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

Two thyroperoxidase-inhibiting chemicals induce shared transcriptional changes in hippocampus of developing rats

Thyroid hormone (TH) system disrupting compounds can impair brain development by perturbing TH action during critical life stages. Human exposure to TH system disrupting chemicals is therefore of great concern. To better protect humans against such chemicals, sensitive test methods that can detect effects on the developing brain are critical. Worryingly, however, current test methods are not sensitive and specific towards TH-mediated effects. To address this shortcoming, we performed RNA-sequencing of rat brains developmentally exposed to two different thyroperoxidase (TPO) inhibiting compounds, the medical drug methimazole (MMI) or the pesticide amitrole. Pregnant and lactating rats were exposed to 8 and 16 mg/kg/day(d) MMI or 25 and 50 mg/kg/d amitrole from gestational day 7 until postnatal day 16. Bulk-RNA-seq was performed on hippocampus from the 16-day old male pups. MMI and amitrole caused pronounced changes to the transcriptomes; 816 genes were differentially expressed, and 425 gene transcripts were similarly affected by both chemicals. Functional terms indicate effects from key cellular functions to changes in cell development, migration and differentiation of several cell populations. Of the total number of DEGs, 106 appeared to form a consistent transcriptional fingerprint of developmental hypothyroidism as they were similarly and dose-dependently expressed across all treatment groups. Using a filtering system, we identified 20 genes that appeared to represent the most sensitive, robust and dose-dependent markers of altered TH-mediated brain development. These markers provide inputs to the adverse outcome pathway (AOP) framework where they, in the context of linking TPO inhibiting compounds to adverse cognitive function, can be used to assess altered gene expression in the hippocampus in rat toxicity studies.

Nonlethal Furfural Exposure Causes Genomic Alterations and Adaptability Evolution in Saccharomyces cerevisiae

Furfural is a major inhibitor found in lignocellulosic hydrolysate, a promising feedstock for the biofermentation industry. In this study, we aimed to investigate the potential impact of this furan-derived chemical on yeast genome integrity and phenotypic evolution by using genetic screening systems and high-throughput analyses. Our results showed that the rates of aneuploidy, chromosomal rearrangements (including large deletions and duplications), and loss of heterozygosity (LOH) increased by 50-fold, 23-fold, and 4-fold, respectively, when yeast cells were cultured in medium containing a nonlethal dose of furfural (0.6 g/L). We observed significantly different ratios of genetic events between untreated and furfural-exposed cells, indicating that furfural exposure induced a unique pattern of genomic instability. Furfural exposure also increased the proportion of CG-to-TA and CG-to-AT base substitutions among point mutations, which was correlated with DNA oxidative damage. Interestingly, although monosomy of chromosomes often results in the slower growth of yeast under spontaneous conditions, we found that monosomic chromosome IX contributed to the enhanced furfural tolerance. Additionally, terminal LOH events on the right arm of chromosome IV, which led to homozygosity of the SSD1 allele, were associated with furfural resistance. This study sheds light on the mechanisms underlying the influence of furfural on yeast genome integrity and adaptability evolution. IMPORTANCE Industrial microorganisms are often exposed to multiple environmental stressors and inhibitors during their application. This study demonstrates that nonlethal concentrations of furfural in the culture medium can significantly induce genome instability in the yeast Saccharomyces cerevisiae. Notably, furfural-exposed yeast cells displayed frequent chromosome aberrations, indicating the potent teratogenicity of this inhibitor. We identified specific genomic alterations, including monosomic chromosome IX and loss of heterozygosity of the right arm of chromosome IV, that confer furfural tolerance to a diploid S. cerevisiae strain. These findings enhance our understanding of how microorganisms evolve and adapt to stressful environments and offer insights for developing strategies to improve their performance in industrial applications.

