cDNA read map parameter for gsAssembler

[shaojingwang]

When I use gsAssembler from 454 to do cDNA assembling, I was confused about this parameter '-rip'.
'-rip' --> Flag to output each read in only one contig.
and this option default false.
As I known for genome assembly, one read should be assembled to only one contig.
I want to know, for cDNA assembly, reads should be assembled to multiple contigs or not? If yes, i can't understand, who can tell me why. If not, in what situation reads should be assembled to multiple contigs?
what message can we get by assembling one read to multiple contigs?

urgent for answer...

[westerman]

Interestingly, in Newbler 2.5 (recently released) there is a warning message when using both the -cdna and the -rip switch:

The -rip option has no effect for cDNA assembly projects.

After looking at the 2.5 output, it appears to be ripping reads apart. I still need to play around with the 2.5 results a bit more in order to be sure. Thus the answer to your question of "reads should be assembled to multiple contigs or not?" is, at least for Newbler, "yes". As for why, I suspect that it is because Newbler likes making Isotigs which, more or less, represent alternative splicing.

[shaojingwang]

Originally Posted by westerman

Interestingly, in Newbler 2.5 (recently released) there is a warning message when using both the -cdna and the -rip switch:



After looking at the 2.5 output, it appears to be ripping reads apart. I still need to play around with the 2.5 results a bit more in order to be sure. Thus the answer to your question of "reads should be assembled to multiple contigs or not?" is, at least for Newbler, "yes". As for why, I suspect that it is because Newbler likes making Isotigs which, more or less, represent alternative splicing.

Thanks for reply and sorry for being offline due to the time difference. I still work with Newbler 2.3, maybe I need a uprade.
I ran a set of sff file using Newbler 2.3 with '-rip' and parallel without '-rip'，then compare the result '454IsotigsLayout.txt' and '454ReadStatus.txt' file. Profound differences were found between the two '454IsotigsLayout.txt' file, I am confused by which is more close to the truth.
Another aspect, even using '-rip' option, '454ReadStatus.txt' file still contain reads assembled to multi contigs, cause difficult to get the expression level.
I'm still confused about this.

GK033JC03GCSGP Assembled contig00261 7 + contig00262 281 -
GK033JC03G0PHY Assembled contig00261 7 + contig00262 436 -
GK033JC03GNYQQ Assembled contig00261 7 + contig00262 470 -
GK033JC03G292T Assembled contig00261 1 + contig00262 434 -
GK033JC03GT171 Assembled contig00261 1 + contig00262 378 -

[flxlex]

Originally Posted by shaojingwang

I want to know, for cDNA assembly, reads should be assembled to multiple contigs or not? If yes, i can't understand, who can tell me why.

Both for genome assemblies and cDNA assemblies (and contrary to many other assemblers), newbler sometimes places parts of reads in different contigs. The reason why I try to explain in my blog on newbler:
http://contig.wordpress.com/2010/02/...newbler-works/

The -rip option has been discussed previously:
http://seqanswers.com/forums/showthread.php?t=7210
http://seqanswers.com/forums/showthread.php?t=7261

Best of luck,

flxlex

[shaojingwang]

That's just what I want to know, thank you.