Construction Of Protein Databases For Disease-related Mutations And Protein-centered Interactions

Human genetic diseases have been a threat to human health and life for a long term. With the development of genetics and molecular biological techniques, many gene mutations which can cause some human genetic diseases by changing amino acid sequences have been identified. A disease-related integrated human mutated protein sequence dataset, called as dIMS, which collected 34,891 dIMS from OMIM, PMD and SwissProt, was constructed. The initial analysis for dIMS was conducted from three aspects, including the classification of dIMS according to disease information, amino acid mutational spectrum analysis, and the analysis of the relationship between disease-related point mutations and functional domains. Based on the dIMS, a web-based system, SysPIMP (the Systematic Platform for Identifying Mutated Protein; http://syspimp.starflr.info/) was constructed not only for browsing dIMSs, but also for identifying disease-related human mutated proteins from the mass spectrometry results. Almost all proteins conduct their own functions through interactions with all kinds of other molecules including proteins. For better understanding the mechanisms of human genetic diseases caused by gene mutations, the research for the interaction networks these disease-related proteins are involved in is necessary. For satisfying this, a web-based integrated protein-centered interaction database (IPID; http://ipid.starflr.info/), collecting 2,065,735 protein-related interactions from 25 public interaction databases, covering five different interaction types, was constructed. After removing redundancy, IPID collected 560,442 non-redundant protein-related interactions, including 198,947 human non-redundant protein-related interactions. InterX!Tandem implemented in IPID is used to identify proteins from mass spectrometry results and provide all kinds of interactions stored in IPID which are related to those identified proteins. On the basis of IPID, 22 disease-related human interaction networks determined by Human Disease Networks were investigated. The construction of these two systems, SysPIMP and IPID, will be helpful for further researches and diagnoses of human genetic diseases.