Scienta Unveils AI Model to Speed Drug Discovery
Scienta has developed a new AI model named EVA designed to accelerate drug discovery for aging-linked inflammation, according to Longevity Technology. The model aims to bridge the gap between preclinical research and patient therapies. This development is part of a broader trend using AI to find new treatments for age-related diseases.
- The Paris-based startup Scienta Lab developed EVA as a multimodal AI, meaning it integrates diverse biological data types—like gene activity (transcriptomics), tissue samples (histology), and protein levels—from both human and animal studies to improve its predictions. Its model was pre-trained on over 500,000 samples to understand the complexities of the immune system. - This model operates in a competitive landscape of "longevity-focused" startups that also use AI for drug discovery, including Gero, Insilico Medicine, and BenevolentAI, which have formed partnerships with major pharmaceutical companies like Sanofi and Novartis. The field is attracting significant capital, with tech leaders like OpenAI's Sam Altman and Amazon's Jeff Bezos investing in companies such as Retro Biosciences and Altos Labs. - Venture capital funding for AI-driven biotech startups has been volatile but strong, rebounding to approximately $6.7 billion in 2024 after a dip in 2023. Mega-rounds for companies like Xaira Therapeutics ($1 billion) and Isomorphic Labs ($600 million) underscore investor confidence in using AI to accelerate the years-long drug development process. - For consumer health apps, the long-term value of such advanced AI models lies in translating their discoveries into personalized wellness insights. Data from wearables, which can be integrated via APIs from Apple HealthKit, Fitbit, and Oura, could eventually be combined with genomic data to offer tailored health recommendations based on the biological pathways identified by models like EVA. - Handling the sensitive data required to train these AI models raises significant regulatory hurdles; consumer-facing health apps must navigate a complex web of laws including HIPAA, which governs protected health information (PHI), and state-level rules like the California Privacy Rights Act (CPRA). Non-compliance can lead to severe fines and loss of investor trust. - The co-founder and CTO of Scienta Lab is Julien Duquesne, who has stated that the goal of EVA is not to replace researchers but to act as a "co-pilot" to help them make better-informed decisions at every stage of drug development. This reflects a common strategy in the industry, where AI is positioned as a tool to augment, not replace, human scientific expertise.
