You know the reaction. Where is the enzyme that runs it?
Innophore’s Catalophore™ point-cloud technology searches all of protein space by the shape and chemistry of an active site, not by sequence, to find and engineer enzymes that run your reaction under your conditions. The method the press once called the “Google for enzymes.” Patented, GPU-accelerated, and published since 2014.
A reaction is easy to name. The enzyme that runs it is hard to find.
Sequence search finds close homologs of enzymes you already know. It misses the enzyme from an unrelated family that happens to do exactly what you need, and it says nothing about whether a candidate survives your reactor.
Name the reaction
You know the substrate, the product and the conditions. The chemistry is defined. The question is which protein runs it best.
Find it by function
The best biocatalyst may sit in a protein family you would never query by sequence. Convergent active sites are invisible to a sequence-only BLAST.
Survive the reactor
An enzyme that works at 25°C and pH 7 is a different molecule from one that must run at 40°C, pH 5 and 200 mM substrate.
Two enzymes from unrelated families can share an active site shaped alike. Sequence search never sees it. Cavity similarity does.
The convergent evolution principleSearch protein space by what an enzyme does, not what it looks like.
Catalophore™ describes active sites as 3D point clouds of their physico-chemical property fields, then searches the structural proteome, your own sequences and public metagenomes for functionally similar cavities, independent of sequence, fold or family. The reaction goes in. A ranked list of enzymes with a structural rationale comes out.
Industrial biocatalysis fails for measurable reasons. AI structure prediction is useful, but it still gets the chemistry wrong often enough to matter, with stereochemistry errors in 4.4% of cases by one benchmark. At proteome scale, that is thousands of subtly wrong active sites.
Every structure that informs a recommendation is energy minimized, refined and reviewed before it reaches you. If the model is wrong, the enzyme list is wrong.
From a target reaction to an enzyme your process can use.
One continuous path. Each stage hands the next exactly what it needs.
Reaction & constraints
Substrate, product, and the process limits: pH, temperature, solvent and scale. Plus any in-house sequences.
Innophore
Point-cloud active-site matching across PDB, public metagenomes and your sequences, then structure-guided engineering for stability, selectivity and activity.
Wet-lab hand-off
Ranked candidates with structural rationale, mutation lists and expression-ready files. Optional production and testing via partner CMOs.
Proven across enzyme classes, published in peer review.
Cavity-based enzyme discovery is not a concept. The same method has found new biocatalysts in peer-reviewed work since the founding paper.
Promiscuous activity by active-site mining
The proof-of-principle: mining structural databases by active-site constellation to find promiscuous ene-reductase activity, the method Innophore was founded on.
Fatty acid photodecarboxylases
Cavity-based discovery of new fatty acid photodecarboxylases, showing the search generalizes to enzyme classes far from where it started.
Imine reductases, data-driven
Data-driven construction of imine reductase libraries with machine learning and bioinformatic modeling, for reductive amination toward chiral amines.
Renewable feedstock cascades
Biocatalytic cascades that turn renewable feedstocks into high-value compounds, from cofactor-independent terpene hydration to enzymatic synthesis from eugenol.
Function from raw sequence
SeqScan of every open reading frame in six reading frames, full structural modeling, and cavity-based matching against the catalytic motif you actually need.
The same engine as drug discovery
The cavity search that flags drug off-targets across the human proteome is the one that finds new biocatalysts. One method, validated across all of protein space.
Patented method, ten years of biocatalysis publications.
The cavity point-cloud method that drives every project is a granted patent, accelerated on production hardware, and has discovered new biocatalysts in peer-reviewed work since 2014.
Protected core method
The point-cloud cavity method is a granted patent (US 2015/0302142 A1, also US 10,825,547 B2) and protected across 15 jurisdictions.
100x on production GPUs
Cavity matching throughput accelerated by a factor exceeding 100 on enterprise GPU systems, in partnership with NVIDIA on BioNeMo.
The “Google for enzymes”
The founding Catalophore™ method (Nature Communications, 2014) won the international CPhI Pharma Award and the OEGMBT research award the same year.
Cavity-based discovery
Peer-reviewed proof that cavity matching finds new biocatalysts, from the 2014 founding paper to fatty acid photodecarboxylases in 2024.
The same cavity search that discovers new biocatalysts for fragrance or food is the one that finds new fatty acid photodecarboxylases in academic work. One method, validated across protein space.
Cavity-based biocatalyst discovery, publishedPlugs into the R&D you already run.
Run it as a service, or take the platform in-house. Standard interfaces, defined inputs and outputs, and your IP on every enzyme and variant delivered.
- API
Documented REST API
Programmatic structure submission and retrieval of ranked cavity matches.
- I/O
Lab-ready outputs
Ranked candidates with PDB and PyMOL files, mutation lists and a written report.
- IP
You keep the IP
Intellectual property on identified, built or modified enzymes belongs to the customer.
# find enzymes for a target reaction POST /v1/cavity/search { "query": "<active-site>", "scope": "proteome+metagenome", "filters": "pH, temperature" } # → ranked enzyme candidates { "hits": [ { "enzyme": "…", "similarity": 0.89, "evidence": "cavity" } ] }
Bring us a reaction.
Whether you run discovery programs or build the platforms behind them, there is a way in.
Request a pilot
Bring a reaction, a substrate or a genome. We run a cavity search and walk you through the candidates and the structural evidence.
Start a pilot →Explore the platform
Take cavity-based enzyme discovery in-house with Catalophore Pro, or connect it to your pipeline via a standard API.
Talk integration →