Hi,
I am have a question related to the bed files for each DNase-seq experiment in ENCODE. I wanted to download all DNA-seq bed files and found that I could do this through “Experiment Search” at “https://www.encodeproject.org/search/?type=Experiment” and by clicking on the “Download” button, this would give me a list of URLs to bed files for all the experiments I've selected.
Using the bed file accession “ENCFF553WEU” as an example, I found through the file summary page (https://www.encodeproject.org/files/ENCFF553WEU/ ) that the corresponding Experiment accession is ENCSR224IYD. However, when I visit the Experiment Summary page at https://www.encodeproject.org/experiments/ENCSR224IYD/, I found (under the heading “ENCODE4 v3.0.0-alpha.2 GRCh38 (ENCAN672OTQ) processed data” in the File Details tab) that this experiment has, in fact, 2 other “bed narrowPeak” files besides “ENCFF553WEU”: one being ENCFF112RNW and the other, ENCFF617YMX.
My question is: Why are there more than one “bed narrowPeak” file for this experiment? And, why did the downloaded list of bed file accessions from the “Experiment Search” only give the bed file accession “ENCFF553WEU” and not the other 2 “bed narrowPeak” files for this experiment?
Thanks in advance for your help.
I have the same question, couldn't figure out what are the differences between the files.
Apparently, according to the file names, the differences are the p-values used for the IDR: 0.05 and 0.01.
Hi Elisheva, thanks for your reply. Sorry but I did not get an alert that you had replied to my question. You are indeed right. I learnt that one narrowPeak bed file has 5% False Discovery Rate (FDR) peak calls and the other narrowPeak bed file has 0.1% FDR peak calls. And the narrowPeak bed file with the highest number of mapped reads (with 0.1% FDR) is given as the default downloaded file, which in the example I cited is “ENCFF553WEU”