Volume 8 Number 1 2026
Advancing Functional Toxicogenomics: Machine Learning and Deep Learning Approaches for Omics-Based Toxicity Assessment
Hares Bin Syed Md. Rifat 1*
Bioinfo Chem 8 (1) 1-8 https://doi.org/10.25163/bioinformatics.8110588
Submitted: 24 July 2026 Revised: 15 September 2026 Accepted: 24 September 2026 Published: 26 September 2026
Chemical safety science is moving away from slow, animal-intensive observation toward prediction grounded in molecular mechanism, and functional toxicogenomics now sits at the centre of that shift. This review examines how high-throughput multi-omics profiling and artificial intelligence are being combined to forecast chemical toxicity in ways that are faster, more human-relevant, and increasingly acceptable to regulators. We first trace how transcriptomic, proteomic, metabolomic, and epigenomic platforms generate dense molecular readouts across human cell lines, organoids, and microphysiological systems, capturing early Molecular Initiating Events and Key Events rather than only late apical outcomes. We then evaluate the computational layer, benchmarking classical machine-learning classifiers—random forests, support vector machines, and gradient boosting—against deep and graph-based architectures, including multi-task neural networks, graph neural networks, and generative models such as Tox-GAN. Reported performance generally climbs as models move from tabular descriptors toward integrated chemical–omics representations, although gains come with real costs in interpretability and data demand. A third thread concerns mechanistic anchoring: how transcriptomic Benchmark Doses, Points of Departure, quantitative in vitro–to–in vivo extrapolation, and physiologically based kinetic modelling convert molecular signals into human-equivalent exposure limits for Next Generation Risk Assessment. Finally, we consider the bottlenecks that still slow adoption—technical noise, batch effects, the “black-box” problem, and fragmented data—and the emerging responses, from Explainable AI and biological knowledge graphs to FAIR, auditable infrastructures such as OpenTox 2.0. Taken together, the evidence suggests a field maturing into a reproducible, mechanism-anchored, and genuinely animal-free discipline, though data standardisation and model transparency remain the decisive hurdles to full regulatory adoption.
Keywords: functional toxicogenomics; multi-omics integration; machine learning; deep learning; adverse outcome pathway; Next Generation Risk Assessment; new approach methodologies; explainable AI
Abdelwahab, A. A., & Hardy, B. (2026). OpenTox 2.0: A perspective on the principles for predictive toxicology and risk assessment applications in the era of integrated new approach methods, computational modelling and artificial intelligence. Edelweiss Connect Preprint, 1–35.
Albanese, A., Swaney, J. M., Yun, D. H., Evans, N. B., Antonucci, J. M., Velasco, S., & Chung, K. (2020). Multiscale 3D phenotyping of human cerebral organoids. Scientific Reports, 10(1), 21487. https://doi.org/10.1038/s41598-020-78130-7
Alnasser, S. M. (2025). Evolution of genetic toxicology: High-throughput screening and data-driven computational models for genotoxicity assessment. BioData Mining, 18(33), 1–19. https://doi.org/10.1186/s13040-025-00447-8
Ankley, G. T., Bennett, R. S., Erickson, R. J., Hoff, D. J., Hornung, M. W., Johnson, R. D., Mount, D. R., Nichols, J. W., Russom, C. L., & Schmieder, P. K. (2010). Adverse outcome pathways: A conceptual framework to support ecotoxicology research and risk assessment. Environmental Toxicology and Chemistry, 29(3), 730–741. https://doi.org/10.1002/etc.34
Aslam, B., Basit, M., Nisar, M. A., Khurshid, M., & Rasool, M. H. (2017). Proteomics: Technologies and their applications. Journal of Chromatographic Science, 55(2), 182–196. https://doi.org/10.1093/chromsci/bmw167
