| Proteomics is a novel discipline of the research on proteins expressed in cells or tissues. It makes possible to explain the structures and functions of protein on the whole level. Proteomics has become one of the most active areas of life science research in the post-genomic era. There are three basic technologies of proteomic researching:two-dimensional electrophoresis technology, computer image analysis and large-scale data processing technology as well as mass spectrometry based technology. Among them, the mass spectrometry based protein identification technology has become the most important one, because of its high sensitivity, fast speed and easy to carry out automation. The process of protein identification using biological mass spectrometry has two steps:experimental and calculation. Through them, the mass spectrum and spectrum analysis are obtained from protein sample, respectively. The existing methods of spectrum analysis include sequence database searching, map database searching, de novo sequencing and combination of de novo sequencing with fault-tolerant search methods. And the method of sequence database searching is wildly used. At present, Mascot, SEQUEST, X! Tandem are most commonly used databases searching tools. The basic progress of sequence database searching method is the candidate proteins in database are digested as peptides theoretically to simulate fragment spectrum generated by theoretical enzyme peptides. Second, match the theoretical maps with the experimental maps, high reliable peptide identification results will be obtained according to the similarity scoring by the quality of a particular peptide control method. Third, it is to deduce proteins and achieve protein identification, in terms of corresponding relationship between amino acid sequence of peptides and proteins.Human liver proteome is one of the focuses in China because of the physiology and pathyology significance of liver. Our laboratory has built the liver proteome library, collecting the liver proteome expression profiling, protein localization and protein interaction data, which are generated by experiments in the "Tenth five-year" period. An important part of the general library is the liver protein expression profile database (DBLEP), which is now a mature human liver database internationally. The purpose of DBLEP is to manage, browse, query and download the existing expression spectrum data sets. (including proteomes of human fetal liver, French adult liver, Chinese adult liver, liver sub-cellular). Furthermore, the focus of DBLEP is to display data fully from various angles and provide a variety of statistics information about expression profiles, reflecting the research status of protein expression profile on human liver.Thousands of high-confidence proteins have been identified in expression profile database of French and Chinese adult liver. Due to the diversity of mass spectrometer, the distinct database searching software and construction methods of theoretical patterns, the data is of high complexity, which mainly reflected in data format, more originality and the complex relationship among these data in the process of protein identification. The diverse massive proteomics data processing is not only time-consuming and labor-intensive, but also unguaranteed of the quality because of the manual methods. Some data analysis tools are frequently used internationally, such as TPP (Trans-Proteomic Pipeline), CellMapBase and so on. They have been integrated with more comprehensive functionality but are not suitable in MS data processing for our lab. So a platform for data processing named MSDataCruiser has been developed. The large-scale mass spectrometry data could be processed on the modularized and integrated platform, which also display the massive data sets in an intuitive and easy to operate way.There are totally 51 batches data sets of liver organelle protein expression profile of human and mouse, responding to different technical lines. These data were generated by seven domestic laboratories including BPRC (Beijing Proteome Research Center). In this paper the procedure of data process is first discussed, including format conversion about LCQ and LTQ data, processing of identified results generated by different database searching software, and the annotation information of identified proteins. In the first part of the paper, the integration of standard files to be imported into the DBLEP, the publication and sharing of data sets are also explained. The processing of MS data was implemented through java programs and tools developed by our lab. In the second part, the BLAST server was added and modified to support sequence-similarity searches against the new data sets and the all data sets in human liver protein expression profile database for users. The development environment as well as design and implementation of each module in MSDataCruiser are illustrated in the last part of this paper. Six main functional modules have been integrated into the platform:format conversion, the processing about identification results of SEQUEST PFF, Mascot PFF and Mascot PMF Combine, the obtaining of protein documents and protein annotation information. In addition, FAQ and some tools for downloading have also been provided by MSDataCruiser.MyEclipse6.6 was choosed as the development environment. MSDataCruiser was developed using "JSP+JavaBean+Servlet" technology, which follows the MVC (Model View Controller) model. Apache Tomat5.5 was used as Web containers. The MSDataCruiser is suitable for processing expression profile data generated on all MS platforms. User-friendly web interfaces are provided and the modular designation makes the platform versatile and easy to operate. |
