In the process of identification and design of a biocatalyst, experience and skills are often required along with tedious cycles of design-trial-optimization. Computational modeling technique is an alternative way from traditional experimental methods to study bioactive molecules for target biocatalysts. In industry, it is critical to predict the specificity of the target protein and the interactions for a novel compound from its chemical structure. In pharmaceutical applications, computational modeling approaches, such as two-dimensional and three-dimensional quantitative structure-activity relationship (QSAR), pharmacophore mapping and machine learning methods have resulted in statistically valid predictions.

Several principles of computational tools are used to analysis a novel compound that can fit active sites: i) Modeling based on the 2D structure of known active compounds to search for similar compounds in large chemical databases. ii) Modeling based on structures of ligand-target interaction patterns in the crystal complex. Exclusion volumes can be added on residues lining the binding site, thereby mimicking the steric constraints of the pocket. iii) Modeling based on the 3D space defined by the shape of a known substrate. Therefore, other active chemicals can be modeled to fit the same space. iv) 3D models where diverse compounds from a chemical database docked into the binding pocket. Those in silico methods have been employed for the investigation of various proteins. Successful applications include drug development, enzyme characterization and substrate identification, toxicology and others.

Creative Enzymes’ service of computational modeling helps customer find the most suitable substrates for their target. With variety of in silico tools, the professional services provide much insight into the desired biocatalysts at a relatively low cost:

• Chemical database search.
• Three-dimensional structure prediction and determination.
• Virtual screening of substrates.
• Rational design of the biocatalyst.

Figure 1. Principles of commonly applied computational tools exemplified on the crystal structure of 17β-HSD1
(Journal of Steroid Biochemistry and Molecular Biology, 2017)