Protein kinases are important enzymes that are responsible for many of the cell signaling pathways in the human body. Years of evolution in proteins have resulted in distinct binding sites and protein function. While there is protein data available for computational analysis, the amount of experimental structural data is small relative to the number of known protein sequences. Protein homology models that are generated computationally can potentially reveal more accurate depictions in computational protein analysis. We propose the use of homology models in the CCORPS method to predict kinase binding affinity.