Research

AI-Powered Biotech Research

Our four research pillars cover the full spectrum of AI-enabled biotechnology — from finding new drug targets to automating the physical experiments that validate them.

Four Pillars of AI Biotech

Each area leverages AI as its primary research accelerant — working in concert to compress discovery timelines from decades to years, and years to months.

Drug Discovery

AI Drug Discovery

Target identification, generative molecular design, and ADMET property prediction are transforming early-stage drug development. Our AI platforms screen billions of virtual compounds, identify previously unknown disease targets, and optimize candidate molecules for safety and efficacy — without setting foot in a wet lab until the most promising candidates are ready.

Sub-areas: Target Identification · Generative Chemistry · ADMET Modeling · Combination Therapy Design · Repurposing Discovery

Explore Drug Discovery →
Approach
Generative AI + Biology

Our generative models learn the chemical grammar of known drugs and apply it to unexplored molecular space — producing candidate molecules optimized for target binding, selectivity, and metabolic stability simultaneously.

Impact
From Years to Months

Traditional lead identification takes 2–4 years. Our AI pipelines compress that to 3–6 months by running in silico experiments that would take a team of chemists a decade to attempt manually.

Protein Science

Protein Science & Folding

Proteins are biology's molecular machines — and their 3D shape determines their function. Our structure prediction and protein design programs use deep learning to model how amino acid sequences fold into functional forms, enabling us to engineer enzymes, design antibodies, and understand disease mechanisms at atomic resolution.

Sub-areas: Structure Prediction · De Novo Protein Design · Enzyme Engineering · Antibody Modeling · Protein-Protein Interaction Mapping

Explore Protein Science →
Approach
Deep Learning Structure Models

We deploy and extend state-of-the-art protein language models trained on hundreds of millions of sequences to predict structure, stability, and function — and to design proteins that evolution never explored.

Impact
1,000× Faster Than X-Ray Methods

Structural biology has historically required months of crystal growth and synchrotron analysis per protein. AI structure prediction delivers comparable accuracy in seconds — unlocking structural insights across entire proteomes.

Genomics

Genomics & Precision Medicine

The human genome contains 3 billion base pairs and interacts with environment, microbiome, and lifestyle in ways that produce unique disease risk profiles for every individual. AI is the only tool capable of reading and interpreting genomic data at the scale and speed needed to make precision medicine real for the general population.

Sub-areas: Variant Discovery · Polygenic Risk Scoring · Gene Expression Modeling · Rare Disease Genomics · Treatment Matching

Explore Genomics →
Approach
Population-Scale AI Analysis

Our genomics AI models process multi-omics data — whole genome sequences, RNA expression profiles, epigenetic marks — across thousands of samples simultaneously to identify disease-relevant signals hidden in genetic noise.

Impact
Personalized Treatment at Scale

By linking genomic variants to treatment response data, our models guide clinicians toward therapies matched to a patient's specific biology — reducing trial-and-error prescribing and improving outcomes.

Lab Automation

Lab Automation & Biomanufacturing

Even the best AI drug candidate must eventually be synthesized and tested in the physical world. Our lab automation programs deploy AI-controlled robotic systems that run experiments, optimize bioprocess parameters, and scale biological manufacturing — closing the loop between computational discovery and physical validation.

Sub-areas: Robotic Liquid Handling · AI-Optimized Bioprocess Control · Smart Bioreactor Systems · Self-Driving Lab Platforms · Quality Control AI

Explore Lab Automation →
Approach
Self-Driving Laboratories

Our self-driving lab platforms use reinforcement learning to design, execute, and analyze experiments autonomously — iterating toward optimal results without human intervention between experimental cycles.

Impact
24/7 Experimental Throughput

Robotic labs don't sleep, don't make pipetting errors, and don't need weekends. AI-controlled systems run experiments continuously — compressing weeks of manual work into days while maintaining reproducibility standards human labs cannot match.

Active Research Programs

Explore the specific funded programs operating across our four research areas.

View All Programs
An initiative of the Odegard Foundation · Founded 2017 by Philip Odegard