Integrative toxicogenomics: Advancing precision medicine and toxicology through artificial intelligence and OMICs technology

More information about a person's genetic makeup, drug response, multi-omics response, and genomic response is now available leading to a gradual shift towards personalized treatment. Additionally, the promotion of non-animal testing has fueled the computational toxicogenomics as a pivotal part of the next-gen risk assessment paradigm. Artificial Intelligence (AI) has the potential to provid new ways analyzing the patient data and making predictions about treatment outcomes or toxicity. As personalized medicine and toxicogenomics involve huge data processing, AI can expedite this process by providing powerful data processing, analysis, and interpretation algorithms. AI can process and integrate a multitude of data including genome data, patient records, clinical data and identify patterns to derive predictive models anticipating clinical outcomes and assessing the risk of any personalized medicine approaches. In this article, we have studied the current trends and future perspectives in personalized medicine & toxicology, the role of toxicogenomics in connecting the two fields, and the impact of AI on personalized medicine & toxicology. In this work, we also study the key challenges and limitations in personalized medicine, toxicogenomics, and AI in order to fully realize their potential.

Multi- and Transgenerational Developmental Impairments Are Induced by Decabromodiphenyl Ethane (DBDPE) in Zebrafish Larvae

A novel brominated flame retardant decabromodiphenyl ethane (DBDPE) has become a ubiquitous emerging pollutant; hence, the knowledge of its long-term toxic effects and underlying mechanism would be critical for further health risk assessment. In the present study, the multi- and transgenerational toxicity of DBDPE was investigated in zebrafish upon a life cycle exposure at environmentally relevant concentrations. The significantly increased malformation rate and declined survival rate specifically occurred in unexposed F2 larvae suggested transgenerational development toxicity by DBDPE. The changing profiles revealed by transcriptome and DNA methylome confirmed an increased susceptibility in F2 larvae and figured out potential disruptions of glycolipid metabolism, mitochondrial energy metabolism, and neurodevelopment. The changes of biochemical indicators such as ATP production confirmed a disturbance in the energy metabolism, whereas the alterations of neurotransmitter contents and light-dark stimulated behavior provided further evidence for multi- and transgenerational neurotoxicity in zebrafish. Our findings also highlighted the necessity for considering the long-term impacts when evaluating the health of wild animals as well as human beings by emerging pollutants.

Human functional genomics reveals toxicological mechanism underlying genotoxicants-induced inflammatory responses under low doses exposure

Understanding the toxicological mechanisms of chemicals is essential for accurate assessments of environmental health risks. Inflammation could play a critical role in the adverse health outcomes caused by genotoxicants; however, the toxicological mechanisms underlying genotoxicants-induced inflammatory response are still limited. Here, functional genomics CRISPR screens were performed to enhance the mechanistic understanding of the genotoxicants-induced inflammatory response at low doses exposure. Key genes and pathways associated with the activities of immune cells and the production of cytokines were identified by CRISPR screens of 6 model genotoxicants. Gene network analysis revealed that three genes (TLR10, HCAR2 and TRIM6) were involved in the regulation of neutrophil apoptosis and cytokine release, and TLR10 shared a similar functional pattern with HCAR2 and TRIM6. Furthermore, adverse outcome pathway (AOP) network analysis revealed that TLR10 was involved in the molecular initiating events (MIEs) or key events (KEs) in the inflammatory response AOPs of all the 6 genotoxicants, which provided mechanistic links between TLR10 and genotoxicants-induced inflammation and respiratory diseases. Finally, functional validation tests demonstrated that TLR10 exhibited inhibitory effects on genotoxicants-induced inflammatory responses in both epithelial and immune cells. This study highlights the powerful utility of the integration of CRISPR screen and AOP network analysis in illuminating the toxicological causal mechanisms of environmental chemicals.

Identifying the acute toxicity of contaminated sediments using machine learning models

Ecological risk assessment of contaminated sediment has become a fundamental component of water quality management programs, supporting decision-making for management actions or prompting additional investigations. In this study, we proposed a machine learning (ML)-based approach to assess the ecological risk of contaminated sediment as an alternative to existing index-based methods and costly toxicity testing. The performance of three widely used index-based methods (the pollution load index, potential ecological risk index, and mean probable effect concentration) and three ML algorithms (random forest, support vector machine, and extreme gradient boosting [XGB]) were compared in their prediction of sediment toxicity using 327 nationwide data sets from Korea consisting of 14 sediment quality parameters and sediment toxicity testing data. We also compared the performances of classifiers and regressors in predicting the toxicity for each of RF, SVM, and XGB algorithms. For all algorithms, the classifiers poorly classified toxic and non-toxic samples due to limited information on the sediment composition and the small training dataset. The regressors with a given classification threshold provided better classification, with the XGB regressor outperforming the other models in the classification. A permutation feature importance analysis revealed that Cr, Cu, Pb, and Zn were major contributors to toxicity prediction. The ML-based approach has the potential to be even more useful in the future with the expected increase in available sediment data.

