10x Science: AI Redefines Drug Discovery
Le brief IA que les pros lisent chaque soir
Les 7 actus IA du jour, décryptées en 5 min. Gratuit.
Inclus dès l'inscription : notre sélection des meilleurs guides & comparatifs IA.
Choisis ton rythme
Gratuit · Pas de spam · Désabonnement en 1 clic
The Impact of AI on Drug Discovery
Artificial intelligence has already made its mark in the scientific field, particularly through Google DeepMind's initiative, which used a deep learning model to predict the complex structures of proteins. These molecules play a crucial role in almost all cellular processes. However, as these AI models continue to generate an increasing number of candidates for new treatments, a major challenge arises: the need to characterize these candidates for testing and large-scale production.
10x Science: An Innovative Solution
This is where 10x Science comes in, a startup founded in December 2025. It recently announced that it raised $4.8 million in a funding round led by Initialized Capital, with support from Y Combinator, Civilization Ventures, and Founder Factor. The founders, David Roberts and Andrew Reiter, experienced biochemists, along with Vishnu Tejas, a serial entrepreneur specializing in computing and AI models, have set out to solve this problem.
David Roberts explained to TechCrunch that while predictive tools are highly effective, each drug candidate must undergo a rigorous characterization process. This involves measuring and analyzing each molecule, a process that can be lengthy and complex.
The Importance of Protein Structure
Understanding protein structures is essential for researchers developing biological drugs. These drugs, produced in living cells, are designed to target specific diseases with high precision. For example, Keytruda, a drug sold by Merck, helps the immune system identify and attack cancer cells by targeting specific cells.
Expertise from Cutting-Edge Research
The founders of 10x Science collaborated in the lab of Dr. Carolyn Bertozzi, a Nobel Prize laureate, at Stanford. Their research focused on the interactions between cancer cells and the immune system. Frustrated by the limitations of traditional methods for understanding these interactions at a molecular level, they decided to create a more efficient solution.
Mass Spectrometry: A Key Technique
Mass spectrometry is the most accurate method for evaluating molecules. It involves measuring the atomic structure of molecules in an electric field. Although this technique generates highly detailed data, interpreting it requires considerable expertise and can be time-consuming.
A Platform Combining AI and Chemistry
The 10x Science platform integrates deterministic algorithms from chemistry and biology with AI agents capable of interpreting mass spectrometry data. The team has dedicated significant effort to training these models so they can analyze data in a traceable manner, a crucial aspect for ensuring regulatory compliance.
Testimonial from a Satisfied User
Matthew Crawford, a scientist at Rilas Technologies, uses the 10x Science platform for chemical analyses. He has found that the platform significantly accelerates his work. He was impressed by the model's ability to explain its conclusions and adapt to evaluating different types of molecules.
Crawford emphasized that, unlike other AI tools he has tested, this one makes reasonable assumptions thanks to the expertise of the creators of 10x Science. For instance, he was able to submit a protein to the system, which then searched online databases for the sequence of that protein without requiring additional programming.
Collaboration with the Pharmaceutical Industry
The leaders of 10x Science are also collaborating with several major pharmaceutical companies and academic researchers. The initial funding will be used to recruit more engineers and refine the model while offering it to new clients. The goal is to expand the offering to provide a deeper understanding of biology by combining protein structure with other cellular data.
A New Definition of Molecular Intelligence
David Roberts stated that the ambition behind their project is to redefine molecular intelligence. For investors, 10x Science represents an opportunity in the biotechnology field that is not tied to the success of a specific drug. If the company succeeds, it could become an essential tool for drug development, regardless of the commercial success of the final products.
Zoe Perret, a partner at Initialized, highlighted that the SaaS platform of 10x Science is a service that pharmaceutical companies will need to use regularly to evaluate potential candidates. The founders' experience is a significant asset in standing out from the competition.
A Tool to Unlock New Techniques
According to Crawford, the platform could help researchers access techniques they cannot deploy due to a lack of time or resources. This would simplify the process of researching and developing new drugs, providing quick and accurate answers from mass spectrometry.
Brief IA — L'actualité IA en français
L'essentiel de l'actualité de l'intelligence artificielle, décrypté et expliqué chaque jour.