The OASIS Consortium: Can Multi-Omics Improve Chemical Safety Assessment?

Summary - How is Multi-Omics Changing Toxicology? 

The OASIS Consortium is a global initiative that combines multi-omics technologies, like transcriptomics, proteomics, and imaging-based Cell Painting, to build more predictive, human-relevant toxicology models. By comparing animal models to scalable, data-rich assays, OASIS aims to accelerate drug and chemical safety evaluation, cut costs, and improve regulatory confidence in Next-Generation Risk Assessment approaches. 

Are animal models a reliable predictor of chemical safety? 

Toxicology relies heavily on animal testing to predict how chemicals and drugs might affect human health. But while animal models have shaped regulatory frameworks, their predictive power for human biology remains limited due to differences in metabolism, gene regulation, and physiology. Approximately one-third of drug candidates fail in the first phase of clinical trials, often due to undetected toxicity in animal models, leading to stalled developmental pipelines and wasted investment. 

The international Omics for Assessing Signatures for Integrated Safety (OASIS) Consortium now aims to tackle these issues head-on through a Next Generation Risk Assessment framework that harnesses high-content, human-relevant, and data-rich systems, such as Cell Painting, high-throughput transcriptomics, and proteomics. Alithea Genomics supports the OASIS Consortium as a technology partner by providing scalable transcriptomic solutions such as MERCURIUSTM DRUG-seq.  

Read on to learn how the OASIS Consortium is developing the tools to improve safety assessment by adopting more predictive, efficient, and cost-effective toxicological assays, all while reducing animal use. 

What Are ‘New Approach Methodologies’ in Toxicology? 

Approximately 92% of drugs fail in clinical trials despite demonstrating sufficient efficacy and safety in current preclinical models (1). This high failure rate now means the cost to bring a new drug to market has risen to over two billion US dollars, depending on therapeutic class and estimate criteria, with a typical discovery and development process of between 10 and 15 years (2).  

At least some of this excessive cost and time is associated with suboptimal early toxicological assessment (3). As a result, a shift in early-stage toxicology from traditional single-endpoint animal models to more relevant human-centric models that predict liver toxicity of potential new medicines stands to benefit not only pharma but also the agritech and industrial chemical sectors, where effective safety assessment is also paramount (3).  

The driving forces behind this push for Next Generation Risk Assessment are innovative ‘New Approach Methodologies’, such as Cell Painting, high-throughput transcriptomics, like DRUG-seq and TempO-seq and novel proteomics methods (3).  

What Are the Main Research Goals of the OASIS Consortium?   

Launched in 2023 by the Health and Environmental Sciences Institute (HESI Global), The Broad Institute, and over 60 global partners spanning industry, government, academia, and other research institutes, with support from the Massachusetts Life Sciences Center, the OASIS Consortium intends to address three core questions:  

1. How effectively can different omics technologies capture toxicity mechanisms in rat and human cell models and predict adverse effects across pharmaceuticals, agrochemicals, and industrial chemicals, and does Cell Painting add value to omics to improve in vitro to in vivo translation? 
   
   
2. How do different cell models, including cell lines, primary cells, and organ-on-chip systems, compare in predicting human outcomes, and what can comparisons with rat models reveal about species-specific differences? 
   
   
3. How well do in vitro responses, combined with internal exposure estimates (e.g. quantitative in vitro to in vivo extrapolation), inform in vivo dose–response relationships, and how do points of departure from different omics and Cell Painting approaches compare? 
   
   

How Does the OASIS Consortium Study Chemical Safety? 

To answer these three questions, the OASIS Consortium has established a strategy that includes different chemical classes (pharmaceuticals, plant protection products, and industrial or personal care chemicals); various cellular systems (U2OS, primary human hepatocytes, HepG2, iPSC-derived human liver organoids, and others); and multiple high-content in vitro methods (transcriptomics, proteomics, and Cell Painting) to establish internal exposure estimates, followed by in vivo assessments in rats and humans (Fig. 1). 

Figure 1. Schematic of the approach used by the OASIS Consortium. Figure taken from (3). 

Using this strategy, the OASIS Consortium will link chemical structures to different omics measurements as predictive variables to identify which in vitro readouts best predict in vivo toxicity across species and doses (3). The goal is to build confidence in these New Approach Methodologies and define best practices for assessing liver toxicity, with broader applications across toxicology.  

Initially, the OASIS Consortium intends to improve the prediction of liver toxicity, as it’s already a key endpoint for risk assessment, but the consortium hopes the findings will be applicable to broader toxicological predictions. As the project progresses, the intention is also to use machine learning strategies to integrate the different data modalities from Cell Painting, transcriptomics and proteomics to bridge the gap between in vitro and in vivo models and enable better prediction of human safety risks (4). 

