bwa sampe error

[m_elena_bioinfo]

Dear,
I run BWA 0.7.5a
>bwa aln genome.fasta read1.fastq > read1.sai
>bwa aln genome.fasta read2.fastq > read2.sai
> bwa sampe -r '@RG\tID:s\tLB:ss\tPL:ILLUMINA\tPU:l1\tSM:' genome.fasta read1.sai read2.sai read1.fastq read2.fastq > alignment.sam

but during sampe step it returns me this error:

[bwa_sai2sam_pe_core] refine gapped alignments... bwa: bwase.c:182: bwa_refine_gapped_core: Assertion `re - rb == rlen' failed.
Aborted

anyone has any idea?

[lh3]

Seems a bug. You may consider to try 0.6.2. The sampe/samse components are more robust there.

[m_elena_bioinfo]

the problem is the index of genome.fasta
if I use bwa0.6.2 for sampe, it returns me:

[bns_restore_core] fail to open file '/dati1/reference/genomes/hg19.fasta.nt.ann'. Abort!

i have tried to use bwa mem (and not aln+sampe) to generate sam file. but gatk want read group in header. could you fix this gap?

[N311V]

Maybe re-index genome.fasta with bwa0.6.2 and then run samtools sampe?

[sunhh]

Maybe re-index genome.fasta with bwa0.6.2 and then run samtools sampe?

No, it doesn't work if you only change the index files.
And if you redo it from the aligning step with the same index files using
0.6.1, it works well.
So it seems a bug in 0.7.5a-r405 .

[jp.]

Hi there
Can someone post here the general commands to run BWA to align paired-end .fastq files ? I just need to have a general look. I have never used BWA but bowtie and tophat using hg19.
(Optional) Can someone also explain the difference while giving commands structure between BWA and bowtie (Optional) ?
Thank you in advance

[dpryan]

Hi there
Can someone post here the general commands to run BWA to align paired-end .fastq files ? I just need to have a general look. I have never used BWA but bowtie and tophat using hg19.
(Optional) Can someone also explain the difference while giving commands structure between BWA and bowtie (Optional) ?
Thank you in advance

Read the bwa webpage (you have a penchant for asking questions that could be easily solved by reading documentation). Also, don't post the same question in multiple threads.

[jp.]

oh..I am sorry..
I got your point, won't do it again.

Read the bwa webpage (you have a penchant for asking questions that could be easily solved by reading documentation). Also, don't post the same question in multiple threads.

[mducar]

Anyone hear about an update on this bug? I'm looking to upgrade the version of bwa we're using and got the same error with 0.7.5a:
[bwa_sai2sam_pe_core] refine gapped alignments... bwa: bwase.c:182: bwa_refine_gapped_core: Assertion `re - rb == rlen' failed.

There hasn't been an update since May -- any idea if this has been fixed and will be available soon?

[g781]

I had the same problem for bwa-0.7.5a. I used the bwa-0.7.4 instead and solved the problem.

The bwa-0.7.5a seems not stable, maybe u can try more stable version of bwa-0.6.2.

Thanks

[probean]

I had the same problem for bwa-0.7.5a. I used the bwa-0.7.4 instead and solved the problem.

The bwa-0.7.5a seems not stable, maybe u can try more stable version of bwa-0.6.2.

Thanks

You saved my day! Thank you!

If anyone else bumps into this problem (especially using 0.7.5a), use 0.7.4 with the same index files... works like a charm.

[zinky]

i use bwa 0.7.5a to do mapping step , it seems good. however, only one of my 21 samples encounter this problem, which makes me confusing. any help message?

[cecile75]

Hi,

I noticed exactly the same thing : bwa 0.7.5 (sampe) worked for 13 of 16 samples, and failed for 3 , without any explanation...
I re-indexed, and re-aligned with bwa 0.6.2 and it worked for all of my samples.

It's a mystery....

[earonesty]

Hi,

I noticed exactly the same thing : bwa 0.7.5 (sampe) worked for 13 of 16 samples, and failed for 3 , without any explanation...
I re-indexed, and re-aligned with bwa 0.6.2 and it worked for all of my samples.

It's a mystery....

I also see this issue. I only use 0.7.5 for "mem", not for anything else.