bwa sampe 0.5.7 error?

[rcorbett]

Hi all,

I have paired genomic data (75bp) that I have "aln"-ed and am attempting to "sampe"

Everthing seems to get off to a decent start, getting through a few chunks of the file before anything unusual happens. Then it appears to estimate a crazy insert size, after which nothing seems to happen. I have aligned hundreds of lanes the same way, its just this one causing me grief. Any ideas?

The end of the stderr/stdout stream looks like this:
.....
[infer_isize] (25, 50, 75) percentile: (341, 364, 385)
[infer_isize] low and high boundaries: 253 and 473 for estimating avg and std
[infer_isize] inferred external isize from 171939 pairs: 361.940 +/- 33.239
[infer_isize] skewness: -0.331; kurtosis: 0.079
[infer_isize] inferred maximum insert size: 593 (6.95 sigma)
[bwa_sai2sam_pe_core] time elapses: 16.97 sec
[bwa_sai2sam_pe_core] changing coordinates of 7533 alignments.
[bwa_sai2sam_pe_core] align unmapped mate...
[bwa_paired_sw] 30928 out of 33986 Q17 singletons are mated.
[bwa_paired_sw] 1868 out of 2024 Q17 discordant pairs are fixed.
[bwa_sai2sam_pe_core] time elapses: 10.67 sec
[bwa_sai2sam_pe_core] refine gapped alignments... 0.76 sec
[bwa_sai2sam_pe_core] print alignments... 1.72 sec
[bwa_sai2sam_pe_core] 786432 sequences have been processed.
[bwa_sai2sam_pe_core] convert to sequence coordinate...
[infer_isize] (25, 50, 75) percentile: (317166847, 804719032, 1441294659)
[infer_isize] low and high boundaries: 75 and -2147483648 for estimating avg and std
[infer_isize] inferred external isize from 139276 pairs: 816598822.655 +/- 607205032.664
[infer_isize] skewness: 0.413; kurtosis: -0.949
[infer_isize] inferred maximum insert size: -897747085 (4.25 sigma)
[bwa_sai2sam_pe_core] time elapses: 17.41 sec
[bwa_sai2sam_pe_core] changing coordinates of 4973 alignments.
[bwa_sai2sam_pe_core] align unmapped mate...

Thats as far as it gets. It just hangs after that.

[rcorbett]

For anyone who is following this.
I re-ran this lane with bwa 0.5.5 and didn't have any problems.

[pietagina]

I am experiencing the same problem with paired end Solexa data against the a yeast genome index generated with bwa-0.5.7.

[bwa_sai2sam_pe_core] convert to sequence coordinate...
[infer_isize] (25, 50, 75) percentile: (1693643, 3723845, 6454208)
[infer_isize] low and high boundaries: 51 and 15975338 for estimating avg and std
[infer_isize] inferred external isize from 57253 pairs: 4250912.238 +/- 3028173.616
[infer_isize] skewness: 0.548; kurtosis: -0.640
[infer_isize] inferred maximum insert size: 16969241 (4.20 sigma)
[bwa_sai2sam_pe_core] time elapses: 4.68 sec
[bwa_sai2sam_pe_core] changing coordinates of 11998 alignments.
[bwa_sai2sam_pe_core] align unmapped mate...