Scienta Lab

Scienta Lab develops a predictive modeling platform focused on human immune response, with the goal of improving translational research in immunology and inflammation. The company addresses a core challenge in drug development: approximately 70% of immunology drug candidates fail to translate from preclinical models into real clinical benefit for patients. Scienta's approach introduces a computational layer that converts early biological data into predictions of clinical outcomes, covering the full arc from target identification through clinical trial design.

The platform serves biopharmaceutical teams working across autoimmune, inflammatory, and immune-mediated diseases, and is supported by a combination of proprietary AI models, curated datasets, and in-house scientific expertise in immunology and data science.

Core Products and Technology

• EVA — Scienta's foundation model, built specifically for immuno-inflammatory diseases. EVA is trained on more than 500,000 transcriptomics samples and 20 million histology tiles, learning shared immune mechanisms across multiple conditions, tissues, and species. It is described as a cross-species, multimodal model and is reported to double clinical trial success rates while outperforming existing biological AI benchmarks.
• ImmunAtlas — Scienta Lab's proprietary training dataset used to develop EVA. It integrates multi-omic, clinical, and histopathology data from more than 96,000 patient and biosample profiles, with ongoing expansion as new datasets are incorporated. The broader platform references over 200,000 patient and biosample profiles across diseases.
• Expert guidance — Each project is supported by immunologists and data scientists who translate model outputs into concrete development decisions, spanning target selection through trial strategy.

Key Capabilities and Modules

• Target and mechanism prioritization — Targets and mechanisms of action are ranked by predicted probability of clinical success, drawing on integrated multi-omic data, disease biology, and early experimental signals. This is intended to help teams prioritize programs before committing major resources, with benefits including early estimation of translational risk and stronger preclinical investment decisions.
• Preclinical-to-human translation — Human efficacy is predicted from preclinical and target-level data by simulating drug impact on immune pathways and mapping these effects to expected clinical outcomes. This supports objective candidate comparison and early go/no-go decisions ahead of IND filing, as well as candidate ranking and portfolio optimization.
• Patient stratification and trial optimization — Patient-level response is predicted from baseline and post-treatment data to identify biomarkers and support trial enrichment strategies. This module is aimed at improving signal detection and accelerating decision-making in Phase II–III development, with applications in biomarker discovery and trial design optimization.

Therapeutic Areas Covered

• Immunology
• Rheumatology
• Endocrinology
• Dermatology
• Immuno-inflammation
• Gastroenterology
• Neurology
• Respiratory

Platform Scale and Performance

• 65% accuracy reported for clinical trial success predictions
• Coverage of more than 60 immuno-inflammatory diseases
• Integration of data from more than 70 tissues across datasets
• More than 200,000 patient and biosample profiles across diseases

Partners, Customers, and Scientific Engagement

• OSE Immunotherapeutics has partnered with Scienta Lab, with the stated aim of accelerating innovation and supporting development of more personalized treatments, as noted by Nicolas Poirier, Scientific Director at OSE Immunotherapeutics.
• Pr. Dennis McGonagle, Professor of Investigative Rheumatology at the Leeds Institute of Rheumatic and Musculoskeletal Medicine (UK), has noted that Scienta Lab's technology advances understanding of immune pathways in immune-mediated and inflammatory diseases.
• Scienta has presented scientific work at the 21st Congress of ECCO (European Crohn's and Colitis Organisation) in 2026, including an abstract on using machine learning to predict clinical efficacy of IBD drug candidates from preclinical data.
• The company is also participating in the BIO International Convention.

Scienta Lab's published work includes a paper introducing EVA as a universal model of the immune system, as well as blog content and conference abstracts addressing the translational gap in immunology drug development, particularly in inflammatory bowel disease.