Research Of Protein Purification System Based On Data Mining

Genetic engineering recombinant protein drugs are one of the important development directions in new drug development. Recombinant protein refers to the use of recombinant DNA techniques in the production of the protein, whose species construct to fermentation, purification technology research development and industrialization of the recombinant protein is the important research content. At present, the recombinant protein purification process of exploration is mainly based on the professional experiment repeatedly, which makes the process of rely heavily on the experiences of professionals, and long development cycle, high cost. In view of the above problems, this paper presents a protein purification system based on data mining, according to the existing purification experience information for purification process development staff to provide effective suggestions to shorten the cycle of process research.Protein purification is largely associated with the nature of the protein. At present, there are hundreds of recombinant protein purification technology research reports in the open literature, that we can find the association between purification process and the properties of protein by data mining technology effectively. Based on this idea, this paper uses K-means algorithm to take the recombinant protein for clustering analysis, and then get the main purification method of each category. But the recombinant protein properties data cannot be used directly for K-means algorithm in clustering, that the data attributes quantification, denoising and standardized treatment before the clustering. In actual production application, some properties of the recombinant protein are associated with some step in the process of the protein purification process, and then, some rules will be used to adjust and improve the process we get form clustering algorithm. We choose dozen of the known recombinant proteins to determine the purification process by using the proposed method, that the advice process and practical application process is proof that the proposed method is effective. Finally, based on the proposed method, we use of Visual Studio 2008+SQL Server 2005 developed a protein purification system, and the system run well.