BWA mem on interlaced fastq

[mstagliamonte]

Hi, everybody,

I am trying to use bwa mem (v. 0.7.12) on some interleaved fastq files, i.e. the reads are paired end, but they are in one file (the ith read is forward; ith+1 is the reverse).

after indexing the reference genome, I have used the command:

bwa mem -M my_ref.fa -p myfile.interleaved.fastq > ./mapped/my_file.sam

When I read the log file though, I get this:

[M:: bwa_idx_load_from_disk] read 0 ALT contigs
[M:: process] read 117486 sequences (10000027 bp)...
[M:: process] 117486 single-end sequences; 0 paired-end sequences
[M:: process] read 118316 sequences (10000081 bp)...

and so on. Does this mean bwa is not recognizing my reads as paired ends? Have I misunderstood the manual regarding the -p option?
Looking for advice

Thanks,
MAx

[piet]

Do the ith read and the ith+1 read have EXACTLY the same name? Please note, that option '-p' is called 'smart pairing'.

[mstagliamonte]

Thank you, I did not realize it would check for names, thought it would just bluntly take the order of the reads.
The reads have a bit of an uncommon name, being:

read1.1 1
read1.2 1
read2.1 2
read2.2 2
read3.1 3
read3.2 3

I'll do a find and replace, see if it works.