A Novel Classification Patterns Of Cellulases And New Tool Of Analyzing Southern Blotting Based On Bioinformatics Approach

The work reported in this article mainly includes two parts: developing a new method to classify cellulases and designing a novel tool to assist Southern (?)lotting scheme and to analyze data of Southern blotting. Methods of researching gene function and detecting foreign DNA segment's insertion are introduced in this article as well as researches on cellulase classification.Researchers have developed many ways to classify cellulases, the main of those methods only employed sequence similarities or conserved clusters to classify them. In this article, a fairly different process is developed to group cellulases via both sequence similarities and information of conserved domains. Moreover, detailed analysis is performed to discover the phyletic distribution of different cellulases.After acquired sequences from GenBank, those sequences were divided into raw groups via multiple alignment and a simple program named TreeConvert. Conserved domains were dug out by HMMER program, and data mining process was employed to find association rules of domains with apriori algorithm. By this way, nearly 2000 enzyme sequences were classified into 10 patterns of conserved domains. Used information of both sequence similarities and conserved domains but not only one of them, this process can avoid the influence of difference in sequence similarities and conserving degree of domains. Therefore, this process is a commonly usable protocol to classify any kind of enzymes.Southern blotting has been widely used in molecular biological research, and its general usage is to confirm whether a foreign segment has been inserted into a target genome. Other experiment methods can also be further employed to ensure insertion sites and number of copies of foreign fragments. Sometimes, using different enzymes digest genome respectively and counting signal numbers of Southern blotting, copy number of foreign segments can be cursorily deduced. In the article, a software package named SouthLoc, which utilizes genomic sequence to design optimal enzyme combination scheme of Southern blotting, is developed to enhance the accuracy of the above cursory method. Therefore, the locations and copy numbers of foreign segments can be detected with Southern blotting and SouthLoc. Resorting to this powerful tool, the application range of Southern blotting is extended.SouthLoc uses two rating policies to rate the enzyme combinations for Southern blotting. The first rating policy is to calculate the score of average appearance times of total segment combinations' location regions in genome, and the second is to calculate the score of average cover range. The former is fit for finding enzyme combination which will generate the fewest location regions, and the latter is suitable to find enzyme combinations that can produce location regions with the shortest average size. SouthLoc rates different combination schemes and finds out the best scheme. Because of huge computation amount, we only rate two-enzyme combinations at first. Then, suitable transform policies and efficient algorithms are applied to calculate the optimal multi-enzyme combination schemes.In order to confirm the correctness of SouthLoc, Southern blotting date of Aspergillus fumigatus is used as a sample in this article to forecast the optimal enzyme combinations and to analyze insertion sites. The results of SouthLoc match the actual experiment results very well. Therefore, SouthLoc can be used as an efficient tool to analyze Southern blotting data. insertion sites of foreign segments can be calculated accurately to facilitate further experiments. Since SouthLoc uses data of Southern blotting as the only source data and it focuses on special genome, it is very valuable not only in analysis of foreign gene's insertion status but also in large-scale analysis of insertional mutagenesis. Therefore, the tool has great application potential.