Artificial intelligence is no longer futuristic talk in terms of using the technology’s capabilities to aid in the discovery and development of new possible drugs and therapies. Researchers at MIT used machine learning to discover an antibiotic compound in 2020 that can potentially kill E. coli other disease-resistant bacteria and in a fun twist, named it halicin (after HAL9000). In 2022, the number of AI-discovered small-molecules grew rapidly and today there are approximately 250 companies focused on AI-driven drug discovery. The FDA also recently published a Digital Health and Artificial Intelligence Glossary for commonly used terms in this space. While there’s currently no approved drug that has been fully AI-discovered, the field is still emerging as is the technology’s potential of automating time-consuming processes and speeding up timelines to finding breakthroughs.

In this edition, we highlight several Biocom California member companies that are making inroads in using AI for drug discovery. If you want to engage further in the conversation, join our Converge group on LinkedIn.

Building Breakthroughs

A top name in the field is insitro, which is focused on using machine learning and generative AI to advance drug discovery and development for a variety of indications. Founded in 2018 by Daphne Koller, a prominent computer scientist who co-founded Coursera while she was a professor at Stanford, the Bay Area-based company has been called “biotech AI’s quiet unicorn” and made the Forbes AI 50 list. The company’s platform uses multi-modal phenotypic cellular data generated in its laboratories with clinical data from human cohorts, and employs machine learning to interpret this complex data set and develops phenotypic models of disease states for further analysis. It’s currently identifying possible therapies for unmet needs in epilepsy and seizure conditions, ALS, oncology (solid tumors) and metabolic pathologies such as Metabolic Dysfunction–Associated Liver Disease (MASLD). The company notes it has partnerships with Bristol Myers Squibb, UK Biobank and Genomics England, this month they entered into three strategic agreements with Eli Lilly to advance new therapies for metabolic diseases. Earlier this spring, they announced Emily Fox, Ph.D., a professor in the Department of Statistics and Department of Computer Science at Stanford University, as their new senior vice president of AI/machine learning.

Supporting Research Institutions, Nonprofits and More

Based in Japan, Axcelead Drug Discovery Partners was formed as a spin-off of Takeda Pharmaceutical and specializes in supporting large biotech companies, startups and academic institutions with drug discovery research. The company boasts an extensive research base and library of 1.5 million compounds, and uses AI and High Throughput Screening (HTS) to discover and develop novel medicines to help address unmet medical needs in a variety of therapeutic areas, such as oncology, immunology, neuroscience, and cardiovascular and metabolic diseases. This fall, the company made two big announcements regarding new partnerships: one with Eli Lilly to research multiple drug discovery programs; and another with Acadia Pharmaceuticals on supporting its research in finding new medicines for CNS diseases and other indications. The company also notes it’s the first drug discovery solutions provider in Japan.

Beyond the Genome

Sapient Bioanalytics specializes in using next-generation mass spectrometry and AI to do large-scale profiling of dynamic biomarkers—proteins, metabolites and lipids—to help biotech and biopharma companies advance their drug discovery and development programs. The San Diego company’s multi-omics biomarker discovery approach goes beyond the genome, where much of biomarker discovery has occurred to date, to explore the broader molecular landscape of human biology and better understand dynamic disease processes over time. The company notes that the biomarkers it measures read out that more than 80% of disease risk stems from non-genetic factors and exposures, providing a more complete view of influences contributing to disease and drug response. Sapient says this data can help accelerate development of precision medicines, for example by combining discovery proteomics and AI-based analysis to support target discovery and validation. The Sapient team also recently contributed to a groundbreaking study that was published in Nature Metabolism journal on assaying metabolic activities in intact human liver tissue ex vivo by using global 13C isotope tracing and non-targeted mass spectrometry. Sapient performed the mass spectrometry analysis and Dr. Mo Jain, founder of Sapient, was an author on the paper.

