DenovAI Biotech

DenovAI Biotech is an Israeli biotech startup founded in 2023 and operated through the AION Labs venture studio in Rehovot, Israel. The company addresses a fundamental challenge in drug discovery: the slow, expensive, hit-to-lead optimization cycle required to discover therapeutic antibodies and protein binders. DenovAI's innovation is a computational platform that combines generative AI and molecular biophysics to design therapeutic proteins "de novo"—from scratch—without requiring prior knowledge of binder structures, large experimental screening campaigns, or iterative optimization cycles. This represents a potential paradigm shift in drug discovery timelines and the breadth of targets that can be addressed with biological therapeutics.

DenovAI's founding team brings deep expertise in computational biology and protein design. CEO Kashif Sadiq is a computational biologist with protein design expertise; COO Tal Leibovich-Rivkin brings operational leadership; and Head of Computational Biology Yoav Raz contributes deep technical knowledge in biophysics and algorithm design. The team is grounded in rigorous computational and experimental biology, positioning the company at the intersection of machine learning, structural biology, and pharmaceutical development.

The platform integrates multiple computational methodologies: generative models trained on natural protein sequences and known structural data, physics-based molecular dynamics simulations to predict protein folding and stability, energy landscape calculations to optimize binding affinity and specificity, and iterative refinement loops to design proteins meeting multiple performance criteria (folding, binding, manufacturability, immunogenicity, half-life). The technology is modality-agnostic and target-agnostic, theoretically applicable to therapeutic antibodies, miniproteins, multispecifics, scaffold proteins, and other protein-based modalities. The workflow is designed for speed: from target definition to candidate selection often within weeks, compared to months or years required by traditional screening-based approaches.

Validation came in 2024–2025 with peer-reviewed publications in high-impact journals. In June 2025, DenovAI's results were published in *Molecular Systems Biology*, demonstrating that AI-designed proteins fold as intended and function in living systems—critical validation for drug discovery applications. The paper presented de novo-designed proteins targeting complex bacterial immune mechanisms (Type III secretion systems), achieving functional efficacy and opening new therapeutic possibilities that were previously intractable with traditional methods.

Strategic backing is exceptionally strong for an early-stage biotech startup. AION Labs, DenovAI's parent initiative, counts among its investors and corporate partners: AstraZeneca, Merck, Pfizer, Teva (Israel-based pharma), Amazon Web Services (AWS), the Israel Biotech Fund, Amiti Ventures, and support from the Israel Innovation Authority. This multi-party backing model gives DenovAI direct strategic input from major pharmaceutical companies and ensures early validation of its platform's value proposition. The arrangement also provides access to industry expertise, assay platforms, disease target pipelines, and manufacturing knowledge that would take independent startups years to accumulate.

DenovAI operates in a competitive landscape that includes other AI-driven drug discovery platforms (DeepMind's AlphaFold, Isomorphic Labs, OpenAI Codex, and other protein-design companies). However, DenovAI's differentiation rests on combining physics-based modeling with generative AI in a way that produces functionally validated, manufacturable therapeutic candidates rather than just predicted structures. The company's approach is grounded in biophysical first principles, making it applicable to challenging therapeutic targets where traditional approaches have struggled.

From a commercialization perspective, DenovAI operates a dual-track model: milestone-driven partnerships with pharma companies (where DenovAI designs candidate proteins for specific targets and pharma retains development and commercialization rights) and internal drug discovery programs (where DenovAI discovers, develops, and potentially brings candidates toward clinical development). The partnership model mitigates cash burn and provides early revenue validation; the internal programs provide ownership upside and differentiation in the market. Early partnerships are likely with AION Labs member companies (AstraZeneca, Merck, Pfizer), providing both commercial traction and strategic validation.

Regulatory and development path considerations: protein therapeutics developed de novo still require preclinical and IND-enabling studies (manufacturing, toxicology, immunogenicity) before entering human trials. DenovAI's platform accelerates the design phase but does not obviate the need for classical biotech/pharma development infrastructure. The company will likely rely on partnerships with pharma companies or contract research organizations (CROs) for development through regulatory approval. Time from AI design to IND application could compress to 12-18 months (vs. traditional 2-4 years), but total development timelines to approval remain multi-year. Successful commercialization requires both technical validation (which DenovAI has begun) and sustained partnership relationships with pharma, CROs, and regulatory bodies.