obtaining BAM files from "load from Encode 2012"

[Rachel Melamed]

Hello!

I am teaching a genomics class using IGV and had a big problem because apparently in the newer versions of IGV there's no BAM files when you do "load from Encode 2012"

I'm thinking it would just be easier for me to download it and give the students the files but I can't find the files corresponding to the ones on IGV. For "K562 RnaSeq Alignments rep 1" I looked at  some below but they are clearly not the right files.

https://hgdownload-test.cse.ucsc.edu/goldenPath/hg19/encodeDCC wgEncodeCshlShortRnaSeq/releaseLatest/wgEncodeCshlShortRnaSeqK562CellShortAln.bam

rsync://hgdownload.soe.ucsc.edu/gbdb/hg19/bbi/wgEncodeGisRnaSeqK562CytosolPapAlnRep1.bam

Can anyone help?

Appreciated!

Rachel

[igv-help]

Hi,  BAM files from Encode have never been available through the IGV menu.  This is not possible because Encode does not index their BAM files.    I'm sorry I can't answer questions about Encode data files, and their is honestly no one in that project who would respond to problems with 2012 data.   Also, I am not sure there is a guarantee that every bigwig file will have a corresponding BAM file.  You might go to the Encode portal and find a more up-to-date consistent set of files to download.

[Mark Maienschein-Cline]

Hi,

I will reiterate Rachel's note, for exactly the same reasons: I teach a bioinformatics workshop course, and we have utilized the BAM files from ENCODE to view in IGV in the past. They absolutely have been there, as we have done it for at least 7 years in a row. They are no longer available in v2.17. I have an older version of IGV (2.10) that does show them just fine.

We can download and put example BAM files somewhere else for the future, but this has been a useful resource for teaching purposes.

Thanks,

Mark

[George Chlipala]

Rachel -

If you can get this message, we had been using the three alignment files from the MCF-7 cell line (see attached picture)

We found the underlying URLs...

• http://hgdownload.cse.ucsc.edu/goldenPath/hg19/encodeDCC/wgEncodeCaltechRnaSeq/wgEncodeCaltechRnaSeqMcf7R2x75Il200AlignsRep1V2.bam
• http://hgdownload.cse.ucsc.edu/goldenPath/hg19/encodeDCC/wgEncodeCaltechRnaSeq/wgEncodeCaltechRnaSeqMcf7R2x75Il200AlignsRep2V2.bam
• http://hgdownload.cse.ucsc.edu/goldenPath/hg19/encodeDCC/wgEncodeCaltechRnaSeq/wgEncodeCaltechRnaSeqMcf7R2x75Il200AlignsRep3V2.bam

We had setup the following shortcut URLs to redirect to the longer URLs.    These work if you specify using the "Load from URL..." menu item in IGV.   Just leave the index field blank.  The index (.bai) files exist as expected (just append .bai to URL) and it seems that IGV can figure that out if you leave the index field blank.

You can use these short URLs if they help. 

• https://wd.cri.uic.edu/ngs/rnaseq_mcf7_A.bam
• https://wd.cri.uic.edu/ngs/rnaseq_mcf7_B.bam
• https://wd.cri.uic.edu/ngs/rnaseq_mcf7_C.bam

These are setup as HTTP redirects. So, the data is not passing through our servers and would be just as fast as using the full length URLs.

Thanks

- George

[James Robinson]

Yes George, you are correct.  I have not worked with the Encode pilot data for years, the production release at the Encode repository (https://www.encodeproject.org/)  does not include bam index files, so is not loadable.   I mistakenly thought this true of the pilot data also.

I will look into what happened here.   Coincidentally I have had it on my list to replace this pilot data with links to the production data,  which we have already done for IGV web,  but this might be a reason to keep the pilot data available.