Capone, P., Chiarella, P., & Sisto, R. (2023). Advanced technologies in genomic toxicology: Current trend and future directions. Current Opinion in Toxicology, 36, 100444. https://doi.org/10.1016/j.cotox.2023.100444
Chen, X., Roberts, R., Tong, W., & Liu, Z. (2022). Tox-GAN: An artificial intelligence approach alternative to animal studies—A case study with toxicogenomics. Toxicological Sciences, 186(2), 242–259. https://doi.org/10.1093/toxsci/kfab157
Crute, C., et?al.?(2025). Advancing toxicology through Next Generation Risk Assessment and multi-omics approaches: The genesis of the Omics for Assessing Signatures for Integrated Safety Consortium. Toxicological Sciences, 208(2), 225–237. https://doi.org/10.1093/toxsci/kfaf128
Cui, H., Miao, S., Esworthy, T., Zhou, X., Lee, S., Liu, C., & Zhang, L. G. (2018). 3D bioprinting for cardiovascular regeneration and pharmacology. Advanced Drug Delivery Reviews, 132, 252–269. https://doi.org/10.1016/j.addr.2018.07.014
Deepika, D., & Kumar, V. (2023). The role of “physiologically based pharmacokinetic model (PBPK)” new approach methodology (NAM) in pharmaceuticals and environmental chemical risk assessment. International Journal of Environmental Research and Public Health, 20(4), 3473. https://doi.org/10.3390/ijerph20043473
Dent, M. P., Amaral, R. T., Da Silva, P. A., Hirose, A., Kopp-Schneider, A., Buesen, R., & Carmichael, P. L. (2021). Next generation risk assessment for safety decision-making. Regulatory Toxicology and Pharmacology, 125, 105026. https://doi.org/10.1016/j.yrtph.2021.105026
Dimitrov, S. D., Diderich, R., Sobanski, T., Pavlov, T. S., Chankov, G. V., Chapkanov, A. S., & Mekenyan, O. G. (2016). QSAR Toolbox—Workflow and major functionalities. SAR and QSAR in Environmental Research, 27(3), 203–219. https://doi.org/10.1080/1062936X.2015.1136680
Djidrovski, I. (2026). ToxMCP: Guardrailed, auditable agentic workflows for computational toxicology. bioRxiv. https://doi.org/10.1101/2026.02.06.703989
ENVESOME Consortium. (2024). An integrative exposome framework for environmental health risk assessment and pathway-based dose–response analysis. Frontiers in Public Health, 12, 116195.
Ewart, L., Fabre, K., Chakilam, A., & LeCluyse, E. (2018). Bridging the gap between in vitro and in vivo: Dose and schedule predictions for an organotypic human liver microphysiology system. Toxicology in Vitro, 45, 65–74. https://doi.org/10.1016/j.tiv.2017.08.004
Farmahin, R., Williams, A., Evenson, L. O., & Yauk, C. L. (2017). Recommended practices for the analysis of transcriptomic benchmark dose data. Regulatory Toxicology and Pharmacology, 91, S14–S26. https://doi.org/10.1016/j.yrtph.2017.09.014
Fortino, V., Kinaret, P., Fyhrquist, N., Alenius, H., & Greco, D. (2014). A robust and accurate method for feature selection and prioritization from multi-class OMICs data. PLoS ONE, 9(12), e114627. https://doi.org/10.1371/journal.pone.0114627
Hardy, B., Douglas, N., Helma, C., Rautenberg, M., Jeliazkova, N., Jeliazkov, V., & Willighagen, E. (2010). Collaborative development of predictive toxicology applications. Journal of Cheminformatics, 2, 7. https://doi.org/10.1186/1758-2946-2-7
Harrill, J. A., Viant, M. R., Yauk, C. L., Sachana, M., Gant, T. W., Auerbach, S. S., Beger, R. D., Bouhifd, M., O’Brien, J., & Burgoon, L. (2021). Progress towards an OECD reporting framework for transcriptomics and metabolomics in regulatory toxicology. Regulatory Toxicology and Pharmacology, 125, 105020. https://doi.org/10.1016/j.yrtph.2021.105020
Hemmerich, J., & Ecker, G. F. (2020). In silico toxicology: From structure–activity relationships towards deep learning and adverse outcome pathways. WIREs Computational Molecular Science, 10(4), e1475. https://doi.org/10.1002/wcms.1475
Hinterberger, M., & Stelmach, R. (2026). Toward human-relevant toxicology: A decision-oriented roadmap integrating modern tools for the safety evaluation of novel foods and functional ingredients. Archives of Toxicology, 100, 1–28.
