Entering edit mode
8.6 years ago
Sissi
▴
60
Hi guys,
I've read a lot around but I still can't find a solution. I need to map (with BWA bwasw) contigs on a Reference genome and they share only 70% of identity. Using the default parameters I don't see any mapped contigs. So, I would like to allow more mismatches. Which parameters should I change?
Thanks a lot
Hi Agata!
Thanks for your reply! I tried to change -b in bwa bwasw and -B in bwa mem, but when I export the mapped contigs with samtools, the fasta file is empty. Something is still wrong :/
Sil
Could you write commands for bwa i samtools, please.
Agata, here the commands I used:
bwa bwasw -b 10 ../Ref.fasta oases-all.fasta | samtools view -bS - > G10.bam samtools view -F 4 G10.bam | awk '{print">"$1"\n"$10}' > G10.fasta
How big is your BAM file? Maybe wrong reference?
The bam file is 62kb and the reference quite short, just 2000bp more or less. By blastn I'm sure there are some contigs that match to this Reference, that's why I'm pretty sure that If I map them with less stringency I should get something :/
It looks like you are trying to align Oases output, which is a transcriptome assembler. Is that correct? If so, then you should be using a splice-aware aligner like TopHat or STAR.
Why BWA-SW specifically? Why not try BWA-MEM? Or a different aligner entirely?
There is no specific reason actually..
Why do you need only 70% of identity? Can you explain the application scene? Maybe I can write a simple aligner for your requirement.
Hi biomaster!
I'm doing de novo assembly of viral siRNAs. I blasted (blastn) the contigs and found that the closest virus has only 70% of identity. Now I want to map this contigs on this viral reference sequence to reconstruct my consensus, but with the default parameters of BWA I can't get any mapped contigs (too many mismatches I guess). Hope I explained my self :)