CRAM and gVCF file in IGV

[adripalu]

Hi everyone,
I'm new to IGV and I have a question that might be stupid.
What are the differences in using CRAM and gVCF instead of BAM and VCF on IGV? There are some details I should be aware of, like uploading the reference FASTA file with the CRAM? Or can I just manage them as BAM and VCF?
Thanks for your help.

[jrobinso]

Mostly you don't need to do anything special unless your CRAM is aligned to a genom we don't have in our list. Note that the GATK and 1KG aligned to a special fasta for hg38, so we provide "hg38_1kg" for those. Also, at the moment CRAM 3.1 files are not supported.

[adripalu]

Thanks for the answer.
The CRAM file are produced by Dragen pipeline that align data to the hg38 reference genome. Given that, should I upload the fasta file or can I just use IGV as usual? What happens if I don't provide the fasta file?
What about the differences of interpretation of the gVCF compared to the VCF?
Thanks

[jrobinso]

You can use hg38. I don't understand the question wrt gVCF but just load it and have a look.

[jrobinso]

I meant you don't have to load the fasta (use IGV's hg38).

[adripalu]

Ok, so basically even if the alignment was performed using dragen pipeline (and data) I can easilly upload the CRAM file on IGV and do my stuff without uploading FASTA reference genome, correct?
Sorry for not being clear. I'd like to know if there are any differences in using the gVCF file instead of the VCF file in IGV. For example, since non-variant regions are also present in the gVCF file, it will be less straightforward to interpret the variants sites, just because their visual data will be less "clean"? Or is there no difference compared to VCF?
I hope I have been clear this time.

[jrobinso]

I think you just have to try it. I don't remember all the details, but I think non-variants sites are a subtle grey. And I don't know anything about dragen.

[adripalu]

Ok, thanks for your help !!