Thread: 454 assemly
View Single Post
Old 01-20-2010, 10:57 AM   #10
sklages
Senior Member
 
Location: Berlin, DE

Join Date: May 2008
Posts: 628
Default

Quote:
Originally Posted by donniemarco View Post
Hi Sven,

Thanks for your helpful reply.

If I extract all the cluster forming sequences say 213, 161 and 406 (actual sequence 214, 162, 407: thanks for pointing this as I had missed this) from the file and try to assemble them with say MIRA.
-Q int, --reindex int: reindex sequence ids.

may change the default indexing start.

Quote:
Do you not think that Mira not knowing these being one cluster will again be restricted to perform assembling the same way as these sequences would be assembled from original file (example the transitive closure property that assemblers lack).
Yes, I'd expect MIRA to assemble these reads similar as if done with the whole dataset, maybe. It's not always easy to directly compare different clustering/assembly approaches. It depends on what you are expecting.

For 454-generated EST datasets I currently use MIRA3 alone, quite successful IMHO. Datasets range from ~300,000 to 1,500.000 reads.

Quote:
I am thinking of using these cluster to generate single supercontig (and then probably rerun with my remainging singleton data with MIRA) however I am looking any particular program to do this. Any help would be appreciated.
Thanks!!
What do want to achieve? What is a 'supercontig' in this context?

A common way is to use a cluster program (wcd,tclust) and assemble these clusters with "cap3". This works quite fine. You could also use phrap.
I'd use MIRA as a "complete program to assemble ESTs".

If you want to further "join" contigs, just do so with cap3 or phrap
(although I would not do this with my datasets ;-))

Again, everthings depends on what you want / are expecting.

cheers,
Sven
sklages is offline   Reply With Quote