Protein kinases are important regulatory enzymes in signal transduction and in cell regulation. Understanding inhibition mechanisms of kinases is important for the further development of new therapies for cancer and inflammatory diseases. I have developed a statistical approach based on the Mantel test to find the relationship between the shapes of ATP binding sites and their affinities for inhibitors. My shape-based dendrogram shows clustering of the kinases based on similarity in shape. I investigate the pocket in terms of conservation of surrounding amino acids and atoms in order to identify the key determinants of ligand binding. I find that the most conserved regions are the main chain atoms in the hinge region and I show that the tetrahydropyran ring of staurosporine causes induced-fit of the glycine rich loop. I apply multiple linear regression to select distances measured between the distinctive parts of residues which correlate with the binding constants. This method allows me to understand the importance of the size of the gatekeeper residue and the closure between the first glycine of the GXGXXG motif and the aspartate of the DFG loop, which act together to promote tight binding to staurosporine. I also find that the greater the number of hydrogen bonds made by the kinase around the methylamine group of staurosporine, the tighter the binding to staurosporine. The website I have developed allows a better understanding of cross reactivity and may be useful for narrowing down the options for a synthetic strategy to design kinase inhibitors.