Problem uploading single VCF file

[Former user]

We have created VCF files from agilent SureCall software.We modified the VCF file to have the SRF,SRR,SAF,SAR and are trying to analyze a single sample VCF by uploading it.The error message is the following: Error Error in SNPitty::calculateALTRatios(inputVCF, ratioCalculation): You don't have the required fields to calculate REF/ALT ratio. Either AF, AD, AO & RO or SAF & SAR & SRF & SRR is needed.

Does this mean the problem is with the actual file ? We are not sure if we mounted to docker properly,so some feedback would be appreciated .

Thank you very much for your help !

[Job van Riet]

Ah, I see the confusion.

The AF, AD, AO & RO or SAF & SAR & SRF & SRR should be added to the VCF as GENO fields rather than INFO fields. This allows for sample-specific plotting (if multi-sample VCF).

This means that these values should be added to the rightmost column and the GENO fields identifiers added to the FORMAT column, e.g.:

FORMAT  HS2_0001vcq5_KV_P22_08MB_R1_mgdlanes.fastq
GT:PL:GQ:SRF:SRR:SAF:SAR    1/0:0,122,255:10:1109:876:3:3

I've added the first line of the VCF as an example and this file worked correctly.

Good luck! If anything else doesn't work, just let me know.

[Maria Kalyva]

Hello Job , thanks a lot for the previous help. I am now trying to upload a gtf file that I converted from bed to gtf with https://github.com/pfurio/bed2gtf. Is there any other way you d recommend to do that? I get this error. Any ideas why? thank you again :)

[Job van Riet]

Mhmm, I don't seem to have any errors when using that GTF on my VCF files. Could you maybe upload your VCF so I can check if that throws the error?

[Job van Riet]

Hi,

The VCF was missing the FORMAT tags which tell the VCF what the order and data in the GENO fields are. I've added those and it worked. :)

You might want to check if the order of the fields are correct as now I just named them in the order as they appear in the header.

The GTF was missing a gene_name column. So I've renamed the peak_id to gene_name to fix that issue. This wasn't really obvious so I've added a GTF sanity check and warning to future users during uploading.

Good luck! Hope this works.

[Maria Kalyva]

Works fine!

Now excited to see some results with snpitty! Do you know if I can merge more than 3 samples? or 3 is the maximum it can visualize?

Thanks again ,

maria

[Job van Riet]

Hi Maria,

We've done merging with 8+ samples without complications. It does become a bit of a mess with all the colors but it should work just fine. I would advise to normalize the VCF to split multi-allelic variants if you're merging multiple samples using: bcftools norm

If you don't, then all alternative alleles from the same reference base and position will be summed up. E.g.:

CHR POS REF ALT BAF
chr1 100 C T,A 0.4

versus normalized:

CHR POS REF ALT BAF
chr1 100 C T 0.1
chr1 100 C A 0.3

Unless you just want to know the difference between REF and ALT and don't particularly care about the specific mutation.

[Maria Kalyva]

Hello Job,

Thank you very much for all your help. Our plots are very nice and informative.

I only have one last question, is there an easy way to connect from a host pc to the local pc running snpitty ? If I read the docker manual would I solve this or should I be aware of anything alse ?

Thanks again

Maria

[Job van Riet]

Hi Maria,

If you start SNPitty with an open port on the host, it should work automatically. E.g. if you have an open port (firewall) on a PC with IP: 10.190.3.4, you could do start the docker with: docker run -p <yourport>:3383 -p 3382:3382 ccbc/snpitty

And connect with a browser to: http://10.190.3.4:<yourport>

If you can't connect, than the port probably isn't open for external connections. If you're using Ubuntu or similar distributions, you can open a port with: sudo ufw allow <yourport>

Hope it helps. :)