I have downloaded a .bed a ChiP-seq data set from the GEO database, and whenever I try to upload it to the UCSC browser for viewing I get an error.
This is the error: Error line 2 of custom track: Expecting + or - in strand
How do I fix this? I am new to handling and analyzing Chip-seq data.
Thanks!!
Can you send the first two lines of the file and/or the URL where you downloaded it from GEO?
https://www.ncbi.nlm.nih.gov/geo/query/acc.cgi?acc=GSM1480611
https://www.ncbi.nlm.nih.gov/geo/query/acc.cgi?acc=GSM1480613
Angie Hinrichs I have attached the two links. Is there a specific software needed to be able to edit chip-seq files? I am very new to chip-seq, so any guidance is much appreciated.
Hi Misha Fini , I took a look at the ".bed" file https://www.ncbi.nlm.nih.gov/geo/download/?acc=GSM1480611&format=file&file=GSM1480611%5FCD4%5FAb%5Ffdr0%2E05%5Fpeaks%2Ebed%2Egz and it includes non-standard columns that don't conform to the BED format (https://genome.ucsc.edu/FAQ/FAQformat#format1). In order to use it as a custom track, I suggest downloading the file locally and then extracting the first 3 or 4 columns to get standard BED.
1. Save the file GSM1480611_CD4_Ab_fdr0.05_peaks.bed.gz locally.
2. If you have standard Linux or Mac OS X command-line tools, then you can use a terminal to keep only the standard columns and lose the non-standard columns, saving to a new file:
gunzip -c GSM1480611_CD4_Ab_fdr0.05_peaks.bed.gz | cut -f 1-4 | gzip -c > GSM1480611_CD4_Ab_fdr0.05_peaks.fixed.bed.gz
3. Upload the new file to UCSC as a custom track. That particular file is not prohibitively large, but if you work with larger files that cause upload timeouts then you might consider converting large .bed files to the bigBed format (https://genome.ucsc.edu/FAQ/FAQformat#format1.5).
The fourth column, which will be displayed as the item's name, is simply the item length which may not be as interesting as some of the other columns. The file includes a header line that labels the columns:
gunzip -c GSM1480611_CD4_Ab_fdr0.05_peaks.bed.gz | head -1
#chrom start end length foldChange p.value summit
If you would rather use the 6th column ("p.value") instead of the 4th column, then you can use "cut -f 1-3,6" instead of "cut -f 1-4" in the command pipe above.
Hope that helps. Also, for future reference the UCSC Genome Browser team operates both public/archived/searchable and private email lists for questions about how to use the Genome Browser, see https://genome.ucsc.edu/contacts.html .
- Badges
- Science topic
- Similar topics
- Filing