[mira_talk] repeat clusters
- From: bio5yz <bio5yz@xxxxxxxxxxxxxx>
- To: mira_talk@xxxxxxxxxxxxx
- Date: Mon, 7 Jun 2010 00:26:34 -0600
Hi all,
I have a 280k 454 titanium readset run with MIRA ~ pre-quality, vector,
polyA clipped. There is already external pairwise alignment information (90%
identity, min 60 bp hit length) for this set and a list of potentially
'problem' reads that have large number of alignment hits.
One particular read of interest consists of 5171 hits in the 5' region and
separately 7138 on the 3'. Examining MIRA output, I found this read to be
clustered to a 'rep_c' cluster of only 2 members. Tracing through the logs
show that it in fact only has one hit in the 'int_posmatchc_pass.1.lst'
files and no mention in the 'repeat_resolve.1' files.
Would anyone know if there is a additional pass I should be tracing? I
understand that this is labeled as a large repeat cluster and was wondering
how it is separated only to 2 members and where the repetitive information
may be stored.
MIRA 3.0.5 command:
mira --project=SETTMV --job=denovo,est,normal,454 COMMON_SETTINGS
-SK:not=10:pr=92:mchr=4096 454_SETTINGS -LR:mxti=no:ft=fastq
-CL:qc=no:cpat=no:lcc=no -AS:mrl=80 -AL:bip=19 > log_assembly.txt
Thanks very much,
Michael.
Other related posts:
