EvoSelect® is a machine learning enzyme engineering technology. It uses evolutionary information from sequence alignments of homology families to generate novel sequences with improved properties.
Key Differentiators in Enzyme Engineering
EvoSelect® integrates into the "Design Phase" of our IsoZym® platform. The platform supports the full spectrum of enzyme discovery and development. Therefore, each step operates either as a standalone work package, depending on the partner’s stage, or as part of a cohesive workflow supporting end-to-end enzyme production.
By partnering with us, you access a comprehensive suite of services. Consequently, this ensures consistency, efficiency, and faster timelines across all development stages.
Moreover, Isomerase offers consultancy support at every step. This includes risk assessment and technoeconomic analysis.
We provide insights into cost optimization, risk mitigation, and sustainable innovation, ensuring both economic viability and operational excellence.
EvoSelect® - Machine Learning Enzyme Engineering Workflow
EvoSelect® applies unsupervised learning to detect statistical patterns from sequence alignments of homology families. These are drawn from proprietary databases, rich in sequences from thermophilic genomes. Specifically, co-evolutionary analysis captures inter-residue relationships (couplings) within a protein, which are used to build a probabilistic pairwise model (Potts model[2]). By establishing couplings, we incorporate epistasis into mutational fitness prediction. As a result, this approach offers improved accuracy compared to single-site models. Subsequently, a generative probabilistic interpretation is built out of the model, from which we iteratively sample mutations, simulating evolution and generating new sequences with improved fitness.
Example Pipeline Using Machine Learning for Enzyme Engineering
EvoSelect® streamlines enzyme development from target selection to material supply in just 8 weeks. The process begins with the input of the target seed sequence. Next, ML-driven sequence optimization initiates, where EvoSelect® iteratively mutates sequences to align with target criteria. Following codon optimization, 40 enzyme candidates are overexpressed in our IsoChassis® hosts and evaluated for expression and target activity. The top-performing candidate is then selected, and utilizing our proprietary drop-in high cell density fermentation methods, we can produce 1-10 grams of the optimized enzyme. Importantly, our process is designed for royalty-free commercialization, allowing freedom to operate without downstream licensing fees.
Contact us
Reach out to explore how our machine learning enzyme engineering platform can refine and accelerate your enzyme development, bringing your innovations to market faster.
In addition, we have extensive experience working with international partners, with the majority of our clients based in the United States. We welcome inquiries from organizations worldwide to explore potential collaborations.