The Three Comparisons Of The Prediction Algorithms Of The Transmembrane Proteins And Its Achievement Through The Software

The prediction algorithms of transmembrane proteins are a special one, which is to forecast the numbers of transmembrane segments in one piece of proteins sequence. According to the amount of data analyzing of transmembrane proteins regions, especially the large analyzing in the topology structure of the transmembrane proteins, to find out the key problems by using the biologic informatics to resolve the prediction of the structure of the transmembrane proteins and then to use of these analysis results to write out the different prediction algorithms aiming at the different proteins (hydrophobicity ,water-solubility ,etc) transmembrane regions .And to carry out these prediction algorithms by using software, which is convenient to query. For instance, given an unknown proteins sequence of transmembrane segments, input this proteins sequence into the software by using the prediction algorithms, the result is as following: how many transmembrane segments in the piece of proteins sequence, and the transmembrane segments' starting and ending location and a series of related coefficient.With the social development and the scientific advancement, computer win the all-time application in all traditional science domain, there are all kinds of boom exploitation in related software .The prediction algorithms of transmembrane proteins which discussed in this article are just written out by MATLAB, aiming atto provide a more convenient, normative, tending to operate query tool in proteins transmembrane regions. So it is meaningful no matter in science or in application.Especially under the context in Sichuan university chemical information center, which is on going to establish cluster servers and chemical remote compute platform on Web, let the prediction algorithms of transmembrane proteins construct on the platform, and the users predict the amount of transmembrane regions in proteins sequence by using remote access. The users can use Web to access conveniently for the share and open of the platform.To sum up, the main research result is as following:1. By study the related knowledge, to analyze the important meaning in the prediction algorithms of transmembrane proteins.2. By reading the documents of prediction algorithms, rebuild three kinds of classical prediction algorithms. And also analyze the procedures and testing results and the discussion of three kinds of prediction algorithms concretely.3. MATLAB finished the software implement of three kinds of prediction algorithms,and took the software integrating to the Internet platform, which is convenient to search and update.