Matchmaking: Genomic signatures 22/11/06. By the Wellcome Trust A new 'connectivity map' lets researchers match drugs to diseases.

The map, developed by Todd Golub and colleagues (Broad Institute in Cambridge, Massachusetts), is founded on patterns of gene activity in disease. In an individual with a particular condition, the activity of some genes goes up, but that of others falls. So each condition has a 'genomic signature'. With specially developed software, this is used to interrogate a database to identify drugs that are active on conditions with similar genomic signatures.

Two studies of cancer cells illustrate the system in action. With drug resistance a problem in cancer therapy, the US group identified a genomic signature associated with acute lymphoblastic leukaemia cells' sensitivity to glucocorticoid. This profile matched the profile generated by rapamycin, suggesting that rapamycin-treated cells should be responsive to glucocorticoid – a prediction confirmed in lab studies.

In a second study, Dr Golub and colleagues looked at prostate cancer caused by overactivity in the male hormone receptor pathway. Having screened for compounds that inhibited receptor activation, they determined the genomic signatures of these compounds and used the connectivity map to identify drugs with similar profiles. These drugs presumably also interfere with the male hormone receptor pathway, and might therefore be suitable treatments for prostate cancer.

Image credit: Worden Sports College

References

Hieronymus H et al. Gene expression signature-based chemical genomic prediction identifies a novel class of HSP90 pathway modulators. Cancer Cell 2006;10(4):321–30. Abstract

Lamb J et al. The Connectivity Map: using gene-expression signatures to connect small molecules, genes, and disease. Science 2006;313(5795):1929–35. Abstract

Wei G et al. Gene expression-based chemical genomics identifies rapamycin as a modulator of MCL1 and glucocorticoid resistance. Cancer Cell 2006;10(4):331–42. Abstract