About IIDB

Background

Transcriptional regulation underlies macrophage responses to TLR signaling. Differential transcriptional activity in response to TLR signaling tailors macrophage responses to different pathogens. In spite of dramatic recent achievements, the cost and difficulty of comprehensive experimental identification of transcription factor binding sites (e.g. using ChIP-chip technology) continues to be high, both financially and also in terms of manpower and time needed. It is therefore highly advantageous to use computational methods to make an initial prediction of key transcription factors regulating macrophage gene expression changes in responses to pathogens.

IIDB contains annotations for over 2000 TLR-responsive mouse genes annotated using data from over 100 microarray experiments, the ENSEMBL database, NCBI database, computationally predicted TFBS, predicted cis- regulatory modules, evolutionary conserved regulatory sequences, DNase hypersensitive sites, co-expression clusters, and sample ChIP-chip data. For each mouse gene, we have analyzed the sequence from at least 20 kb upstream of the transcription start site to at least 10kb downstream of the transcription end site, including all exons and introns. Annotations for DNase hypersensitive sites, as well as exon/intron boundaries, CpG islands, repeats, Affymetrix GeneChip expression array probes (Mouse 320 2.0) are included in our database to facilitate analysis for users with ChIP-chip data. Users can interactively interrogate IIDB using a web interface. Search results are automatically annotated on the genomic DNA sequence and imported via .gff2 and .gff3 files into the Argo genome browser

Construction and Content

IIDB uses the MYSQL relational database system to store, retrieve and manage the data. The web interface between the user and IIDB is coded in PERL/CGI.

Acknowledgements

M. Korb designed and implemented the IIDB database. MK and AR jointly implemented the user interface. A. Rust, V. Thorsson, C. Battail, B. Li, D. Hwang developed the statistics and algorithms for the transcription factor binding site predictions. All computational work was carried out under the guidance of H. Bolouri and I. Shmulevich. K. Kennedy, J. Roach, C. Rosenberger, M. Gilchrist , D. Zak, C. Baldwin, B. Marzolf performed the ChIP-Chip and microarray experiments under the guidance of A. Aderem.

We thank R. Engels at the Broad Institute for his assistance with Argo customization and E. Deutsch at the Institute for Systems Biology for technical advice.

This research was supported in part by NIAID grants U54AI057160 and 5K08AI056092. Support was also provided by the NIGMS grant PM50 GMO76757/Center for Systems Biology.