April 15, 2005 (Vol. 25, No. 8)
URL:
http://robotics.stanford.edu/~erans/cancer
Rating:
Strong Points: Very ambitious plan well carried out
Weak Points: A bit overwhelming
Summary:
Information in microarrays is rapidly overtaking sequence databases as the king/queen of overwhelming datasets produced by the new biology. In contrast to sequences, which are largely of interest for protein coding and identification of transcription factor binding sites, microarray information is amazingly more complex, since the interpretation of the data requires comparisons of up/down levels of expression of enormous numbers of genes under various changes in condition. The aim of the Module Map site is to attempt to make sense of such data in tumor cells by looking at expression of sets of genes (modules) that seem to move up or down in concert with each other. Almost 2,000 microarray datasets from over 20 tumor types went into creating this database. Data is organized in several fashions, but the most innovative form is that of a clickable image of a microarray grid hyperlinked to relevant genes.
