bioinformation matrices were constructed to identify the sequence of response events at the molecular-cellular level. These events are initiated by the transformation of the protein-peptide profile of human blood plasma, which determine the metabolome. The study clarifies the elements of involvement of 20 target proteins in the pathogenesis of metabolic disorders associated with hypertension, dyslipidemia, obesity, hepatosis, and cognitive dysfunction associated with chemical combined exposure. The criteria for the safe content of 10 contaminants were substantiated, taking into account their combinations in human biological media. Predictive assessments of pathogenetic pathways were confirmed by the facts of their implementation at the cellular-tissue, organ and body level as metabolic disorders and existing diseases of the cardiovascular, nervous systems, lipoprotein metabolism, etc., proven to be associated with effects produced by airborne pollutants, including combined ones. The study expands the existing methodological approaches to assessing combined effects of chemicals taking into account parameterized cause-effect relations of biomarkers of exposure and effects and quantitative assessment of additional cases of risk occurrence. Assessment of the developed digital models revealed that combined chemical exposures were predominantly synergic and emergent (up to 70 % cases). We developed conceptual grounds and architecture of iterative risk prediction and the development of risk-associated diseases, including real harm to health upon elevated expression of protein targets. Thus, the digitized version of the forecast (digital platform), as a multi-level cascade model, is a tool for scientific analysis of a hygienic situation with the parameterization of expected negative outcomes. It determines methods for their correction and prevention, which increases the reliability of hygienic assessments and the validity of management decisions.

Molecular Fingerprints Optimization for Enhanced Predictive Modeling

Identification of Optimal Machine Learning Algorithms and Molecular Fingerprints for Explainable Toxicity Prediction Models Using ToxCast/Tox21 Bioassay Data

Predicting the Reproductive Toxicity of Chemicals Using Ensemble Learning Methods and Molecular Fingerprints.

Predicting the Activity of Unidentified Chemicals in Complementary Bioassays from the HRMS Data to Pinpoint Potential Endocrine Disruptors

Improved Prediction Of Drug-Target Interactions Based On Ensemble Learning With Fuzzy Local Ternary Pattern

ADMET property prediction through combinations of molecular fingerprints

FP-ADMET: a compendium of fingerprint-based ADMET prediction models

Count-Based Morgan Fingerprint: A More Efficient and Interpretable Molecular Representation in Developing Machine Learning-Based Predictive Regression Models for Water Contaminants' Activities and Properties.

An Explainable Supervised Machine Learning Model for Predicting Respiratory Toxicity of Chemicals Using Optimal Molecular Descriptors

Toxicity prediction using target, interactome, and pathway profiles as descriptors

Predicting the Cytotoxicity of Chemicals Using Ensemble Learning Methods and Molecular Fingerprints

Comparison of Descriptor- and Fingerprint Sets in Machine Learning Models for ADME-Tox Targets

Discriminant Models on Mitochondrial Toxicity Improved by Consensus Modeling and Resolving Imbalance in Training.

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

Hyper-parameter optimization of multiple machine learning algorithms for molecular property prediction using hyperopt library

MolFPG: Multi-level Fingerprint-Based Graph Transformer for Accurate and Robust Drug Toxicity Prediction.

Computational Models to Predict Endocrine-Disrupting Chemical Binding with Androgen or Oestrogen Receptors.

Accurate Clinical Toxicity Prediction using Multi-task Deep Neural Nets and Contrastive Molecular Explanations

Permutationally Invariant Deep Learning Approach to Molecular Fingerprinting with Application to Compound Mixtures

Scientific and methodological grounds for iterative prediction of risk and harm to human health under chemical environmental exposures: From protein targets to systemic metabolic disorders