Which Multi-Omics Technologies are Used in the OASIS Consortium for in vitro Chemical Safety Assessments? 

Until recently, there was a lack of scalable assay systems that retain physiological relevance while providing multiplexed, high-dimensional readouts suitable for toxicological insights.  

The three technologies evaluated by the OASIS Consortium are: 

1. Transcriptomics – for gene expression profiling

Traditional transcriptomics approaches, like RNA sequencing (RNA-seq), are too expensive and time-consuming to perform at the screening scale required by the OASIS Consortium. Similarly, while high-throughput transcriptomics has determined perturbation responses in many research studies and is gradually gaining regulatory acceptance, there has so far been limited use in the pharmaceutical or agrochemical industries for regulatory decision-making due to challenges with standardization, data interpretation, and the complexity of integrating transcriptomic data with existing regulatory frameworks (5). 

To address this, scalable targeted RNA-seq methods, like TempO-seq, that rely on probe hybridization to mRNA molecules allow users to screen hundreds of samples for defined gene panels. The limitation is that novel transcripts are missed, and only a small fraction of a transcript is assessed. 

In contrast, 3’ mRNA-seq technologies, like DRUG-seq, are RNA-extraction free and utilize unique sample barcoding of mRNA poly(A)-tails to allow multiplexing of hundreds to thousands of samples to generate unbiased, robust expression readouts, even for novel transcripts (6). New technologies like MERCURIUS™ Total DRUG-seq and MERCURIUS™ Full Length DRUG-seq take this a step further and now provide full-length transcript readouts for both the non-coding and protein-coding transcriptomes at scale for users wanting to assess fusion transcripts, alternative splicing, and transcript isoform levels. 

Want to explore MERCURIUSTM DRUG-seq data for yourself? Access our free Hap1, HeLa, and HepG2 demo data. 

2. Proteomics – for protein-level changes 

Similar to traditional RNA-seq methods, mass spectrometry or ELISA-based proteomics methods were previously too expensive and low-throughput for broad screens. Now, new methods are cost-effective enough to analyze sufficient numbers of compounds to quantitatively evaluate whether the technologies can predict toxicity in a dose and time-resolved way (7). 

Proteomics data can also be generally analyzed in a similar way to transcriptomics data, presenting an opportunity to directly compare methods assessing different biological layers and directly evaluate which technology is most useful. 

3. Cell Painting – for morphological and phenotypic profiling 

While not an omics technology, the high-content, high-throughput nature of Cell Painting makes it well-suited to screening scale use. It is a phenotypic profiling assay that labels cells with six fluorescent probes to visualize eight different cellular components by microscopy (8).  

 Advanced image analysis algorithms then measure thousands of features, like cell shape and texture, to provide robust insight into cellular state from mechanisms of chemical cytotoxicity to the detection of shared mechanisms of action (9,10). Unlike the transcriptomics and proteomics methods, which measure signals in bulk cell populations, Cell Painting can also provide single-cell resolution useful for investigating heterogeneous populations.  

What Progress Has the OASIS Consortium Made So Far? 

Established in 2023, the consortium is already analyzing over 1,500 compounds with known in vivo toxicity data from rats or humans in public databases like DILIlist and ToxRefDB (11,12). Industry partners have also provided data for an additional 200 compounds.  

This initial benchmarking strategy will assess how well in vitro points of departure and in silico exposure estimates from different omics and cell models align with known in vivo data, using statistical thresholds to define predictive reliability. 

Benchmarking is crucial at this stage, as it will help regulators, toxicologists, and industry scientists to trust and adopt multi-omics-based methods with greater confidence. It will also provide insight into their limitations and will direct future investment to overcome any technological challenges. 

What Are the Benefits of the OASIS Consortium for Industry and Regulators? 

As the OASIS Consortium is a multi-sector, pre-competitive collaboration, the project will make all data and results publicly accessible upon publication. The OASIS Consortium has already published a preprint in early 2025 with Cell Painting data that assessed cytotoxicity and mode of action in primary human hepatocytes, clearly demonstrating the project's momentum (13). 

Ultimately, the OASIS Consortium looks set to profoundly change the face of translational toxicology. By critically evaluating New Approach Methodologies that could allow faster, cheaper, and more biologically relevant safety assessment frameworks, the pharmaceutical, agrochemical, and industrial chemical sectors will hopefully reduce animal usage, shorten developmental pipelines, boost clinical success rates, and develop safer products for all.  

Want to learn more about how Alithea Genomics and MERCURIUS™ DRUG-seq could boost your toxicological safety assessment pipelines? Contact us for a free consultation. 

Key Takeaways - The OASIS Consortium Explained 

• What is the OASIS Consortium? 
  A global pre-competitive initiative led by HESI Global and The Broad Institute to evaluate multi-omics-based safety testing across pharmaceuticals, agrochemicals, and industrial chemicals. 

