A system for integrative and post-planned analysis of 2-DE/MS centered proteomics data
Full article [PDF] (903,54 kB)
doi:10.2390/biecoll-jib-2005-12
August 17, 2005
November 10, 2005
Juhui Wang, Christophe Caron, Xuefeng He, Audrey Carpentier, Michel-Yves Mistou, Alain Trubuil, Christophe Gitton, Céline Henry and Alain Guillot
Correspondence should be addressed to:
Juhui Wang
INRA, Lab. of Applied Mathematics and Informatics, 78352 Jouy-en-Josas, France
rf.arni.yuoj@nullgnaW.iuhuJ
Abstract
Proteomic analysis is intrinsically an iterative, incremental process. Information is usually acquired gradually by researchers, and in different projects. At the same time, there are relatively few examples of biological data management systems which take into account this reality, most of them usually treat the experiment generated data as static and unchangeable: data are never reconsidered, or seldom, whereas technology becomes more powerful or that other researchers have brought information on data correction. And yet, postplanned analysis which involves multiple iterations and subsequent re-investigations of previously prepared data might bring tremendous benefits. Named PARIS (Proteomic Analysis and Resources Indexation System), the system we developed here seeks to address this requirement. Compliant with the majority of 2-DE analysis and MALDI-TOF based protein identification softwares, it automatically takes data from them and stores the raw and processed data in a relational database suitable for advanced exploration. Taking into account the standards proposed by PSI (Proteomics Standard Initiative), the system exports the stored data in XML format for data exchange and knowledge sharing. PARIS also manages information about experiments and their biological contexts, and allows the user to search and analyze a large data collection in a global manner. It provides tools for data integration and advanced, cross multi-experiment, multi-experimenter data exploration, and supports visual verification and correction of the analysis results. Implemented in Java, the system is platform independent, accessible to multiple users through Internet. It is also scalable for use for one or many laboratories, and therefore suitable to inter-institute collaborative work.