Targeting HER2-Driven Cancers

Iambic Therapeutics was founded by two chemists and its AI platform is informed by physics principles to discover and design possible drug candidates. The company says by using high-throughput chemistry and biology, it’s able to create and test novel molecular structures against fresh biological data on a weekly basis, as well generate biochemical, metabolic and cellular data. The San Diego-based company’s pipeline includes possible a new therapy for HER2 cancers and a first-in-class small molecule inhibitor targeting CDK2 and CDK4, cyclin-dependent kinases that are often dysregulated in a variety of cancers. 2024 has been a busy year for Iambic: Earlier this year, they dosed the first patient in a Phase 1 clinical study of IAM1363 for the treatment of HER2-driven cancers; they recently announced a partnership with Lundbeck to research new therapies for neurological conditions; and in September, the company made Endpoints News’ list of Biotech’s Most Promising Startups. Co-founder Thomas Miller will also be a guest speaker at our upcoming Global Partnering & Investor Conference in February 2025.

Faster Formulations

Persist Ai is a newer company founded in 2022 that came out of Y Combinator, and was noted for using AI and robotics to reduce the formulation time for injectable drugs that are used to treat chronic illnesses from 5 years all the way down to 2 years. The Sacramento-based company says their mission in building a next-generation formulation platform to better aid drug discovery is a personal one. Founder Karthik Raman, Ph.D., has shared that his mother was diagnosed with brain cancer when he was young, and that a drug that was newly-released at the time helped her live well beyond the initial prognosis of six months—she lived for 15 more years. The company says scientists can use their Cloud Lab to build formulations, that their technology allows one scientist to do the work of ten and can do thousands of experiments a month. In August, Persist announced a partnership with Nivagen to bring a novel long-acting injectable and drug product manufacturing platform to market.

Unearthing Moleculer Gems

Genesis Therapeutics was spun out of Stanford University by founder and CEO Evan Feinberg back in 2019, and the company has received media attention for its proprietary advanced molecular platform, GEMS, which uses generative AI to discover molecules and possible drug candidates that could treat challenging or previously undruggable targets. The company’s current pipeline includes optimizing small molecules for biological targets in breast and colorectal cancers, as well as a variety of autoimmune conditions. Genesis recently agreed to a $35 million deal with Gilead Sciences for AI-based drug discovery work on three undisclosed targets, and it has also partnered in previous years with Eli Lilly and Genentech.

Searching for Antibodies

BigHat Biosciences is focused on discovering antibodies for next-generation therapies, and uses a combination of machine learning and synthetic biology to both hunt for them and engineer them. Leading the search is their proprietary Milliner platform, which the company says integrates a wet lab for high-speed characterization along with machine learning technologies to seek better antibodies that could possibly be used in therapies to treat solid cancers, inflammatory diseases and immune conditions. An article about the company in Genetic Engineering & Biotechnology News notes that without the use of AI, searching for novel antibodies would be “a slog.” Peyton Greenside, BigHat’s founder and CSO, notes their technology makes searching for and engineering antibodies happen at a much faster pace than could be done manually, which had been the traditional process. Instead, hundreds of recombinant antibodies can be produced and analyzed in a single, weekly workcell. The Bay Area-based company was recently named to Fierce Biotech’s Fierce 15 and has inked several notable partnerships, including a collaboration with Janssen Biotech on accelerating protein design and guiding the selection of antibodies for a variety of neuroscience targets.

Peyton Greenside is a previous Catalyst Awards winner and was a guest on Season 1 of our LifeLines podcast—check out her episode to delve more into BigHat’s innovative technology.

Springing Into Action

There’s custom tools for just about every profession, and Spring Science developed AI software just for scientists and researchers, along with a high-content image analysis suite, that can aid with identifying and sorting cell populations, automate quality control, pinpoint novel phenotypes and work through interpretating complex imaging data, to start. The Bay Area-based company’s technology was used in vaccine development through a grant with the Gates Foundation, and it also offers a suite of AI and wetlab services and in-house experts experienced in drug discovery. The company regularly shares educational articles on its website, including a recent piece on the rise of AI virtual cells .