Redundant(?) report problem in tophat .sam file?

[Gangcai]

Hi everyone,
I have checked the tophat sam file from tophat1.0.11, and found one reads have been reported twice(same genome location but different optional fields). Why two different mismatch( NM:i:0 , NM:i:1) number for the same reads and same mapping location?


HWI-EAS244_1_1_61_293_716_0_1_6480169;1 19 chr1 1231272 255 47M76N28M = 1231188 0 ACTTCTTTTCCACGTATTTGTCCTTGATCCAGGCCTCCTTGTCCTGCCGGGAGCTGCTGGCTGTGGGTTTCCTGC P\MMY]LR\`a]a``^[FWa`]a[aa\^`a`a^\``X`X```]ZZa\V\aaaaaaa__aa_ababbabbbbaab` NM:i:0 XS:A:- NS:i:0
HWI-EAS244_1_1_61_293_716_0_1_6480169;1 19 chr1 1231272 255 47M76N28M = 1231188 0 ACTTCTTTTCCACGTATTTGTCCTTGATCCAGGCCTCCTTGTCCTGCCGGGAGCTGCTGGCTGTGGGTTTCCTGC P\MMY]LR\`a]a``^[FWa`]a[aa\^`a`a^\``X`X```]ZZa\V\aaaaaaa__aa_ababbabbbbaab` NM:i:1 XS:A:- NS:i:0

[thinkRNA]

is this mouse or human? Have you tried to blat to see where it should really be landing? this is indeed puzzling and I wonder whether it is a bug in tophat. Even with bowtie's option -a, the same read should not be reported twice at the same mapping location(I think).

[Gangcai]

It's human.
And the blat result is :
ACTIONS QUERY SCORE START END QSIZE IDENTITY CHRO STRAND START END SPAN
---------------------------------------------------------------------------------------------------
browser details HWI-EAS244_1_1_61_293_716_0_1_6480169;1 74 1 75 75 100.0% 1 + 1231272 1231422 151
browser details HWI-EAS244_1_1_61_293_716_0_1_6480169;1 23 50 73 75 100.0% 2 + 220481765 220481795 31
browser details HWI-EAS244_1_1_61_293_716_0_1_6480169;1 20 48 67 75 100.0% 5 + 170003947 170003966 20

So the tophat mapping location is right. The only difference is NM report.