Identification of biomarkers that distinguish chemical contaminants based on gene expression profiles

Systematic approaches to machine learning models for predicting pesticide toxicity

Pesticides play an important role in modern agriculture by protecting crops from pests and diseases. However, the negative consequences of pesticides, such as environmental contamination and adverse effects on human and ecological health, underscore the importance of accurate toxicity predictions. To address this issue, artificial intelligence models have emerged as valuable methods for predicting the toxicity of organic compounds. In this review article, we explore the application of machine learning (ML) for pesticide toxicity prediction. This review provides a detailed summary of recent developments, prediction models, and datasets used for pesticide toxicity prediction. In this analysis, we compared the results of several algorithms that predict the harmfulness of various classes of pesticides. Furthermore, this review article identified emerging trends and areas for future direction, showcasing the transformative potential of machine learning in promoting safer pesticide usage and sustainable agriculture.

Transcriptome analysis of a springtail, Folsomia candida, reveals energy constraint and oxidative stress during petroleum hydrocarbon exposure

Petroleum hydrocarbon (PHC) contamination in soil is ubiquitous and poses harmful consequences to many organisms. The toxicity of PHC-impacted soil is difficult to predict due to variations in mixture composition and the impacts of natural weathering processes. Hence, high-throughput methods to assess PHC-impacted soils is required to expedite land management decisions. Next-generation sequencing is a robust tool that allows researchers to investigate the effects of contaminants on the transcriptome of organisms and identify molecular biomarkers. In this study, the effects of PHCs on conventional endpoints (i.e., survival and reproduction) and gene expression rates of a model springtail species, Folsomia candida were investigated. Age-synchronized F. candida were exposed to ecologically-relevant concentrations of soils spiked with fresh crude oil to calculate the reproductive EC25 and EC50 values using conventional toxicity testing. Soils spiked to these concentrations were then used to evaluate effects on the F. candida transcriptome over a 7-day exposure period. RNA-seq analysis found 98 and 132 differentially expressed genes when compared to the control for the EC25 and EC50 treatment groups, respectively. The majority of up-regulated genes were related to xenobiotic biotransformation reactions and oxidative stress response, while down-regulated genes coded for carbohydrate and peptide metabolic processes. Promotion of the pentose phosphate pathway was also found. Results suggest that the decreased reproduction rates of F. candida exposed to PHCs is due to energy constraints caused by inhibition of carbohydrate metabolic processes and allocation of remaining energy to detoxify xenobiotics. These findings provide insights into the molecular effects in F. candida following exposure to crude oil for seven days and highlight their potential to be used as a high-throughput screening test for PHC-contaminated sites. Adverse molecular effects can be measured as early as 24 h following exposure, whereas conventional toxicity tests may require a minimum of four weeks.

Multimetric feature selection for analyzing multicategory outcomes of colorectal cancer: random forest and multinomial logistic regression models

Colorectal cancer (CRC) is one of the most common cancers worldwide, and a leading cause of cancer deaths. Better classifying multicategory outcomes of CRC with clinical and omic data may help adjust treatment regimens based on individual's risk. Here, we selected the features that were useful for classifying four-category survival outcome of CRC using the clinical and transcriptomic data, or clinical, transcriptomic, microsatellite instability and selected oncogenic-driver data (all data) of TCGA. We also optimized multimetric feature selection to develop the best multinomial logistic regression (MLR) and random forest (RF) models that had the highest accuracy, precision, recall and F1 score, respectively. We identified 2073 differentially expressed genes of the TCGA RNASeq dataset. MLR overall outperformed RF in the multimetric feature selection. In both RF and MLR models, precision, recall and F1 score increased as the feature number increased and peaked at the feature number of 600-1000, while the models' accuracy remained stable. The best model was the MLR one with 825 features based on sum of squared coefficients using all data, and attained the best accuracy of 0.855, F1 of 0.738 and precision of 0.832, which were higher than those using clinical and transcriptomic data. The top-ranked features in the MLR model of the best performance using clinical and transcriptomic data were different from those using all data. However, pathologic staging, HBS1L, TSPYL4, and TP53TG3B were the overlapping top-20 ranked features in the best models using clinical and transcriptomic, or all data. Thus, we developed a multimetric feature-selection based MLR model that outperformed RF models in classifying four-category outcome of CRC patients. Interestingly, adding microsatellite instability and oncogenic-driver data to clinical and transcriptomic data improved models' performances. Precision and recall of tuned algorithms may change significantly as the feature number changes, but accuracy appears not sensitive to these changes.

