NextGen Enzyme Engineering AI-volution

“I wish we employed Aminoverse’s AI platform 4 years ago, who knows where our enzyme performance would be now” – we could not have said it better than this customer.

If enzyme performance is one of your strategic pillars, it is time to change your approach. Aminoverse’s wet lab data generation and machine learning algorithms combine the best of both worlds with a clear focus:

Design the best possible enzyme in the shortest time at minimal budget.

Traditional approaches like random mutagenesis and in silico design usually target less than 10% the enzyme’s sequence space resulting in an equally limited knowledge base and thus low success chance. Future developments of the same or similar enzymes always start at square one.

Aminoverse’s AI platform brings your enzyme development to the next level. By pairing state-of-the-art wet lab workflows with latest advancements in machine learning, a fundamental and holistic knowledge base will be generated for your target enzyme. This does not only allow to identify the very best mutations for mutagenesis, but also to unveil valuable principles about the target enzyme class and its behavior under reaction conditions.

Such a data-driven approach significantly reduces the risk of enzyme engineering projects and at the same time provides a head-start for future developments targeting the same enzyme or structurally and evolutionarily related enzymes.

Discover up to 100% of all point mutations that improve your enzyme’s performance – at a fraction of the cost. Our proprietary AI technology ‘EnzyMAP’ is rooted in real wet lab data and allows to predict the effect of all point mutations across the entire enzyme sequence, requiring 80 % less screening capacity and budget.

If you want to bring your enzyme development to the next level and obtain fundamental knowledge about your enzyme spanning the entire amino acid sequence instead of just 5-10%, we have you covered. This is not only applicable to wild type enzymes but also of particular use for engineered enzymes where no additional improvements could be achieved.

Per each screened property, an enzyme fitness landscape is predicted and visualized as heatmap. Simultaneous screening for several enzyme properties allow the identifcation of mutations that improve multiple parameters at the same time or which mutations have detrimental effects on enzyme performance.

The holy grail of enzyme development: A safe way to recombine top mutations to create superior enzyme performance. Our proprietary AI technology “EnzyREC” allows to predict the probability of recombinatorial enzyme mutants to fold correctly and be fully functional while combining all the benefits in one superior enzyme.

With more than a centillion of possible amino acid combinations, any enzyme’s sequence is an incredibly complex puzzle where all pieces have to fall in place correctly to yield the desired performance.

Our recombination approach with EnzyREC AI is unique, as we do not strive to predict recombinant variants with improved performance. There aren’t sufficient data sets available to properly train an algorithm for this. Instead, we focus on the prediction of stable and correct folding of recombinant enzymes. The performance improvement comes automatically through the selection of the best performance improving mutations identified by EnzyMAP AI.