Our previous work has described the gene expression patterns of human visceral adipose tissue (VAT) at the transcriptome level and reported that the strongly expressed genes in VAT showed an uneven distribution throughout the genome. The aim of the present work was to focus on the unclassified genes and known expressed sequence tags (ESTs) strongly expressed in VAT and analyze their structure and function with bioinformatics. Among the 400 ESTs strongly expressed in the VAT, 340 clones were classified into known genes through searching the latest Genbank database. Functional classification showed that 85 clones were unclassified known genes, and approx 90% of them were found to be expressed in adipose tissue for the first time. Among the 85 unclassified genes, only two share similarities in the coding sequences with all species examined, and six genes had so far no obvious similarity to any genes across different species. The protein products of 7 genes had putative signal peptide and 11 had transmembrane domains. The protein products of 39 genes had relative specific motifs or prosites on primary structure. In silico Northern blot showed that 21 known ESTs were abundantly specifically expressed in adipose tissue, which may provide clues to identify novel genes closely related to adipocyte function with potential pathophysiological implications.