Creative Biolabs offers customers a knowledge-based protein design service for novel molecule studies of protein engineering.
Knowledge-based protein design refers to the design of a protein with computational principles embedded in the statistics of protein sequence and structure databases. This approach reduces the combinatorial complexity of predicting protein sequences that determine protein characteristics. It leads to a better understanding of the physical chemistry of proteins. Knowledge-based potentials derived from secondary and tertiary protein structures have become the key elements of nearly every protein design algorithm. They have provided significant motivation to resolve the protein folding problem. Additionally, knowledge-based protein design can further refine and simplify the potential functions used in protein design algorithms.
Figure 1. Knowledge base of protein microenvironments linked to ligand fragments. For each protein-ligand complex from the PDB, residue atoms interacting with the ligand were identified, and the ligand atoms proximal to them (semi-transparent shaded regions) were also noted (top). Next, the FEATURE microenvironments of the residue atoms are calculated (semi-transparent circles) (center). Ligand atoms were also mapped to their pre-computed fragment lists and linked to their proximal microenvironments to form the knowledge base (bottom). (Grace W. Tang et al., 2014)
Taking knowledge-based potentials for example, they can be used for the modification of existing protein function, the redesign of natural protein folds and the complete design of a non-natural protein fold. What’s more, such potentials can provide information on the global topology of amino acid interactions in natural proteins.
De novo design still remains as a challenging problem, like the creation of novel protein folds, binding interfaces, or enzymatic activities. A more immediate approach is the redesign of existing proteins, to enhance their thermostability, altered binding specificity, improved binding affinity, enhanced enzymatic activity, or alter the substrate specificity. Simple energy minimization of a single protein is often not adequate for the above-mentioned predictions. Multi-objective searches have to be used for overall improvement of protein properties.
Human intervention can be seen in most design methods. For example, compared with a fully automated methodology, the use of hand curation is common for selecting or refining designs. Predictions have been aided by visual inspection regarding intermolecular hydrogen bonds or intermolecular hydrophobic contacts. However, the limitations of such hand-curated designs are also noticed, despite their significance for success: firstly, they have limits on the transferability of methods for use in new systems or by other researchers; secondly, they limit the ability to investigate and improve the understanding of the underlying biophysical interactions. We can perform knowledge-based protein design by taking holistic measures of all the important factors involved in protein structure, function and stability.