Dataset Information

The Chemical Effect Predictor (CEP) predictions for conazoles

ABSTRACT: The Chemical Effect Predictor (CEP) is a machine learning predictive tool based on data extracted from a Knowledge Graph (KG) designed for predicting hazard of drugs/chemicals in human health. The KG represents the complexity of human biology and how the action of the compound can affect human health through the intricate network of proteins and their associated biological processes to the disease phenotype. Based on this KG representing biological relationships, different variables describing the association between drug/chemical and diseases are extracted for training a machine learning predictive model based on a random forest algorithm. The machine learning model trained on external data reporting adverse effects caused by drugs is used for predicting hazards for query drugs/chemical of our interest. Each drug/chemical-disease association is assigned a prediction value ranging from 0 to 1, with 1 indicating the highest predicted probability that the query compound will cause the specific adverse effect. To support the analysis of CEP results, and identify robust signals for the prioritization of specific health effect concerns, the Disease Set Enrichment Analysis (DSEA) approach was designed. DSEA uses the classification of diseases in higher-order System Organ Classes (SOCs) and the ranked list of predicted diseases based on the prediction value and calculates an Normalized Enrichment score (NES) and adjusted p-value for each SOC. As a result, the hazard prediction for each compound is represented by 22 SOCs with its NES and adjusted p-value, which can be used to identify disease areas of concern for the compound.

ORGANISM(S): Human

SUBMITTER:

PROVIDER: S-RHER39 | biostudies-other |

REPOSITORIES: biostudies-other

ACCESS DATA

Similar Datasets

Project description:Predictions of interactions between target proteins and potential leads are of great benefit in the drug discovery process. We present a comprehensively applicable statistical prediction method for interactions between any proteins and chemical compounds, which requires only protein sequence data and chemical structure data and utilizes the statistical learning method of support vector machines. In order to realize reasonable comprehensive predictions which can involve many false positives, we propose two approaches for reduction of false positives: (i) efficient use of multiple statistical prediction models in the framework of two-layer SVM and (ii) reasonable design of the negative data to construct statistical prediction models. In two-layer SVM, outputs produced by the first-layer SVM models, which are constructed with different negative samples and reflect different aspects of classifications, are utilized as inputs to the second-layer SVM. In order to design negative data which produce fewer false positive predictions, we iteratively construct SVM models or classification boundaries from positive and tentative negative samples and select additional negative sample candidates according to pre-determined rules. Moreover, in order to fully utilize the advantages of statistical learning methods, we propose a strategy to effectively feedback experimental results to computational predictions with consideration of biological effects of interest. We show the usefulness of our approach in predicting potential ligands binding to human androgen receptors from more than 19 million chemical compounds and verifying these predictions by in vitro binding. Moreover, we utilize this experimental validation as feedback to enhance subsequent computational predictions, and experimentally validate these predictions again. This efficient procedure of the iteration of the in silico prediction and in vitro or in vivo experimental verifications with the sufficient feedback enabled us to identify novel ligand candidates which were distant from known ligands in the chemical space.

Project description:In the quest for advanced superconducting materials, the accurate prediction of critical temperatures (Tc) poses a formidable challenge, largely due to the complex interdependencies between superconducting properties and the chemical and structural characteristics of a given material. To address this challenges, we have developed a machine-learning framework that aims to elucidate these complicated and hitherto poorly understood structure-property and property-property relationships. This study introduces a novel machine-learning-based workflow, termed the Gradient Boosted Feature Selection (GBFS), which has been tailored to predict Tc for superconductors by employing a distributed gradient-boosting framework. This approach integrates exploratory data analyses, statistical evaluations, and multicollinearity reduction techniques to select highly relevant features from a high-dimensional feature space, derived solely from the chemical composition of materials. Our methodology was rigorously tested on a data set comprising approximately 16,400 chemical compounds with around 12,000 unique chemical compositions. The GBFS workflow enabled the development of a classification model that distinguishes compositions likely to exhibit Tc values greater than 10 K. This model achieved a weighted average F1-score of 0.912, an AUC-ROC of 0.986, and an average precision score of 0.919. Additionally, the GBFS workflow underpinned a regression model that predicted Tc values with an R2 of 0.945, an MAE of 3.54 K, and an RMSE of 6.57 K on a test set obtained via random splitting. Further exploration was conducted through out-of-sample Tc predictions, particularly those exceeding the liquid nitrogen temperature, and out-of-distribution predictions for (Ca1-xLax)FeAs2 based on varying lanthanum content. The outcome of our study underscores the significance of systematic feature analysis and selection in enhancing predictive model performance, offering various advantages over models that rely primarily on algorithmic complexity. This research not only advances the field of superconductivity but also sets a precedent for the application of machine learning in materials science.

Project description:Under the ExpoCast program, United States Environmental Protection Agency (EPA) researchers have developed a high-throughput (HT) framework for estimating aggregate exposures to chemicals from multiple pathways to support rapid prioritization of chemicals. Here, we present methods to estimate HT exposures to chemicals migrating into food from food contact substances (FCS). These methods consisted of combining an empirical model of chemical migration with estimates of daily population food intakes derived from food diaries from the National Health and Nutrition Examination Survey (NHANES). A linear regression model for migration at equilibrium was developed by fitting available migration measurements as a function of temperature, food type (i.e., fatty, aqueous, acidic, alcoholic), initial chemical concentration in the FCS (C0) and chemical properties. The most predictive variables in the resulting model were C0, molecular weight, log Kow, and food type (R2=0.71, p<0.0001). Migration-based concentrations for 1009 chemicals identified via publicly-available data sources as being present in polymer FCSs were predicted for 12 food groups (combinations of 3 storage temperatures and food type). The model was parameterized with screening-level estimates of C0 based on the functional role of chemicals in FCS. By combining these concentrations with daily intakes for food groups derived from NHANES, population ingestion exposures of chemical in mg/kg-bodyweight/day (mg/kg-BW/day) were estimated. Calibrated aggregate exposures were estimated for 1931 chemicals by fitting HT FCS and consumer product exposures to exposures inferred from NHANES biomonitoring (R2=0.61, p<0.001); both FCS and consumer product pathway exposures were significantly predictive of inferred exposures. Including the FCS pathway significantly impacted the ratio of predicted exposures to those estimated to produce steady-state blood concentrations equal to in-vitro bioactive concentrations. While these HT methods have large uncertainties (and thus may not be appropriate for assessments of single chemicals), they can provide critical refinement to aggregate exposure predictions used in risk-based chemical priority-setting.

Dataset Information

The Chemical Effect Predictor (CEP) predictions for conazoles

Similar Datasets

OmicsDI is part of the ELIXIR infrastructure

Tweets