vcf to MAF to fasta

Question

[natasha.sernova]

Dear all,

A strange idea has came to my mind recently.

There is no direct way to convert vcf into fasta. But

I've read that through Galaxy I can convert vcf to maf (multiple align),

and then it may be possible to convert maf to fasta.

How much information from the original vcf-file will I loose using this way,

and is it possible to avoid the loss?

THANK YOU!

Natasha

Answer 1

[Pierre Lindenbaum]

"There is no direct way to convert vcf into fasta" : there is. The GATK provides a tool named FastaAlternateReferenceMaker http://www.broadinstitute.org/gatk/gatkdocs/org_broadinstitute_sting_gatk_walkers_fasta_FastaAlternateReferenceMaker.html

[natasha.sernova]

Yes, that's true. I've even used it once.

But this option killed me.

-L input.intervals \
I still haven't known (if I have a particular
vcf-file number 1), how and where
to find the corresponding interval coordinates.

You definitely know. PLEASE, help me!

Thank you very much indeed!

Natasha

[Pierre Lindenbaum]

is -L required ?

[natasha.sernova]

Dear Pierre,

this is optional parameter. But when I omit it, I have just the full fasta-fail for corresponding chromosome, nothing else. I thought that GATK will allow me to cut a fragment corresponding.to vcf-file. But I have to give it "one or more genomic intervals over which to operate". It seems much more complicated. How to find these genomic intervals?

I have vcf-files, refs and chromosome sequences. I made fai-files, dict-files, but it still doesn't give me any hints to the fasta alignment. What else should be done?

Many thanks for your help!

Natasha

[Pierre Lindenbaum]

"one or more genomic intervals over which to operate": I don't understand your problem. You can provide a BED file or even a VCF file: http://www.broadinstitute.org/gatk/gatkdocs/org_broadinstitute_sting_gatk_CommandLineGATK.html#--intervals

[natasha.sernova]

Great. And this will work??? " In order to perform the analysis

at specific positions based on the records present in the file

(e.g. -L file.vcf)" - this is exactly what I need. I will try, that's a miracle.

I was very poor in reading the manual... THANK YOU, Pierre!

[pld]

I guess I missed the goal of OPs question, I just was not sure what they wanted in the fasta file. I presumed that they didn't have a preexisting fasta file with their reference sequences.

Answer 2

[pld]

MAF is Mutation Annotation Format, not a multiple alignment:https://wiki.nci.nih.gov/display/TCGA/Mutation+Annotation+Format+%28MAF%29+Specification+-+v2.4

Fasta is really a flat sequence data format, you might be able to store the variants found in a sequence in the fasta header, but I'm not sure what you're asking for makes sense.

[lh3]

The Multiple Alignment Format (MAF) has been used by UCSC/EnsEMBL etc for ten years or so. It is one of the most widely used multi-alignment formats. The first spec for the Mutation Annotation Format was only released two years ago.

[natasha.sernova]

https://wiki.nci.nih.gov/dosearchsite.action?queryString=Multiple+alignment+format

So my question doesn't make sense at all. OK, thank you.

[pld]

That link just leads me to Mutation Annotation Format.