Hussein, M. N. H. (2026). Toxicogenomics and personalized medicine: Applications, key technologies, and AI integration. Ain Shams Journal of Forensic Medicine and Clinical Toxicology, 46(1), 52–68.
Igarashi, Y., Nakatsu, N., Yamashita, T., Ono, A., Ohno, Y., Urushidani, T., & Yamada, H. (2015). Open TG-GATEs: A large-scale toxicogenomics database. Nucleic Acids Research, 43(D1), D921–D927. https://doi.org/10.1093/nar/gku955
Johnson, W. E., Li, C., & Rabinovic, A. (2007). Adjusting batch effects in microarray expression data using empirical Bayes methods. Biostatistics, 8(1), 118–127. https://doi.org/10.1093/biostatistics/kxj037
Kim, D., & Choi, J. (2025). Big data and artificial intelligence in toxicology: Current status and future perspectives. Environmental Analysis Health and Toxicology, 40, e2025s07. https://doi.org/10.5620/eaht.2025s07
Li, H., Yuan, H., Middleton, A., Li, J., Nicol, B., Carmichael, P., & Tran, L. (2021). Next generation risk assessment (NGRA): Bridging in vitro points-of-departure to human safety assessment using physiologically-based kinetic (PBK) modelling. Toxicology in Vitro, 74, 105171. https://doi.org/10.1016/j.tiv.2021.105171
Lynch, C., Sakamuru, S., Ooka, M., & Xia, M. (2024). High throughput screening to advance in vitro toxicology: Accomplishments, challenges, and future directions. Annual Review of Pharmacology and Toxicology, 64, 215–236. https://doi.org/10.1146/annurev-pharmtox-112122-104310
Magurany, K. A., Chang, X., Clewell, R., Coecke, S., Haugabrooks, E., & Marty, S. (2023). A pragmatic framework for the application of new approach methodologies in one health toxicological risk assessment. Toxicological Sciences, 192(2), 155–177. https://doi.org/10.1093/toxsci/kfad018
Mayr, A., Klambauer, G., Unterthiner, T., & Hochreiter, S. (2016). DeepTox: Toxicity prediction using deep learning. Frontiers in Environmental Science, 4, 2. https://doi.org/10.3389/fenvs.2016.00002
Meier, M. J., Froetschl, R., & Yauk, C. L. (2024). Outcome of IWGT workshop on transcriptomic biomarkers for genotoxicity: Key bioinformatics approaches and recommendations. Environmental and Molecular Mutagenesis, 65(4), 135–151. https://doi.org/10.1002/em.22342
Mirshafiei, M., Rashedi, H., Yazdian, F., Rahdar, A., & Baino, F. (2024). Advancements in tissue and organ 3D bioprinting: Current techniques, applications, and future perspectives. Materials & Design, 240, 112853. https://doi.org/10.1016/j.matdes.2024.112853
Moreau, M., Simms, L., Andersen, M. E., Trelles Sticken, E., Wieczorek, R., Pour, S. J., Chapman, F., Roewer, K., Otte, S., & Fisher, J. (2024). Use of quantitative in vitro to in vivo extrapolation (QIVIVE) for the assessment of non-combustible next-generation product aerosols. Frontiers in Toxicology, 6, 1373325. https://doi.org/10.3389/ftox.2024.1373325
Multi-omics integration strategies and proteomic workflows in toxicology. (2024). Proteomes, 12(1), 25. https://doi.org/10.3390/proteomes12010025
Nguyen, X. H., & Yoo, J. (2025). Current status and future prospects of toxicity assessment using organoids. Toxicological Research, 41(4), 325–333. https://doi.org/10.1007/s43188-025-00298-9
OECD. (2023). Guidance document on reporting elements for the regulatory use of omics data from laboratory-based toxicology studies. OECD Publishing. https://doi.org/10.1787/6bb2e6ce-en
Pathan, S., Perez Santin, E., & Jia, X. (2025). AI-driven target prediction and network toxicology in modern hazard assessment. Computational Toxicology, 30, 100312.