Machine learning-based biomarkers identification from toxicogenomics - Bridging to regulatory relevant phenotypic endpoints

One of the major challenges in realization and implementations of the Tox21 vision is the urgent need to establish quantitative link between in-vitro assay molecular endpoint and in-vivo regulatory-relevant phenotypic toxicity endpoint. Current toxicomics approach still mostly rely on large number of redundant markers without pre-selection or ranking, therefore, selection of relevant biomarkers with minimal redundancy would reduce the number of markers to be monitored and reduce the cost, time, and complexity of the toxicity screening and risk monitoring. Here, we demonstrated that, using time series toxicomics in-vitro assay along with machine learning-based feature selection (maximum relevance and minimum redundancy (MRMR)) and classification method (support vector machine (SVM)), an "optimal" number of biomarkers with minimum redundancy can be identified for prediction of phenotypic toxicity endpoints with good accuracy. We included two case studies for in-vivo carcinogenicity and Ames genotoxicity prediction, using 20 selected chemicals including model genotoxic chemicals and negative controls, respectively. The results suggested that, employing the adverse outcome pathway (AOP) concept, molecular endpoints based on a relatively small number of properly selected biomarker-ensemble involved in the conserved DNA-damage and repair pathways among eukaryotes, were able to predict both Ames genotoxicity endpoints and in-vivo carcinogenicity in rats. A prediction accuracy of 76% with AUC = 0.81 was achieved while predicting in-vivo carcinogenicity with the top-ranked five biomarkers. For Ames genotoxicity prediction, the top-ranked five biomarkers were able to achieve prediction accuracy of 70% with AUC = 0.75. However, the specific biomarkers identified as the top-ranked five biomarkers are different for the two different phenotypic genotoxicity assays. The top-ranked biomarkers for the in-vivo carcinogenicity prediction mainly focused on double strand break repair and DNA recombination, whereas the selected top-ranked biomarkers for Ames genotoxicity prediction are associated with base- and nucleotide-excision repair The method developed in this study will help to fill in the knowledge gap in phenotypic anchoring and predictive toxicology, and contribute to the progress in the implementation of tox 21 vision for environmental and health applications.

Assessing PD-L1 Expression Level via Preoperative MRI in HCC Based on Integrating Deep Learning and Radiomics Features

To assess if quantitative integrated deep learning and radiomics features can predict the PD-L1 expression level in preoperative MRI of hepatocellular carcinoma (HCC) patients. The data in this study consist of 103 hepatocellular carcinoma patients who received immunotherapy in a single center. These patients were divided into a high PD-L1 expression group (30 patients) and a low PD-L1 expression group (73 patients). Both radiomics and deep learning features were extracted from their MRI sequence of T2-WI, which were merged into an integrative feature space for machine learning for the prediction of PD-L1 expression. The five-fold cross-validation was adopted to validate the performance of the model, while the AUC was used to assess the predictive ability of the model. Based on the five-fold cross-validation, the integrated model achieved the best prediction performance, with an AUC score of 0.897 ± 0.084, followed by the deep learning-based model with an AUC of 0.852 ± 0.043 then the radiomics-based model with AUC of 0.794 ± 0.035. The feature set integrating radiomics and deep learning features is more effective in predicting PD-L1 expression level than only one feature type. The integrated model can achieve fast and accurate prediction of PD-L1 expression status in preoperative MRI of HCC patients.

Development of predicitve models to distinguish metals from non-metal toxicants, and individual metal from one another

BACKGROUND

Evaluating the toxicity of chemical mixture and their possible mechanism of action is still a challenge for humans and other organisms. Microarray classifier analysis has shown promise in the toxicogenomic area by identifying biomarkers to predict unknown samples. Our study focuses on identifying gene markers with better sensitivity and specificity, building predictive models to distinguish metals from non-metal toxicants, and individual metal from one another, and furthermore helping understand underlying toxic mechanisms.