• Why is a new approach needed in toxicology? 
  Traditional animal studies often fail to predict human-specific toxicity, causing costly late-stage drug failures. 

• Which technologies are being tested? 
  Cell Painting, high-throughput transcriptomics (e.g., DRUG-seq and MERCURIUS™ DRUG-seq), and proteomics for mechanistic and predictive insight. 

• How does OASIS improve risk assessment? 
  By combining omics data with in-silico exposure modeling to create a reliable link between in-vitro assays and real human outcomes. 

• Who are the partners? 
  Over 60 organizations from academia, industry, and government—including Alithea Genomics, providing advanced RNA-sequencing technology. 

• What’s the ultimate goal? 
  To validate human-relevant, multi-omics methods that can replace animal testing and guide future chemical safety regulations. 

References 

1. Mullard A. Parsing clinical success rates. Nature Reviews Drug Discovery. 2016 Jul 1;15(7):447-8. 
2. Measuring the return from pharmaceutical innovation | Deloitte US [Internet]. Deloitte. 2024. Available from: https://www.deloitte.com/us/en/Industries/life-sciences-health-care/articles/measuring-return-from-pharmaceutical-innovation.html 
3. Rouquié D, Bender A, Cheah J, Crute CE, Dalmas D, Ewald J, Fullerton A, Harrill JA, Kadri S, Kleinstreuer N, Kramer N. The OASIS Consortium: integrating multi-omics technologies to transform chemical safety assessment. Toxicological Sciences. 2025 Sep 15:kfaf128. 
4. Seal S, Mahale M, García-Ortegón M, Joshi CK, Hosseini-Gerami L, Beatson A, Greenig M, Shekhar M, Patra A, Weis C, Mehrjou A. Machine Learning for Toxicity Prediction Using Chemical Structures: Pillars for Success in the Real World. Chemical research in toxicology. 2025 May 2;38(5):759-807. 
5. Gant TW, Auerbach SS, Von Bergen M, Bouhifd M, Botham PA, Caiment F, Currie RA, Harrill J, Johnson K, Li D, Rouquie D. Applying genomics in regulatory toxicology: a report of the ECETOC workshop on omics threshold on non-adversity. Archives of toxicology. 2023 Aug;97(8):2291-302. 
6. Ye C, Ho DJ, Neri M, Yang C, Kulkarni T, Randhawa R, Henault M, Mostacci N, Farmer P, Renner S, Ihry R. DRUG-seq for miniaturized high-throughput transcriptome profiling in drug discovery. Nature communications. 2018 Oct 17;9(1):4307. 
7. Zecha J, Bayer FP, Wiechmann S, Woortman J, Berner N, Müller J, Schneider A, Kramer K, Abril-Gil M, Hopf T, Reichart L. Decrypting drug actions and protein modifications by dose-and time-resolved proteomics. Science. 2023 Apr 7;380(6640):93-101. 
8. Seal S, Trapotsi MA, Spjuth O, Singh S, Carreras-Puigvert J, Greene N, Bender A, Carpenter AE. A decade in a systematic review: The evolution and impact of cell painting. bioRxiv. 2024 May 7. 
9. Way GP, Natoli T, Adeboye A, Litichevskiy L, Yang A, Lu X, Caicedo JC, Cimini BA, Karhohs K, Logan DJ, Rohban MH. Morphology and gene expression profiling provide complementary information for mapping cell state. Cell systems. 2022 Nov 16;13(11):911-23. 
10. Nyffeler J, Willis C, Harris FR, Foster MJ, Chambers B, Culbreth M, Brockway RE, Davidson-Fritz S, Dawson D, Shah I, Friedman KP. Application of Cell Painting for chemical hazard evaluation in support of screening-level chemical assessments. Toxicology and applied pharmacology. 2023 Jun 1;468:116513. 
11. Chen M, Suzuki A, Thakkar S, Yu K, Hu C, Tong W. DILIrank: the largest reference drug list ranked by the risk for developing drug-induced liver injury in humans. Drug discovery today. 2016 Apr 1;21(4):648-53. 
12. Mezencev R, Feshuk M, Kolaczkowski L, Peterson GC, Zhao QJ, Watford S, Weaver JA. The association between histopathologic effects and liver weight changes induced in mice and rats by chemical exposures: an analysis of the data from Toxicity Reference Database (ToxRefDB). Toxicological Sciences. 2024 Aug 1;200(2):404-13. 
13. Ewald JD, Titterton KL, Bäuerle A, Beatson A, Boiko DA, Cabrera ÁA, Cheah J, Cimini BA, Gorissen B, Jones T, Karczewski KJ. Cell Painting for cytotoxicity and mode-of-action analysis in primary human hepatocytes. bioRxiv. 2025 Jan 24.