IGV - Genome problems

[tonio100680]

Hello,

I want to use IGV to see my GAIIx data (CASAVA v1.8). I import my bam files(.bam and .bam.bai) in IGV. I have to use a different genome because I can see my data. I imported the genome project 1000genome. I created an alias file (previously seen in a forum). I watch my reads and my coverage. I can not add annotations as refseqgene ().
Have seen a solution?
Thank you in advance for your help

[Heisman]

Maybe the website is down (or you don't have internet access)? It may not be on your end at all.

[tonio100680]

No, I have Internet access. The .Bam files (one per chromosome generated by CASAVA) should be merged (merge samtools)?

[Jim Robinson]

Hi, that's because we don't have any annotations for a genome with id "b37_CFB". When you created a "custom" genome you can include an annotation file at that time, or you can load them yourself from the file menu, but we only have annotations on our server for assemblies we host.

Is there a reason you can't just select "b37" or "hg19" as your genome? What are the chromosome names like in your BAM file?

Jim

[tonio100680]

I tested with your genomes (hg19 and b37) and I get the same results (ie none). My chromosomes name : chr17.fa for example.
Should we merge the . bam files (one per chromosome) ?

[pmiguel]

I tested with your genomes (hg19 and b37) and I get the same results (ie none). My chromosomes name : chr17.fa for example.
Should we merge the . bam files (one per chromosome) ?

You have to either get the chromosomes names in your .bam file(s) to match those in the hg19 genome annotation, or create an alias file, as described here.

One issue not addressed in the above link is how to find out what your chromosomes are named in your .bam files and what the chromosomes are named in the annotation files. I use samtools to get a list of the chromosomes I have mapped to from the .bam file:

samtools view -H mydata.bam

You apparently already have this information -- so, that is good. Now, if you look at the list of chromosomes for hg18, you will see they look very similar to your names. You just have a ".fa" suffix, whereas hg18 does not. So your alias files would be of the form:

chr17.fa <tab> chr17

where " <tab> " is a single tab character. You need a line for each chromosome of interest. The file should be called "hg18_alias.tab" and needs to be saved in your "genomes" directory.

(BTW, if you happen to be reading this Jim, I always have some difficulty locating this directory. Would be greatly preferred if there were an "import alias table" from the IGV menu that could place the file in the desired location. Also, would be nice if it would do a little sanity checking on the file to make sure the lines are parseable and that the second column names all correspond to names in the current annotation file.)

--
Phillip

[tonio100680]

Yet what I did (hg19_alias.tab):
chr1.fa chr1
chr2.fa chr2
chr3.fa chr3
chr4.fa chr4
chr5.fa chr5
chr6.fa chr6
chr7.fa chr7
chr8.fa chr8
chr9.fa chr9
chr10.fa chr10
chr11.fa chr11
chr12.fa chr12
chr13.fa chr13
chr14.fa chr14
chr15.fa chr15
chr16.fa chr16
chr17.fa chr17
chr18.fa chr18
chr19.fa chr19
chr20.fa chr20
chr21.fa chr21
chr22.fa chr22
chrX.fa chrX
chrY.fa chrY
chrM.fa chrM

[Jim Robinson]

Hi Tonio,

2 things. First, is that alias file tab delimited (its impossible to tell from the posting of course). Secondly, did you try just loading the BAM without the alias file? The alias is a general solution, but certain "well known" mappings are automatically generated. chr1.fa -> chr1 is one of those, as is 1 -> chr1 and chr1 -> 1.

Loading annotations from our server is a separate issue, you should find some annotations for the human genomes we host, more for hg18 than hg19 (or b37) but not zero. If you are seeing nothing please send us a screenshot of any error or message you get (again with our hosted genome selected, not your custom one). You can post it here or email it to igv-help (at) broadinstitute.org.

Jim