Computational Inference of Compound-induced Anti-inflammatory Effects Across Time in an Adjuvant-induced Arthritis Rat Model

A number of diseases, such as arthritis and cardiovascular disorders impacting the lives of many people have strong inflammatory components. To elucidate the antiinflammatory mechanism of a Novartis compound, time-course gene expression data were collected from joint tissue in an adjuvant-induced arthritis rat model, with and without compound treatment. The gene expression data were then analyzed using a dynamic Bayesian network (DBN) inference algorithm. Due to the nature of the biological experiments, a prescreening method to select a gene list and a data pair-up method to prepare the data were developed prior to applying DBN. From the inference analysis, a well-known cell adhesion molecule implicated in several inflammatory diseases was suggested as a potential direct target of the compound. In addition, a biologically plausible downstream pathway, modulated by the compound, was revealed. Experimental validation studies to confirm the compound direct target are in progress.

• Type: Conference
• Year: 2006
• Where: BIOCOMP
• Authors: Jing Yu, Gabriel Helmlinger, Muriel Saulnier, Anna Georgieva
• Keywords: BIOCOMP 2006, BIOCOMP 2007, Compound Direct Target, Gene Expression Data, Novartis Compound