Probabilistic Model Based Hough Transform for Detection of Co-expression Patterns in Three-Color cDNA Microarray Data

Three-color cDNA microarrays built on normal-disease-drug samples can be used to assess the effects of a drug on the genomic scale. We have recently shown that the Hough Transform (HT) applied to a two-dimensional representation of the three color intensities can be used to detect groups of co-expressed genes. However, the standard HT is not well suited for the purpose because: (1) the essayed genes need first to be hard-partitioned into equally and differentially expressed genes, causing the HT to ignore possible information in the former group; (2) the two-dimensional gene representations are negatively correlated and there is no direct way of expressing this in the standard HT; (3) it is not clear how to quantify the association of co-expressed genes with the line along which they cluster. We address these deficiencies by formulating a dedicated probabilistic model based HT. The approach is applied to assess the effects of the drug Rg1 on homocysteine-treated human umbilical vein endothetial cells. Compared with our previous study we robustly detect stronger natural groupings of co-expressed genes. Moreover, the gene groups show coherent biological functions with high significance, as detected by the Gene Ontology analysis.