A novel model to pinpoint human androgen receptor targets developed

A novel computational model to pinpoint androgen receptor targets within the human genome was recently reported by the joint team of Genomatix Software GmbH, Munich, Germany (www.genomatix.de), and Center for Prostate Disease Research (CPDR), Uniformed Services University, Bethesda, Maryland, USA (www.cpdr.org). Identification of androgen receptor target sites is crucial for the understanding of prostate cancer that is the second leading cause of cancer related deaths among men. The unprecedented accuracy of the formula was achieved by training the computer to recognize common motifs within the regulatory sequences of androgen-inducible genes. It came as a surprise that the computer pointed to a composite model indicating consistently tight distance correlation between two transcription factor matrix families. Researchers in the collaborative team systematically tested and improved the model by experimentally verifying androgen receptor binding to predicted target sites within the prostate cancer genome. Results revealed a stunning concordance between computation and wetlab experiments. The teams led by Dr. Albert Dobi (CPDR) and by Dr. Thomas Werner (Genomatix GmbH), concluded on the pages of the Journal of Molecular Biology November 4th issue (JMB, vol. 353, pp.763, 2005) that composite models, biological context and the hand-in-hand application of computational and experimental biology can indeed yield superior tools for cancer biology.

---

---