About

About the Human 8.3 Database

Usage note: Filters and cookies
cisRED manages your 'Filter' settings via a browser 'cookie'. You must allow your web browser to accept cookies from cisred.org for your filter settings to take effect.

Motif discovery and post-processing for promoter regions of ~15K protein-coding target genes (Ensembl v35, NBCI v35, hg17) returned 205K motifs after applying a discovery p-value threshold of 0.01.

Motifs were discovered in sequence sets consisting of regions around the transcription start site (TSS) of a single, canonical transcript for each gene, and corresponding regions from other species. Input sequence sets contained a total of 32 vertebrate species whose genome data were taken from Ensembl, ENCODE and low coverage read files. A typical (mode) input sequence set contained 11 vertebrate species. Search regions were -1.5Kb/+200b relative to a TSS, net of most types of repeats and of coding sequences, which were masked. For v8.1, 'annotation-based' modules, i.e. co-occurring patterns of motifs that were similar to known transcription factor binding site data in TRANSFAC, JASPAR and ORegAnno were added.

This database contains 4,137 annotation-based modules. In this version of the human 8.x database, GoMiner has been used to assign a list of terms from the Gene Ontology (GO) project to each module, based on the known functions of the genes associated with that module. The module-view pages on the cisred.org website display all GO terms assigned to each module with an FDR ≤ 0.01.

Direct SQL queries can be run against the database cisred_Hsap_8_3 at db.cisred.org, with the username 'anonymous', and no password. The newly added pattern_go_terms table contains all the results returned by GoMiner, including GO term assignments with an FDR > 0.01.

An overview of the pipeline and database is available in an NAR 2006 publication. A manuscript that describes the methods currently used to identify conserved genomic elements is in preparation.