IGV displays ">chr1 " as part of genome sequence

[jwfoley]

I was trying to figure out why all my chr1 alignments looked like they were full of mismatches in IGV, and then I noticed they were all off by 6 bp. I zoomed in to the beginning of chr1 to compare it with a FASTA file, and I saw this:


It looks like IGV somehow parsed the ">chr1 " header in the FASTA file as part of the sequence! Only chr1 is affected:


To verify that this wasn't something wrong with my downloaded genome, I removed it and re-downloaded it, but the problem persists. Then I tried downloading some other genomes.

mm10 is affected:


and hg19:


but not hg18, which was preinstalled:


I've reproduced this problem on two different computers using IGV versions 2.3.20 and 2.3.25. Is anyone else seeing this? Is it a bug? It doesn't seem like the kind of thing that would escape everyone's notice; all the coordinates are off by 6.

[westerman]

I do not see the same problem. hg18 was the same as yours (no problem) and just now loading in hg19 and mm10 shows just the 'N's at the start of chr1. Perhaps there was a problem on the 25th? I suggest reloading in the references.