Google DeepMind has significantly advanced science by employing a deep learning model to forecast complex protein structures, which are vital for various cellular functions. Despite the increasing output of potential drug candidates from AI models, the challenge lies in the practical characterisation of these candidates for testing and broader production.
To address this issue, a startup called 10x Science, founded in December 2025, has raised $4.8 million in seed funding, spearheaded by Initialized Capital with support from Y Combinator and other investors. The team’s founders—biochemists David Roberts and Andrew Reiter, alongside AI expert Vishnu Tejas—are passionate about improving the drug development process.
Roberts highlighted that while drug prediction tools are available, all potential candidates must undergo a meticulous characterisation process to ensure proper measurement. This step is crucial for the development of biologic drugs that specifically target diseases, such as Merck’s Keytruda, which assists the immune system in combating cancer.
The founders, who collaborated in Dr. Carolyn Bertozzi’s Stanford lab, became aware of the difficulties in understanding molecular interactions when researching cancer and immune cell dynamics. They realised the need for an advanced approach to analysing molecules using a sophisticated technique known as mass spectrometry. While mass spectrometry is the most accurate method for determining molecular structures, it produces complex data that requires extensive expertise for analysis, which can be time-consuming.
10x Science’s platform integrates deterministic algorithms based on chemistry and biology with AI systems capable of interpreting mass spectrometry data. The founders focused on training their models effectively and ensuring their findings are traceable to comply with regulatory standards.
Matthew Crawford, a scientist at Rilas Technologies, reported a positive experience with 10x Science’s tool, noting its ability to autonomously find required data and adapt to different molecular evaluations. He commended the efficiency of the model, which surprised him with its capacity to provide insightful explanations of its outcomes.
10x Science aims to collaborate with major pharmaceutical firms and academic institutions, intending to leverage the recently acquired funds to expand its team and refine the platform. If successful in protein characterisation, the company envisions merging protein structures with other biological insights to revolutionise the field of molecular intelligence.
Investors view 10x Science as a strategic entry into the biotech sector, as it provides a service independent of a specific drug’s regulatory approval. The company’s success could establish it as a key player in drug development processes, especially if it continues to innovate within the mass spectrometry domain. Crawford suggested that the platform could assist researchers by delivering prompt and simple solutions to complex questions, thereby streamlining their research and advancing drug discovery.
Fanpage: TechArena.au
Watch more about AI – Artificial Intelligence
