Connection up- and down-regulation expression analysis of microarrays (CU-DREAM): a physiogenomic discovery tool

Background: Many microarray experiments have been conducted during recent years, and scores of gene expression data have been archived in public databases. The use of data from multiple experiments can provide valuable information. However, there is a lack of convenient tools to compare datasets in this manner.

Objective: Implement software, called CU-DREAM, to compare the datasets of two microarray experiments. CUDREAM is easy to use and compatible with Gene Expression Omnibus (GEO).

Subjects and methods: Five experiments were used to demonstrate the functionality of CU-DREAM. These are GSE6791, GSE7803, GSE5816, GSE4246, and GSE13638 for studies of cancers and RNA interference.

Results: All six showcases demonstrated the validity of the CU-DREAM approach. One showcase could confirm the regulation of genes identified in two independent experiments on cervical cancer. The statistical significance was lower compared with cervical and lung cancers. In addition, CU-DREAM could identify isoform changes in lung cancer. The last showcase demonstrated that Dicer- and Ago2-depleted cells or Dicer-depleted HeLa and HEK293 cells shared the same gene regulation pathways. CU-DREAM had seven main functions: 1) to identify genes that are up- and down-regulated in an experiment, 2) to validate significantly regulated genes using data from another experiment, 3) to determine if two different diseases have a similar effect on gene regulation, 4) to identify isoform-changed genes, 5) to determine if cells share gene regulation mechanisms, 6) to identify common gene regulation pathways even when comparing two different cell types, and 7) to identify down-stream genes that are regulated by the conditions of the analyzed experiments.

Conclusion: CU-DREAM is an effective tool for the pre-screening of drugs, substances or environmental insults or the identification of the genetic changes that are associated with pathological conditions (CU-DREAM can be downloaded from: http://pioneer.netserv.chula.ac.th/~achatcha/cu-dream).