Imperagen Closes £5M Seed Round for Enzyme Engineering

Imperagen, a Manchester-based techbio company spun out of the University of Manchester, has closed a GBP 5 million seed funding round to advance its closed-loop enzyme engineering AI platform, bringing total funding to GBP 8.5 million.

PXN Ventures led the round via the GMC Life Sciences Fund and NPIF II – PXN Equity Finance, with continued participation from existing investors IQ Capital and Northern Gritstone. The company said proceeds will be used over the next 18 months to accelerate platform research and development, expand wet-lab capacity, grow its in-house AI team, and build out commercial operations across its target sectors, which include pharmaceutical manufacturing, life sciences, personal care, sustainable fine chemicals, and industrial biotech.

Imperagen's first institutional round, a GBP 3.5 million seed co-led by IQ Capital and Northern Gritstone, closed in August 2022 to support the initial spinout from the University of Manchester.

The company was founded in November 2021 by Dr Andrew Almond, Dr Andrew Currin, and Dr Tim Eyes, all researchers from the university's Manchester Institute of Biotechnology. Coinciding with the close of the current round, Guy Levy-Yurista, PhD, joined as chief executive officer. The company described him as a technology and life sciences executive with two prior exits across the US and Europe.

Imperagen's platform combines quantum-physics simulation, problem-specific AI modelling, and automated laboratory robotics in a single integrated system. In the first stage, quantum mechanical modelling simulates millions of enzyme mutation combinations in silico, generating predicted property datasets without physical experimentation. Those outputs train bespoke AI models calibrated to the specific engineering challenge rather than general-purpose protein language models, which the company said are not optimised to predict catalytic performance for a defined reaction under defined conditions. Automated robotic wet-lab systems then physically test the highest-ranked variants, and the resulting experimental data feed directly back into the AI model, tightening predictions with each successive round.

The approach is grounded in established computational enzymology. Enzyme catalysis involves electron transfer and bond-breaking events that can require quantum mechanical treatment to model accurately, a principle underlying the well-validated quantum mechanics/molecular mechanics framework used across the field. By generating large quantum-physics-derived datasets and training problem-specific models on them, the company said it can navigate the non-linear sequence-function landscape of enzyme catalysis more efficiently than conventional directed evolution or zero-shot AI design methods.

The most publicly detailed application to date involved a collaboration with an unnamed Fortune 500 personal care company, in which Imperagen reported productivity improvements of 677x and 572x for two separate enzymes across five rounds of its closed-loop system. No pharmaceutical manufacturing partner has been publicly named, though the company identified that sector as a primary commercial target.

Enzymes are used in pharmaceutical manufacturing to perform stereoselective and other chemically demanding transformations in the synthesis of active pharmaceutical ingredients, often replacing metal-catalysed steps that require high pressure, toxic reagents, or generate significant waste. Engineering enzymes capable of processing complex, non-natural drug-like scaffolds requires precisely the kind of active-site remodelling that quantum-physics simulation is designed to predict. The company's platform is therefore positioned as an enabling tool for biocatalytic API synthesis rather than a direct drug discovery engine.

This article was generated with AI assistance and reviewed and edited by the AllSci editorial team Explore more at AllSci News: https://allsci.com/news/

Spot something wrong? Report an issue with this article

The AllSci Brief