Generative AI is rapidly accelerating industrial innovation, with a January 2024 McKinsey report estimating that it could bring up to $110 billion in annual economic value to the life sciences industry. This technology has the potential to transform drug discovery processes, helping pharmaceutical companies shorten the time it takes to unlock the value of new drugs.
Since 2017, many large pharmaceutical companies, drug development firms, AI startups, and academic institutions have launched AI research and development projects. Initially, these AI-driven drug discovery initiatives focused on predicting bioactive compounds and generating novel small-molecule drugs (lead compounds). In recent years, they have rapidly expanded to cover various fields of biomedical and drug discovery research.
Currently, at least 18 AI-designed drug candidates have entered clinical trials. Below are five unicorn and near-unicorn startups worth watching:
Xaira Therapeutics | 2024 New Unicorn
Xaira boasts a leading team of protein design experts and has rapidly emerged as an AI-driven drug development unicorn this year. Its first round of fundraising attracted $1 billion in investments, marking the second-largest biotech startup fundraising event globally in 2024.
Founded by neuroscientist Marc Tessier-Lavigne in 2024—who previously served as president of Stanford University and Rockefeller University—the company’s research is led by protein design master David Baker. Baker was named to "TIME 100 Health" in 2023 by TIME magazine. Xaira’s team has the capability to develop machine learning models, generate large-scale data, and innovate new therapies.
Their most powerful AI model, RFdiffusion, is an innovative system that combines structure prediction with generative diffusion modeling. It uses deep learning to generate new proteins that are unlike any known proteins found in nature. Traditional design methods require testing thousands of molecules to find one that meets performance expectations, while RFdiffusion requires testing far fewer. The research results were published in the prestigious journal Nature in 2023.
Generate: Biomedicines | 2021 Unicorn (Valuation: $1.4 Billion)
Generate has raised $693 million in funding to date. The company was incubated by the well-known biotech venture capital firm Flagship Pioneering, which merged two projects, FL56 and FL57, in 2019 to form this company.
Generate’s technology platform combines expertise in protein science, structural biology, and machine learning algorithms. This platform can generate effective antibodies via computational processes, significantly reducing the time required for traditional protein optimization to just two years. The company utilizes large-scale data and machine learning to rapidly generate antibodies, enzymes, peptides, cell, and gene therapies for various diseases.
Through in-house development and partnerships with pharmaceutical companies, Generate is advancing its drug candidates. Two of its promising drugs have entered clinical trials, and 19 other potential candidates are in development.
Genesis Therapeutic | Near-Unicorn (Valuation: $960 Million to $1.3 Billion)
Genesis Therapeutics is a leading biotech company founded in 2019 by former senior scientists at Google X. Its core technology integrates machine learning and artificial intelligence to accelerate the drug discovery and development process.
Genesis specializes in using advanced computational methods to predict molecular structures and the mechanisms of compound action, thus improving the efficiency of drug research and development. The company has successfully discovered a new antibiotic using machine learning, and their findings were published in Science.
Genesis’s AI platform, GEMS, combines deep neural networks for 3D structural sensing with molecular simulation techniques and generative engines. This platform can create new therapies for traditionally challenging drug targets. In May last year, Genesis partnered with pharmaceutical giant Eli Lilly to develop up to five new therapies, leveraging Genesis’s AI drug platform. The potential benefits of this collaboration could reach $670 million.
Exscientia | 2021 Unicorn (2021 IPO)
Exscientia is a UK-based startup that uses AI to optimize drug design processes. Founded in 2012, the company has raised over $833 million from its inception through its IPO. One of its co-founders, renowned computational chemist Andrew Hopkins, leads a research team focused on developing AI systems that autonomously learn and rapidly predict and design new, therapeutically valuable drugs.
Exscientia made headlines by designing a drug molecule that entered human clinical trials in less than a year—a global first. Currently, Exscientia has two promising drugs in Phase 1/2 clinical trials, and its AI-designed anti-cancer drug, recently published in Nature, shows great potential.
Recursion Pharmaceuticals | 2020 Unicorn (2021 IPO)
Founded in 2013, Recursion focuses on decoding biology to industrialize the drug discovery process. Over the past decade, the company has built the world’s largest proprietary biological and chemical database, leveraging big data and deep learning to identify potential new drugs and therapies. Recursion’s platform has amassed over 23,000 terabytes of biological and chemical datasets, and they are using NVIDIA’s cloud platform to accelerate the development of the world’s largest AI model for biomolecular generation.
This year, Recursion launched LOWE, a generative AI application that can execute complex drug discovery tasks using natural language interfaces. It simulates the work of experts and scientists, interacting with its massive biological and chemical datasets to design and execute hypothetical test experiments, automating and accelerating drug discovery processes.
Resource (mandarin): 【生成式AI醫級戰場1】全球10獨角獸創新局 快速生成蛋白質新藥
