Additional file 1.
List of modENCODE histone modification profiles along each developmental stage that are incorporated into the prediction of CBS enhancers. For each ChIP-seq profile, the following information is given: histone mark, developmental stage, number of regions significantly enriched on this sample as compared to a control as reported by modENCODE, genome coverage, and the NCBI-GEO accession code. Additional file 2. List of combinations between modENCODE histone modification profiles along each developmental stage that are incorporated into the prediction of CBS enhancers. For each combination we show this information: combination of histone marks, developmental stage, genome coverage and percentage of H3K4Me1 regions that present intersection with the second mark. Additional file 3. List of combinations between modENCODE histone modification profiles along each developmental stage that are incorporated into the prediction of CBS enhancers after removing those regions overlapping RefSeq exons. For each combination, the set of histone marks, developmental stage, and genome coverage is given. Additional file 4. Visualization of CBS information on the modENCODE genome Browser. The following information is displayed (from top to bottom): FlyBase en gene annotation CBS predictions for several TFs that are known to participate in the regulation of this gene, REDfly experimental CRMs on this locus, BLS predictions, and ChIP-seq information about H3K4Me1, H3K27Ac, and H3K27Me3, as provided by modENCODE. Additional file 5. Dynamic regulatory pattern landscape along a genome region in embryos (12–16 h). The following information tracks are displayed along this fragment of 300 kb: (i) modENCODE H3K4Me1 ChIP-seq profile in red, H3K27Ac in blue, and H3K27Me3 in green; (ii) CBS evolutionarily conserved enhancers derived from previous profiles; active enhancers are highlighted in blue, and poised enhancers, in green; and (iii) RefSeq gene annotation and UCSC conservation tracks. Additional file 6. Characterizing putative enhancers in a gene-free region with CBS. A region of 50 kb in chromosome 2L that does not contain any RefSeq annotation is shown. The H3K4Me1/H3K4Me3 profiles, and the set of putative enhancers evolutionarily conserved predicted by CBS at this locus, are shown for embryos of 0–4 h and 12–16 h. Additional file 7. Conservation levels in TF binding regions of five different ChIP-seq experiments from the modENCODE project. Conservation was calculated as the average phastCons value along each hit, as reported by modENCODE. Additional file 8. Accuracy evaluation of CBS predictions on modENCODE GAF/Trl ChIP-seq binding regions. At the top, the distribution of successfully identified ChIP-seq sites for CBS predictions is shown, taking into account different levels of sequence conservation. At the bottom, the ratio between the number of predictions and the number of successfully identified ChIP-seq sites for the same conservation thresholds is given. Additional file 9. Evaluation of CBS annotations in the even-skipped gene stripe 2 enhancer (GenBank: AF042709, dm3: chr2R:5865217–5865890). The following binding sites have been experimentally validated : bicoid (+138, +159, +310, +403, +521), hunchback (+496, +578, +661), and Kruppel (+3, +139, +327, +521, +571, +615). At the top, we show the UCSC multiz15way conservation track. For each TF, we display the MatScan matrix score, CBS annotations, and experimental sites. This figure was graphically customized from original CBS results, incorporating the location of experimentally validated sites and the score of the weight matrix predictions into the final picture.
Format: PDF Size: 404KB Download file
This file can be viewed with: Adobe Acrobat Reader
Blanco and Corominas BMC Genomics 2012 13:688 doi:10.1186/1471-2164-13-688