+43 316 269 205 office@innophore.com

TECHNOLOGY

CHANGE YOUR WAY OF SEARCHING NEW OR NOVEL ENZYMES

with 3D point clouds – so called Catalophores

 

We use a bioinformatics method combined with artificial intelligence to mine structural databases of enzymes using three dimensional search templates termed “catalophores” (i.e. carrier of the catalytic function).

New enzymes, identified with this technique do not share a common structure or sequence base with their currently employed counterparts and therefore potentially feature altered protein properties such as thermo- and solvent stability, substrate spectrum, selectivity and specificity.

3D point clouds - Catalophores - for Enzyme Discovery
3D point clouds - Catalophores - for Enzyme Discovery with different properties
Supercomputing enables Drug & Enzyme Discovery

DATA MINING & DATABASE

 

We mine and homogenize publicly available data sources, such as PDB, PISA, UniProt, ProFam etc.

Our proprietary database covers:

  • 98% of the experimental structures from RCSB PD
  • 100% of the Biological assemblies from ePDB PISA
  • High-throughput distributed automatic comparative model building
  • Tailored in-silico mutant libraries
  • Synthetic proteins including non-canonical amino acids

     

    HOW WE SEE THE WORLD

     

    Our 3D point-clouds cover 19 physico-chemical properties (electrostatics, hydrophobicity, accessibility, potential energies, hydrogen binding potential, elasticity …) that are matched to find similarities to binding sites without the need of sequential or structural similarity.

    1 cavity  – several different Catalophores

    Matching algorithm at work

    MATCHING

    A match is more sustainable when you look behind the surface…

     

    Our tailored ICP (Iterative-Closest-Point based) multi-dimension point-cloud matching algorithms calculate the similarity of cavities including the physical-chemical properties as artificial dimensions enriching the cartesian shape similarity.

    EMPLOYING BIG-DATA AND MACHINE LEARNING FOR GROUNDBREAKING DISCOVERIES IN ENZYME RESEARCH

     

    We feed the matching information into our AI deep learning models to improve accuracy and speed. The more we match, the better we know.

    The non-cartesian view of the cavity world allows us to recognize similarities in deep dimensional space a human observer would hardly find

    One match is quick, it usually takes 3 seconds. We match millions a day. That’s why we run supercomputers.

    Neural networks, Machine Learning and Artificial Intelligence for Drug & Enzyme Discovery

    1 hidden layer  – 100 hidden layer neurons

    ADDITIONAL TOOLS INTEGRATED IN THE PLATFORM

    Homology modeling

    Distributed automatic homology model building

    Mutant generation

    Distributed automatic mutant generation - creating mutant libraries

    Docking

    Distributed automatic docking analysis - docking experiments to predict protein - ligand interactions

    In-silico protein modification

    In-silico global or local protein modification - predicted structures of synthetic proteins

    Molecular dynamics

    Distributed automatic molecular dynamics - predicted protein - ligand interactions

    APPLICATIONS

    Industrial biocatalysis and biotechnology

    • Carry out novel enzymatic reactions

    … e.g. altered substrate scope, selectivity

    • Overcome insufficient properties of existing biocatalysts

    …. e.g. increased pH or thermal stability

    • Circumvent protection by IP rights or patents

    … gain or regain your FTO

    • Get a new starting point for your traditional engineering pipeline

    …if you reached the end and your biocatalyst still does not do the job

    Drug discovery and repurposing in pharma and medicine

    A new tool for the in-silico prediction of unexpected side-effects before clinical trials.
    Screening for repurposing opportunities of existing, novel or abandoned drugs.

    SUPPORT HUMAN RESOURCES

     

    The platform can improve your performance within wet-chemical lab efforts by reducing In vitro screening for enzymes to a minimum. Our in silico approach makes it possible to reduce time to market significantly.

    Shorter time-to-market in the pharma pipeline - early stage R&D

    Check out our services