RESULTS

Based on an independent dataset test, using only 15 gene markers, we were able to distinguish metals from non-metal toxicants with 100% accuracy. Of these, 6 and 9 genes were commonly down- and up-regulated respectively by most of the metals. 8 out of 15 genes belong to membrane protein coding genes. Function well annotated genes in the list include ADORA2B, ARNT, S100G, and DIO3. Also, a 10-gene marker list was identified that can discriminate an individual metal from one another with 100% accuracy. We could find a specific gene marker for each metal in the 10-gene marker list. Function well annotated genes in this list include GSTM2, HSD11B, AREG, and C8B.

CONCLUSIONS

Our findings suggest that using a microarray classifier analysis, not only can we create diagnostic classifiers for predicting an exact metal contaminant from a large scale of contaminant pool with high prediction accuracy, but we can also identify valuable biomarkers to help understand the common and underlying toxic mechanisms induced by metals.

Considerations for Strategic Use of High-Throughput Transcriptomics Chemical Screening Data in Regulatory Decisions

Recently, numerous organizations, including governmental regulatory agencies in the U.S. and abroad, have proposed using data from New Approach Methodologies (NAMs) for augmenting and increasing the pace of chemical assessments. NAMs are broadly defined as any technology, methodology, approach or combination thereof that can be used to provide information on chemical hazard and risk assessment that avoids the use of intact animals. High-throughput transcriptomics (HTTr) is a type of NAM that uses gene expression profiling as an endpoint for rapidly evaluating the effects of large numbers of chemicals on in vitro cell culture systems. As compared to targeted high-throughput screening (HTS) approaches that measure the effect of chemical X on target Y, HTTr is a non-targeted approach that allows researchers to more broadly characterize the integrated response of an intact biological system to chemicals that may affect a specific biological target or many biological targets under a defined set of treatment conditions (time, concentration, etc.). HTTr screening performed in concentration-response mode can provide potency estimates for the concentrations of chemicals that produce perturbations in cellular response pathways. Here, we discuss study design considerations for HTTr concentration-response screening and present a framework for the use of HTTr-based biological pathway-altering concentrations (BPACs) in a screening-level, risk-based chemical prioritization approach. The framework involves concentration-response modeling of HTTr data, mapping gene level responses to biological pathways, determination of BPACs, in vitro-to-in vivo extrapolation (IVIVE) and comparison to human exposure predictions.

Crosstalk between Receptor and Non-receptor Mediated Chemical Modes of Action in Rat Livers Converges through a Dysregulated Gene Expression Network at Tumor Suppressor Tp53

Chemicals, toxicants, and environmental stressors mediate their biologic effect through specific modes of action (MOAs). These encompass key molecular events that lead to changes in the expression of genes within regulatory pathways. Elucidating shared biologic processes and overlapping gene networks will help to better understand the toxicologic effects on biological systems. In this study we used a weighted network analysis of gene expression data from the livers of male Sprague-Dawley rats exposed to chemicals that elicit their effects through receptor-mediated MOAs (aryl hydrocarbon receptor, orphan nuclear hormone receptor, or peroxisome proliferator-activated receptor-α) or non-receptor-mediated MOAs (cytotoxicity or DNA damage). Four gene networks were highly preserved and statistically significant in each of the two MOA classes. Three of the four networks contain genes that enrich for immunity and defense. However, many canonical pathways related to an immune response were activated from exposure to the non-receptor-mediated MOA chemicals and deactivated from exposure to the receptor-mediated MOA chemicals. The top gene network contains a module with 33 genes including tumor suppressor TP53 as the central hub which was slightly up-regulated in gene expression compared to control. Although, there is crosstalk between the two MOA classes of chemicals at the TP53 gene network, more than half of the genes are dysregulated in opposite directions. For example, Thromboxane A Synthase 1 (Tbxas1), a cytochrome P450 protein coding gene regulated by Tp53, is down-regulated by exposure to the receptor-mediated chemicals but up-regulated by the non-receptor-mediated chemicals. The regulation of gene expression by the chemicals within MOA classes was consistent despite varying alanine transaminase and aspartate aminotransferase liver enzyme measurements. These results suggest that overlap in toxicologic pathways by chemicals with different MOAs provides common mechanisms for discordant regulation of gene expression within molecular networks.

