This page requires Adobe Flash Player to display correctly.
This ticker requires Adobe Flash Player to display correctly.

In Silico Kinase Specificity Prediction (IKSP) Services

For many of the well studied protein kinases, it is possible to deduce their substrate specificities by careful amino acid alignment of the phosphosites of known in vitro substrates. However, for more than two thirds of 536 known human protein kinases, there are too few if any known substrates. This lack of knowledge about the specificities of most protein kinases severely handicaps their characterization. Kinexus has developed an algorithm that provides for prediction of the specificities of nearly 500 human protein kinases and appears to have utility in other diverse species as well.

The peptide substrate interaction with kinases reflects the composition and placement of amino acids within the active sites of these enzymes. Since the “typical“ protein kinases have very similar primary and tertiary structures within their catalytic domains, it is possible to use mutual information to predict kinase-substrate interactions. We deduced consensus sequences of over 150 protein kinases following alignment of their known substrates through examination of about 10,000 kinase-substrate phosphosite pairs. In collaboration with the MITACS Group in the Departments of Mathematics and Computer Science at the University of British Columbia and Simon Fraser University, we have “cracked the kinase code” so that we can predict the specificity of a protein kinase from its primary amino acid sequence in its catalytic domain. This involved the precise alignment of 247 amino acids and gap/insert regions in the catalytic domains of these kinases. Through a combination of the application of mutual information and amino acid interaction scoring between a kinase and its substrate phospho-peptide, we have developed an algorithm that demonstrates a high degree of successful substrate prediction capability. Approximately 80 amino acids in the catalytic domain of kinases appear to contribute to their protein substrate recognition.

The creation of amino acid probability frequency tables for all 20 common amino acids at each position surrounding and including the phosphosites has permitted Kinexus to predicted the amino acid sequences of optimal peptide substrates for each of nearly 500 human protein kinase catalytic domains. Our research has revealed that we can also produce substrate amino acid probability matrices for any “typical” protein kinase in another eukaryotic species for which the catalytic domain amino acid sequence has been determined. Identification of optimal and poor substrates phosphosites is invaluable for localizing or ruling out sites of phosphorylation in substrate proteins of interest or even predicting which protein may be substrates for a particular kinase of unknown function.

It is very convenient and inexpensive for clients to take advantage of the power of our kinase substrate prediction capabilities for their own research programs with our In Silico Kinase Specificity Prediction (IKSP) Services. For our Introductory Price of US$179 per kinase, clients only have to provide us with the name and Uniprot or NCBI accession number of the protein kinase that they are interested in. Kinexus will retrieve the primary amino acid sequence of the kinase, carry out the precise alignment of the catalytic domain amino acids, and perform our kinase specificity prediction analysis. In return, the client will receive a unique kinase specificity amino acid frequency matrix that can be used as desired. Such information can be used to perform BLAST searches to identify novel substrates of interesting kinases, design optimal peptide substrates and pseudo-substrate kinase inhibitors.

With our follow up Custom Peptide Synthesis Services, Kinexus can produce synthetic peptides with predicted amino acids sequences for use as optimized substrates to assay kinases or pseudo-substrates (in which the phosphosite is modified to an alanine or phenyalanine residue) to inhibit selected kinases.

To get everything you need to take advantage our this unique service, please download our Customer Information Package.
Credits