**Warning!**In the first version of the Regulatory Elements Database, the DNase data had a labels swapped, mostly on cell-type treatments: (HCF - HCFaa, HFF - HFF_Myc, WI-38 - WI-38-TAM, and SKNMC - SKMC). This error has been fixed in the current version of the database and downloads (as of November 2015).

# Supplemental Material

- Supplemental Material
- Supplemental Figure S1
- Supplemental Figure S2
- Supplemental Figure S3
- Supplemental Figure S4

# Data

- TableS01-concordanceSummary.txt
- TableS02-openSamples_v3.txt
- TableS03-dhs-to-cluster.txt.tar.gz
- TableS04-cluster-to-openCellTypes_v3.txt
- TableS05-overlapSummary.txt
- TableS06-results_tissueMalig_v2.txt
- TableS07-dhsPredictors_v2.bed
- TableS08-results_sex_v3.txt
- TableS09-motifJaspar.txt

## DHS to gene connections:

The allGeneCorrelations files can be a bit confusing so let me here explain. The "p-value" column in the files is just the proportion of random correlations that were less than the true measured correlation. To truly convert this to a p-value I would have to take 1 minus this value for just one of the tails of the distibution. Instead, I'm just leaving it this way, so you can tell whether the correlation is on the high end or the low end of the permutation distribution. I left it as .996/.998 so you can tell enriched from depleted. So, perhaps I'm wrong to label that column p-value. On the website, I just did the 1-p transformation, and then use color coding to show whether its enriched or depleted. For any connections with p-value < .05, you can download the file labeled p.05. This file contains permutation scores from 0 to .05, or from .95 to 1. Some people are interested in less strict connections; for this you can download the p.2 file.