Prioritization of Contaminants of Emerging Concern in Wastewater Treatment Plant Discharges Using Chemical:Gene Interactions in Caged Fish

We examined whether contaminants present in surface waters could be prioritized for further assessment by linking the presence of specific chemicals to gene expression changes in exposed fish. Fathead minnows were deployed in cages for 2, 4, or 8 days at three locations near two different wastewater treatment plant discharge sites in the Saint Louis Bay, Duluth, MN and one upstream reference site. The biological impact of 51 chemicals detected in the surface water of 133 targeted chemicals was determined using biochemical endpoints, exposure activity ratios for biological and estrogenic responses, known chemical:gene interactions from biological pathways and knowledge bases, and analysis of the covariance of ovary gene expression with surface water chemistry. Thirty-two chemicals were significantly linked by covariance with expressed genes. No estrogenic impact on biochemical endpoints was observed in male or female minnows. However, bisphenol A (BPA) was identified by chemical:gene covariation as the most impactful estrogenic chemical across all exposure sites. This was consistent with identification of estrogenic effects on gene expression, high BPA exposure activity ratios across all test sites, and historical analysis of the study area. Gene expression analysis also indicated the presence of nontargeted chemicals including chemotherapeutics consistent with a local hospital waste stream. Overall impacts on gene expression appeared to be related to changes in treatment plant function during rain events. This approach appears useful in examining the impacts of complex mixtures on fish and offers a potential route in linking chemical exposure to adverse outcomes that may reduce population sustainability.

A machine-learning approach for predicting palmitoylation sites from integrated sequence-based features

Palmitoylation is the covalent attachment of lipids to amino acid residues in proteins. As an important form of protein posttranslational modification, it increases the hydrophobicity of proteins, which contributes to the protein transportation, organelle localization, and functions, therefore plays an important role in a variety of cell biological processes. Identification of palmitoylation sites is necessary for understanding protein-protein interaction, protein stability, and activity. Since conventional experimental techniques to determine palmitoylation sites in proteins are both labor intensive and costly, a fast and accurate computational approach to predict palmitoylation sites from protein sequences is in urgent need. In this study, a support vector machine (SVM)-based method was proposed through integrating PSI-BLAST profile, physicochemical properties, [Formula: see text]-mer amino acid compositions (AACs), and [Formula: see text]-mer pseudo AACs into the principal feature vector. A recursive feature selection scheme was subsequently implemented to single out the most discriminative features. Finally, an SVM method was implemented to predict palmitoylation sites in proteins based on the optimal features. The proposed method achieved an accuracy of 99.41% and Matthews Correlation Coefficient of 0.9773 for a benchmark dataset. The result indicates the efficiency and accuracy of our method in prediction of palmitoylation sites based on protein sequences.

Application of the TGx-28.65 transcriptomic biomarker to classify genotoxic and non-genotoxic chemicals in human TK6 cells in the presence of rat liver S9

In vitro transcriptional signatures that predict toxicities can facilitate chemical screening. We previously developed a transcriptomic biomarker (known as TGx-28.65) for classifying agents as genotoxic (DNA damaging) and non-genotoxic in human lymphoblastoid TK6 cells. Because TK6 cells do not express cytochrome P450s, we confirmed accurate classification by the biomarker in cells co-exposed to 1% 5,6 benzoflavone/phenobarbital-induced rat liver S9 for metabolic activation. However, chemicals may require different types of S9 for activation. Here we investigated the response of TK6 cells to higher percentages of Aroclor-, benzoflavone/phenobarbital-, or ethanol-induced rat liver S9 to expand TGx-28.65 biomarker applicability. Transcriptional profiles were derived 3 to 4 hr following a 4 hr co-exposure of TK6 cells to test chemicals and S9. Preliminary studies established that 10% Aroclor- and 5% ethanol-induced S9 alone did not induce the TGx-28.65 biomarker genes. Seven genotoxic and two non-genotoxic chemicals (and concurrent solvent and positive controls) were then tested with one of the S9s (selected based on cell survival and micronucleus induction). Relative survival and micronucleus frequency was assessed by flow cytometry in cells 20 hr post-exposure. Genotoxic/non-genotoxic chemicals were accurately classified using the different S9s. One technical replicate of cells co-treated with dexamethasone and 10% Aroclor-induced S9 was falsely classified as genotoxic, suggesting caution in using high S9 concentrations. Even low concentrations of genotoxic chemicals (those not causing cytotoxicity) were correctly classified, demonstrating that TGx-28.65 is a sensitive biomarker of genotoxicity. A meta-analysis of datasets from 13 chemicals supports that different S9s can be used in TK6 cells, without impairing classification using the TGx-28.65 biomarker.

