Methods: The SKB contains a relational database designed with the features meeting the standards proposed by the proteomics initiative. The database stores the datasets derived from high throughput salivary proteomics and transcriptomics experiments. MySql, Apache, perl, and CGI are the main computational tools used to construct the SKB.
Results: In health subjects, 11248 distinct peptides have been identified from salivary fluids (parotid/sm/sl), inferring 1166 distinct proteins; while 12679 unique peptides have been identified from whole saliva, inferring 1457 distinct proteins. On the transcriptome side, a high confidence set of 1370 exons representing 851 unique mRNA transcripts have been identified. The web interface of the SKB (http://www.hspp.ucla.edu) is equipped with the features of viewing, querying, and downloading salivary proteomics datasets and allows exploration of annotation information. The distributions of the salivaomics datasets have been mapped on the human genome. Comparison has been performed between transcriptomes of the saliva from health and cancer subjects. A sophisticated viewer of the salivary proteome and transcriptome is being built based on the GBrowse, a tool previously used to mine multiple dimensional datasets in genomics studies.
Conclusions: The SKB is the only web resource dedicated to the salivary omics studies and contains valuable information to explore the biology, diagnostic potentials, pharmacoproteomics and pharmacogenomics of human saliva.
Supported by PHS UO1 DE 16275 and R01 DE 15970.