second mapping pass

[GoogleCodeExporter]

Hi,

I'm trying to figure out how to run a second mapping pass, as described in the 
paper

"It is also possible to run a second mapping pass, supplying it with splice 
junction loci found in the first mapping pass. In this case, STAR will not 
discover any new junctions but will align spliced reads with short overhangs 
across the previously detected junctions." (Dobin et al., 2012)

I didt found any examples so I don't understand how to run it...

thanks in advance


M.

Original issue reported on code.google.com by [email protected] on 25 Jun 2013 at 9:15

[GoogleCodeExporter]

Hi Mauro,

for a quicker response, please post your questions at the STAR google group 
(https://groups.google.com/forum/?fromgroups=#!forum/rna-star).

The 2-pass scheme is not automated at the moment. I have attached a simple 
script with an example of a 2-pass run.
You would need to run the 1st pass with usual parameters, than use the 
SJ.out.tab as a splice junction database file to generate a new genome index 
for STAR, and then run the 2nd pass of STAR with the new genome index. You can 
also add annotated junctions to the 1st and 2nd passes.

If you have many samples, you can collect all the novel junctions from all the 
samples (SJ.out.tab files), possibly filter them for reliability, and create 
one common set of novel junctions for all samples by merging them. Then you 
generate a new genome using annotated junctions and the common set of novel 
junctions, and re-run all the samples with this new genome - this would be the 
2-nd pass. 

Original comment by [email protected] on 6 Jul 2013 at 3:14

• Added labels: Type-Enhancement
• Removed labels: Type-Defect

Attachments:

• STAR_2pass.sh