BWA sampe: wierd pairing

[lh3]

You should draw the distribution.

[zlu]

The insert size specified was obtained from ELAND alignment. Do you think increasing the -a will help? And is there a tag in the sam file that reports pairing with wrong insert size?

[lh3]

Bwa fails to infer insert size and will use "-a" to set the maximum insert size in pairing. This may happen if too few reads are mapped or the insert size distribution is bimodal or something alike. You should check the distribution after mapping.

[zlu]

When running bwa (0.5.4) smape I got this line output to the screen over and over again:

[infer_isize] fail to infer insert size: weird pairing

Should I be worrying about this? Does it mean the pairing is not correct?

The following are the commands I used for alignment and sampe:

bwa aln -l 32 -t 2 -q 4 Genomes/Btau_UMD3.fa s_1_1_sequence.fq > Run20_s_1_1_sequence.sai & bwa aln -l 32 -t 2 -q 4 Genomes/Btau_UMD3.fa s_1_2_sequence.fq > Run20_s_1_2_sequence.sai &

bwa sampe -a 253 -o 1000 Genomes/Btau_UMD3/Btau_UMD3.fa s_1_1_sequence.sai s_1_2_sequence.sai s_1_1_sequence.fq s_1_2_sequence.fq > Run20_s_1_pe.bwa.sam

Thank you.