LabGenius

Since the first protein-based drug was approved almost 40 years ago, proteins have been used to treat large debilitating conditions like cancer and autoimmune disease. Today they represent 30% of the total pharmaceutical market, with the top 10 generating over $5B in annual revenue each. Yet the way we discover these treatments has not changed in decades — it is still manual, based on human-led hypothesis, trial-and-error approaches, inconsistent and incomplete data. With low hanging fruits picked many years ago, R&D productivity has massively declined: today, getting just one protein to market takes more than a decade and costs over $2B dollars. It’s a classic example of an industry reaching a bottleneck that can only be solved through digitization and technology.

Abstracting from the specifics of biology, think of discovering proteins like a (very) complex search and maths problem: it requires solving a multi-parameter equation with a huge number of possible combinations (often greater than the number of atoms in the universe) but without having a map of the search space, or being able to test solutions in parallel and at pace. Solving such complexity without technology is virtually impossible.

James Field experienced first hand these limitations during his PhD and wanted to rethink the R&D stack from ground up. He founded LabGenius to build a systematic, machine learning-driven platform that merges the worlds of atoms and bits to fundamentally redesign the process of discovering protein therapeutics.

LabGenius’ platform combines robotic automation, synthetic biology, and machine learning in a closed loop, which enables the team to systematically search through billions of potential protein designs and rapidly test those that are predicted to be high performing. The platform generates a flywheel effect: the more the engine tests, the more proprietary data is accumulated, the more accurate the ML models get, the more complete the biological “maps” become. This industry paradigm shift has the potential to make protein R&D significantly faster, more efficient, and able to surface novel drugs that would not have been found via the “old way”.