Pavan, M., & Kumar, V. (2025). New approach methodologies (NAMs) in chemical risk assessment: Opportunities and challenges for regulatory acceptance. Frontiers in Toxicology, 7, 1632941. https://doi.org/10.3389/ftox.2025.1632941
Phillips, J. R., Svoboda, D. L., Tandon, A., Patel, S., Sedykh, A., Mav, D., Kuo, B., Yauk, C. L., Yang, L., & Thomas, R. S. (2019). BMDExpress 2: Enhanced transcriptomic dose-response analysis workflow. Bioinformatics, 35(10), 1780–1782. https://doi.org/10.1093/bioinformatics/bty878
Ramirez-Hincapie, S., Štampar, M., & Žegura, B. (2023). High-throughput targeted metabolomics platform for hepatotoxicity screening in HepG2 cells. Archives of Toxicology, 97(6), 1723–1738. https://doi.org/10.1007/s00204-023-03474-8
Ren, X., Zhang, Y., & Liu, S. (2025). Using environmental mixture exposure-triggered biological knowledge-driven machine learning to predict health risks. Environmental Science & Technology, 59(4), 1234–1245. https://doi.org/10.1021/acs.est.5c05389
Richard, A. M., Judson, R. S., Houck, K. A., Grulke, C. M., Volarath, P., & Thomas, R. S. (2016). ToxCast chemical landscape: Paving the road to 21st century toxicology. Chemical Research in Toxicology, 29(8), 1225–1251. https://doi.org/10.1021/acs.chemrestox.6b00135
Samantasinghar, A., Sunildutt, N., Ahmed, F., & Choi, J. H. (2025). Revolutionizing biomedical research: Unveiling the power of microphysiological systems with advanced assays, integrated sensor technologies, and real-time monitoring. ACS Omega, 10(10), 9869–9889. https://doi.org/10.1021/acsomega.4c11227
Santorsola, M., & Lescai, F. (2023). The promise of explainable deep learning for omics data analysis: Adding new discovery tools to AI. New Biotechnology, 77, 1–11. https://doi.org/10.1016/j.nbt.2023.06.002
Serra, A., Saarimäki, L. A., Fratello, M., Marwah, V. S., & Greco, D. (2020). Transcriptomics in toxicogenomics, part III: Data modelling for risk assessment. Nanomaterials, 10(4), 708. https://doi.org/10.3390/nano10040708
Short, S., Green Etxabe, A., Swart, E., Rivetti, C., Campos, B., Krishnan, R., Kille, P., & Spurgeon, D. J. (2026). Exploiting omic data to advance predictive ecotoxicology. Environmental Science & Technology, 60(34), 23642–23660. https://doi.org/10.1021/acs.est.6c01198
Singh, A. V., Bhardwaj, P., Laux, P., Pradeep, P., Busse, M., Luch, A., Hirose, A., Osgood, C. J., & Stacey, M. W. (2026). AI and ML-based risk assessment of chemicals: Predicting carcinogenic risk from chemical-induced genomic instability. Frontiers in Toxicology, 6, 1461587. https://doi.org/10.3389/ftox.2024.1461587
Su, R., Wu, H., Xu, B., Liu, X., & Wei, L. (2018). Developing a multi-dose computational model for drug-induced hepatotoxicity prediction based on toxicogenomics data. IEEE/ACM Transactions on Computational Biology and Bioinformatics, 16(4), 1231–1239. https://doi.org/10.1109/TCBB.2018.2858756
Tang, K., Ji, X., Zhou, M., & Li, S. (2021). Rank-in: Enabling integrative analysis across microarray and RNA-seq for cancer and toxicity. Nucleic Acids Research, 49(17), e99. https://doi.org/10.1093/nar/gkab554
Tanveer, T., Shaheen, K., Parveen, S., Misbah, Z. T., Babar, M. M., & Gul, A. (2018). Omics-based bioengineering in environmental biotechnology. In Omics technologies and bio-engineering: Towards improving quality of life (pp. 353–368). Academic Press/Elsevier. https://doi.org/10.1016/B978-0-12-815870-8.00019-X
Ukaegbu, O. F., Zhang, J., & Rouquié, D. (2026). Environmental perturbation-aware virtual cells: Advancing computational toxicology for chemical safety assessment. ALTEX—Alternatives to Animal Experimentation, 43(1), 1–22. https://doi.org/10.14573/altex.2606062