Machine learning in the identification, prediction and exploration of environmental toxicology: Challenges and perspectives

Over the past few decades, data-driven machine learning (ML) has distinguished itself from hypothesis-driven studies and has recently received much attention in environmental toxicology. However, the use of ML in environmental toxicology remains in the early stages, with knowledge gaps, technical bottlenecks in data quality, high-dimensional/heterogeneous/small-sample data analysis and model interpretability, and a lack of an in-depth understanding of environmental toxicology. Given the above problems, we review the recent progress in the literature and highlight state-of-the-art toxicological studies using ML (such as learning and predicting toxicity in complicated biosystems and multiple-factor environmental scenarios of long-term and large-scale pollution). Beyond predicting simple biological endpoints by integrating untargeted omics and adverse outcome pathways, ML development should focus on revealing toxicological mechanisms. The integration of data-driven ML with other methods (e.g., omics analysis and adverse outcome pathway frameworks) endows ML with widely promising application in revealing toxicological mechanisms. High-quality databases and interpretable algorithms are urgently needed for toxicology and environmental science. Addressing the core issues and future challenges for ML in this review may narrow the knowledge gap between environmental toxicity and computational science and facilitate the control of environmental risk in the future.

Machine learning and artificial intelligence in toxicological sciences

Machine learning and artificial intelligence approaches have revolutionized multiple disciplines, including toxicology. This review summarizes representative recent applications of machine learning and artificial intelligence approaches in different areas of toxicology, including physiologically based pharmacokinetic (PBPK) modeling, quantitative structure-activity relationship modeling for toxicity prediction, adverse outcome pathway analysis, high-throughput screening, toxicogenomics, big data and toxicological databases. By leveraging machine learning and artificial intelligence approaches, now it is possible to develop PBPK models for hundreds of chemicals efficiently, to create in silico models to predict toxicity for a large number of chemicals with similar accuracies compared to in vivo animal experiments, and to analyze a large amount of different types of data (toxicogenomics, high-content image data, etc.) to generate new insights into toxicity mechanisms rapidly, which was impossible by manual approaches in the past. To continue advancing the field of toxicological sciences, several challenges should be considered: (1) not all machine learning models are equally useful for a particular type of toxicology data, and thus it is important to test different methods to determine the optimal approach; (2) current toxicity prediction is mainly on bioactivity classification (yes/no), so additional studies are needed to predict the intensity of effect or dose-response relationship; (3) as more data become available, it is crucial to perform rigorous data quality check and develop infrastructure to store, share, analyze, evaluate, and manage big data; and (4) it is important to convert machine learning models to user-friendly interfaces to facilitate their applications by both computational and bench scientists.

Enhancing Data Integration, Interoperability, and Reuse to Address Complex and Emerging Environmental Health Problems

Environmental health sciences (EHS) span many diverse disciplines. Within the EHS community, the National Institute of Environmental Health Sciences Superfund Research Program (SRP) funds multidisciplinary research aimed to address pressing and complex issues on how people are exposed to hazardous substances and their related health consequences with the goal of identifying strategies to reduce exposures and protect human health. While disentangling the interrelationships that contribute to environmental exposures and their effects on human health over the course of life remains difficult, advances in data science and data sharing offer a path forward to explore data across disciplines to reveal new insights. Multidisciplinary SRP-funded teams are well-positioned to examine how to best integrate EHS data across diverse research domains to address multifaceted environmental health problems. As such, SRP supported collaborative research projects designed to foster and enhance the interoperability and reuse of diverse and complex data streams. This perspective synthesizes those experiences as a landscape view of the challenges identified while working to increase the FAIR-ness (Findable, Accessible, Interoperable, and Reusable) of EHS data and opportunities to address them.