Combined Transcriptomics Analysis for Classification of Adverse Effects As a Potential End Point in Effect Based Screening

Environmental risk assessment relies on the use of bioassays to assess the environmental impact of chemicals. Gene expression is gaining acceptance as a valuable mechanistic end point in bioassays and effect-based screening. Data analysis and its results, however, are complex and often not directly applicable in risk assessment. Classifier analysis is a promising method to turn complex gene expression analysis results into answers suitable for risk assessment. We have assembled a large gene expression data set assembled from multiple studies and experiments in the springtail Folsomia candida, with the aim of selecting a set of genes that can be trained to classify general toxic stress. By performing differential expression analysis prior to classifier training, we were able to select a set of 135 genes which was enriched in stress related processes. Classifier models from this set were used to classify two test sets comprised of chemical spiked, polluted, and clean soils and compared to another, more traditional classifier feature selection. The gene set presented here outperformed the more traditionally selected gene set. This gene set has the potential to be used as a biomarker to test for adverse effects caused by chemicals in springtails to provide end points in environmental risk assessment.

Protein submitochondrial localization from integrated sequence representation and SVM-based backward feature extraction

Mitochondrion, a tiny energy factory, plays an important role in various biological processes of most eukaryotic cells.

Cross-species toxicogenomic analyses and phenotypic anchoring in response to groundwater low-level pollution

Background

Comparison of toxicogenomic data facilitates the identification of deregulated gene patterns and maximizes health risk prediction in human.

Results

Here, we performed phenotypic anchoring on the effects of acute exposure to low-grade polluted groundwater using mouse and zebrafish. Also, we evaluated two windows of chronic exposure in mouse, starting in utero and at the end of lactation. Bioinformatic analysis of livers microarray data showed that the number of deregulated biofunctions and pathways is higher after acute exposure, compared to the chronic one. It also revealed specific profiles of altered gene expression in all treatments, pointing to stress response/mitochondrial pathways as major players of environmental toxicity. Of note, dysfunction of steroid hormones was also predicted by bioinformatic analysis and verified in both models by traditional approaches, serum estrogens measurement and vitellogenin mRNA determination in mice and zebrafish, respectively.

Conclusions

In our report, phenotypic anchoring in two vertebrate model organisms highlights the toxicity of low-grade pollution, with varying susceptibility based on exposure window. The overlay of zebrafish and mice deregulated pathways, more than single genes, is useful in risk identification from chemicals implicated in the observed effects.

Electronic supplementary material

The online version of this article (doi:10.1186/1471-2164-15-1067) contains supplementary material, which is available to authorized users.

Sequence-based identification of recombination spots using pseudo nucleic acid representation and recursive feature extraction by linear kernel SVM

Background

Identification of the recombination hot/cold spots is critical for understanding the mechanism of recombination as well as the genome evolution process. However, experimental identification of recombination spots is both time-consuming and costly. Developing an accurate and automated method for reliably and quickly identifying recombination spots is thus urgently needed.

Results

Here we proposed a novel approach by fusing features from pseudo nucleic acid composition (PseNAC), including NAC, n-tier NAC and pseudo dinucleotide composition (PseDNC). A recursive feature extraction by linear kernel support vector machine (SVM) was then used to rank the integrated feature vectors and extract optimal features. SVM was adopted for identifying recombination spots based on these optimal features. To evaluate the performance of the proposed method, jackknife cross-validation test was employed on a benchmark dataset. The overall accuracy of this approach was 84.09%, which was higher (from 0.37% to 3.79%) than those of state-of-the-art tools.

Conclusions

Comparison results suggested that linear kernel SVM is a useful vehicle for identifying recombination hot/cold spots.