Vasant, M., Vinken, M., & Arnesdotter, E. (2024). Omics-based input and output in the development and use of adverse outcome pathways for human health risk assessment. International Journal of Molecular Sciences, 25(20), 11154. https://doi.org/10.3390/ijms252011154
Vinken, M. (2024). Adverse outcome pathway networks as the basis for the development of new approach methodologies: Liver toxicity as a case study. Current Opinion in Toxicology, 40, 100504. https://doi.org/10.1016/j.cotox.2024.100504
Wan, L., Wu, J., & Kong, X. (2024). Application of organoid technology in human health risk assessment. Environment International, 188, 108744. https://doi.org/10.1016/j.envint.2024.108744
Wang, H., Liu, R., Schyman, P., & Wallqvist, A. (2019). Deep neural network models for predicting chemically induced liver toxicity endpoints from transcriptomic responses. Frontiers in Pharmacology, 10, 42. https://doi.org/10.3389/fphar.2019.00042
Wang, T., Shao, W., Huang, Z., Tang, H., Zhang, J., Ding, Z., & Huang, K. (2021). MOGONET integrates multi-omics data using graph convolutional networks allowing patient classification and biomarker identification. Nature Communications, 12, 3445. https://doi.org/10.1038/s41467-021-23774-w
Wilkinson, M. D., Dumontier, M., Aalbersberg, I. J., Appleton, G., Axton, M., Baak, A., & Mons, B. (2016). The FAIR Guiding Principles for scientific data management and stewardship. Scientific Data, 3, 160018. https://doi.org/10.1038/sdata.2016.18
Williams, A. J., Grulke, C. M., Edwards, J., McEachran, A. D., Mansouri, K., & Richard, A. M. (2017). The CompTox Chemistry Dashboard: A community data resource for environmental chemistry. Journal of Cheminformatics, 9(1), 61. https://doi.org/10.1186/s13321-017-0247-6
Wu, L., Yan, B., Han, J., Li, R., Xiao, J., He, S., & Bo, X. (2023). TOXRIC: A comprehensive database of toxicological data and benchmarks. Nucleic Acids Research, 51(D1), D1432–D1445. https://doi.org/10.1093/nar/gkac955
Wu, W., Qian, J., Liang, C., Yang, J., Ge, G., Zhou, Q., & Guan, X. (2023). GeoDILI: A robust and interpretable model for drug-induced liver injury prediction using graph neural network-based molecular geometric representation. Chemical Research in Toxicology, 36(11), 1717–1730. https://doi.org/10.1021/acs.chemrestox.3c00220
Xu, Z., & Yang, M. (2026). Omics technologies in disinfection byproduct toxicology: Toward mechanistic and extrapolative frameworks. Environment & Health, 3(10), 1139–1153. https://doi.org/10.1021/envhealth.5c00874
Yang, R., Shi, Z., Li, Y., Huang, X., Li, Y., Li, X., ... & Li, X. (2024). Research focus and emerging trends of the gut microbiome and infant: a bibliometric analysis from 2004 to 2024. Frontiers in Microbiology, 15, 1459867.
Yi, X., et?al.?(2023). AI data pipeline for toxicity prediction in complex food and environmental matrices. Food Quality and Safety, 10, fyaf052. https://doi.org/10.1093/fqsafe/fyaf052
Zushin, P. H., Mukherjee, S., & Wu, J. C. (2023). FDA Modernization Act 2.0: Transitioning beyond animal models with human cells, organoids, and AI/ML-based approaches. Journal of Clinical Investigation, 133(21), e175824. https://doi.org/10.1172/JCI175824
Network Medicine Approaches to Long COVID: Integrating Multi-Omic Data to Redefine Disease Taxonomy
Reproducibility, Algorithmic Bias, and Clinical Readiness in Medical Imaging AI: A Lifecycle-Based Synthesis and Multi-Institutional Validation Protocol
AI-Driven Target Identification Reshapes Early-Stage Drug Discovery in Triple-Negative Breast Cancer