Bowtie2 hg19 reference for gatk MuTect

Question

[varsha619]

Hello, I understand that the suggested aligner to use with GATK is bwa. If I want to use Bowtie2 as the aligner, which reference file should I be using? The reference in GATK bundle (Homo_sapiens_assembly19.fasta) does not seem to work with Bowtie2 and using hg19 Bowtie2 index from http://bowtie-bio.sourceforge.net/bowtie2/index.shtml gives a compatibility error with dbsnp_138.hg19.vcf file while running GATK MuTect.

ERROR MESSAGE: Input files dbsnp_138.hg19.vcf and reference have incompatible contigs: No overlapping contigs found.

Is there any way to fix this other than switching to the bwa aligner? Is there a different reference or dbsnp vcf file I could use? Thank you for your help.

[h.mon]

The reference in GATK bundle (Homo_sapiens_assembly19.fasta) does not seem to work with Bowtie2

What make you think this reference doesn't work with Bowtie2? Do you encounter any errors? What were the steps you run?

[varsha619]

This is the error I get with the fasta file I downlaoded from - ftp://[e-mail]/bundle/hg19/

bowtie2 -p 4 -x ucsc.hg19.fasta -1 R1_001.fastq.gz -2 R2_001.fastq.gz | samtools view -bS - > 001_out.bam

(ERR): "ucsc.hg19.fasta" does not exist or is not a Bowtie 2 index

Exiting now ...

[samopen] no @SQ lines in the header.

[sam_read1] missing header? Abort!

Do I need to create new indices to be able to use this reference file from the GATK bundle with bowtie2?

[igor]

If you download the sequence from one source and the index from a different source, it is not surprising that they are not be compatible.

[varsha619]

I did keep the reference files consistent through the steps. I meant to say that I tried using both files separately. I can successfully align the files using hg19 Bowtie2 indices but then I get a compatibility error in MuTect with dbsnp_138.hg19.vcf file while running GATK MuTect. If I try to use the reference in GATK bundle to align instead, I get the error: file does not exist or is not a Bowtie 2 index.

[aldhairmedico]

I have another question.

• I want to compare GATK/BWA-MEM + Variant Caller vs GATK/BOWTIE2 + Variant Caller. For that purpose, can I use the GATK interleaved fastq file generated according to their Best Practices or it is necessary to put the original 1 2 paired fastq files?
• This is confusing for me because in Bowtie2 manual there is the option --interleaved that accepts a single interleaved fastq as input but I'm not sure if this is the same file that generated by GATK or not.
• Using the former option I can achieve a more similar result but I'm afraid of using the wrong input.

[igor]

can I use the GATK interleaved fastq file generated according to their Best Practices

GATK Best Practices does not generate FASTQs. FASTQs are the input and come from the user.

[aldhairmedico]

Update: Yes, the Interleaved fastq file generated from unmapped BAMs (uBAMs) can be used as input for Bowtie2

Answer 1

[igor]

Since GATK does not provide a Bowtie2 index, the index does not exist. Thus, you need to generate it yourself.

[varsha619]

Yes this fixed the issue, my bad. Thank you!

[GenoMax]

If an answer was helpful, you should upvote it; if the answer resolved your question, you should mark it as accepted. You can accept more than one if they work.

Answer 2

[h.mon]

Do I need to create new indices to be able to use this reference file from the GATK bundle with bowtie2?

Yes, creating the bowtie2 indices would help.

[varsha619]

Worked after I created the bowtie2 indices for the GATK bundle hg19 genome file, thank you.

Answer 3

[swbarnes2]

Input files dbsnp_138.hg19.vcf and reference have incompatible contigs

Okay, so did you do the obvious thing and check to see that chromosome names are the same between the two files?

[varsha619]

Yes both the files have the format chr1, chr2, ...