[mira_talk] Singlet output from MIRA
- From: "Kvist, Sebastian" <skvist@xxxxxxxxxxxxxxx>
- To: "mira_talk@xxxxxxxxxxxxx" <mira_talk@xxxxxxxxxxxxx>
- Date: Fri, 25 Jan 2013 16:42:26 -0500
Dear Bastien,
Quick question (I hope) about the assembly process. I want to run four
iterations of MIRA3 on my 454 data and when I do so, I'm encountering some
issues:
The data set is cDNA and I used the following command to call MIRA: mira
--project=HT8UG4B01 --job=denovo,est,accurate,454 &
I started of with about 120,000 frags (trimmed and quality controlled), the
first pass with MIRA gives me ~12,000 contigs and the "info_assembly.txt" file
tells me that NO singlets remained after assembly. After the second pass, I am
left with ~600 contigs and, again, the info file tells me no singlets are
present. After the third and fourth passes, I'm left with 16 contigs (again, no
singlets). The N50 increases for each pass, which I'm happy about but when I
look at the 16 contigs there must be a lot of data that has been discarded
somewhere. At least I hope that we haven't effectively sequenced only 16 highly
expressed genes.
I was wondering if the singlets are contained within another file and that the
info file is telling me the wrong thing? I checked the "info_debrislist_pass.1
[…] 5" and it seems as though the singlet names are in there but not the actual
sequences. I could "grep -f" from the original file using the debris file as a
guide but I was wondering if there's an easier way to get to the singlet
sequences?
Also, I was wondering if you have any opinion on performing several iterations
of MIRA on the same data, without loosing any potential singlets (because the
"singlets" in the second pass are effectively contigs from the first pass) and
if you have any ideas on how to do it most effectively?
Sorry about the length of the e-mail for such a trivial question but I thought
that it might be best to give the full picture.
Thanks a lot.
Best,
Sebastian Kvist
Other related posts:
