I looked at the input reads to MIRA. They are fastq files and the first two and the last two entries are legit. That is, matching sequence number plus forward/reverse label. Standard Illumina stuff. So those are good. Then I looked at the readpool.maf file (tail) and that looked to be matching at the end too. But then did a line on the output from miraconvert and I got 3403276. And I don’t think that can be since it’s not divisible by 8. I played with deinterleaving the file as well and it’s clear that there are 4 lines missing. Worse, they are not at the end or the beginning where it’s obvious. Attached is a snippet from a deinterleaved file that shows some context. It appears that it got out of sync right where it went from empty reads (completely clipped I assume) to reads with something left. Somehow 4 lines are being missed/added. Not sure which. This is a smallish file (300 MB) and I can upload it if that helps. I’m waiting for someone to tell me that I’m missing something obvious. It’s been a long Sunday so it’s not unlikely :-( Thanks, Torben @M02545:8:000000000-A92DB:1:1102:16316:1692/2 N + ! @M02545:8:000000000-A92DB:1:1102:11073:1698/2 N + ! @M02545:8:000000000-A92DB:1:1102:18634:1698/2 N + ! @M02545:8:000000000-A92DB:1:1102:16835:1706/1 GGCTCGGCCATAAGGAGCGAGCTTCGGAT + @:FCGGGFGCGGD<EFEFCC7@@FE+8,8 @M02545:8:000000000-A92DB:1:1102:8628:1728/1 TTAGCGGCCATTCCCCTTTTTTTTTTTTT + GGGFC++@@+CC<,;@CC,6,+8@@+++8 @M02545:8:000000000-A92DB:1:1102:20479:1729/1 > On Dec 14, 2014, at 19:16, mira_talk-bounce@xxxxxxxxxxxxx wrote: > > On 15 Dec 2014, at 5:08 , Torben Nielsen <torben@xxxxxxxxxx> wrote: >> Mira 4.9.3 >> Has anyone else seen cases where >> miraconvert -C readpool.maf reads.fastq >> appears to drop one or more reads? I have seen few cases now where one of a >> pair appears to have gone missing? > > Are both reads of the pair present in the MAF in the first place? > > B. > > > -- > You have received this mail because you are subscribed to the mira_talk > mailing list. For information on how to subscribe or unsubscribe, please > visit http://www.chevreux.org/mira_mailinglists.html -- You have received this mail because you are subscribed to the mira_talk mailing list. For information on how to subscribe or unsubscribe, please visit http://www.chevreux.org/mira_mailinglists.html