problem with BWA alignment

Question

[blur]

Hi,

I am trying to align PE reads using bwa and something isn't working... I tried bwa mem ref.fa read1.fq read2.fq > file.sam but I get an error

[M::mem_pestat] skip orientation FF as there are not enough pairs
[M::mem_pestat] skip orientation FR as there are not enough pairs
[M::mem_pestat] skip orientation RF as there are not enough pairs
[M::mem_pestat] skip orientation RR as there are not enough pairs

I also tried using SAMPE: i.e. I created two seperate sai files and ran

bwa sampe ref.fa read1.sai read2.sai read1.fq read2.fq > file.sam

and the file I end up with has like two reads in it...

can someone point out what I am doing wrong? Thanks!

[agata88]

This is not an error, it's an information. Convert created SAM file to BAM file, sort it with samtools and check alignments statistics by samtools flagstat. You will see then how many reads were aligned etc. Best, Agata

[2nelly]

No worries.

check this out: Error while running BWA mem

[Brian Bushnell]

It's possible that your read pairing is broken, or that you're using the wrong reference so nothing is aligning, or the quality is too low for anything to align. Lots of possibilities - I suggest you start by stating what organism you are working with and BLASTing or Sketching the reads to see what they are, and running FastQC and posting the results. Also, the first 10 read headers from each file would be helpful, as well as any information you have about what platform they came from and how they have been preprocessed.

[WouterDeCoster]

I added markup to your post for increased readability. You can do this by selecting the text and clicking the 101010 button. When you compose or edit a post that button is in your toolbar, see image below:

Answer 1

[shwethacm]

What you are seeing is not really an error. For some reason, most of your reads are aligning very poorly to your reference. Are you aligning to the wrong reference? This could result in the kind of results you are seeing. Did you preprocess your fastq files (trim adapters, remove linkers?). This could sometimes mess with alignments.

[blur]

Thanks for all the help - I think the clean-up process I am using is what's ruining my pairs... It a specific protocol that has a collapse for identical reads, and I think this is how pairs get lost (i.e. if I have 10 reads identical in each fastq file, the ones which survive are not the original pairs) If I were to align each read separately and than put them into a single file, will I be loosing too much info? each read is 100bp. (btw the reference genome is correct (I double checked).)

Thanks!