Settings for cap3 assembly of cDNA to genomic data

[mycorr10]

Hello,
Sorry if there has been an answer to this elsewhere- i couldn't find it.
I've been given the task of extending some short ESTs with pieces of genome sequence that have been accumulated for a particularly AT-rich organism. There is no complete genome yet but my lab doesn't want to wait for the assembly-which may be six months down the line.

I've tried to align the ESTs to the genome contigs with cap3 but the assembly consistently fails to capture regions with introns properly.
I have tried to alter the -f,-r, and -h parameters but it's my first time trying cap3 and haven't gotten it right.

Can someone advise me about what are the correct parameters for this task?
thank you.

[aloliveira]

've tried to align the ESTs to the genome contigs with cap3 but the assembly consistently fails to capture regions with introns properly.I have tried to alter the -f,-r, and -h parameters but it's my first time trying cap3 and haven't gotten it right.

Cap3 is an assembler program not an aligner program. If you want an alignment to the genome using EST use the program BLAT.

[mycorr10]

I've already used blat to find the appropriate genome contigs that match the EST but it doesn't fit my needs. What i need is a program that does multiple sequence alignment capable of incorporating large gaps such as introns. BLAT does a good job of ID the right pieces but I'm trying to define the whole-length gene and therefore need a decent alignment program. CAP3 was the program suggested to me but if someone has a better suggestion I'd appreciate it.

[tbanks]

Exonerate is a great piece of software for sequence alignment. Check out the est2genome option.

[mycorr10]

Yes,
I like exonerate but since my genome isn't assembled i don't get a convenient multiple sequence alignment out of it. I often have two or three long genome contigs that align to my EST and the output of the program just gives a one to one cDNA to contig alignment. Again it may be a question of finding the right parameters cause I'm new to exonerate too but so far I haven't found a command line program that can do what a commercial MSA program like CodonCodeAligner or sequencher can do- which is a pity cause I'd love to automate this procedure...
If I'm wrong about the capabilities of exonerate I'd love some clarification.

[pkerrwall]

Did you find a solution to this problem? I need to do a similar job. Wish we knew the clustal options that the Sequencher Large Gap option uses.

[sphil]

Hey,

SSPACE is a tool which migrates cDNA (i.e. RNA-seq data) with standard DNA contigs. It uses PE as well as SE reads and gives you the chance to close gaps etc. using cDNA. Available at:

http://www.baseclear.com/landingpages/sspacev12/

[mycorr10]

Thanks I will check that out. A day or two ago someone suggested an alignment tool called Spicey and then they promptly left town. I can't find a reference or website for it but when they get back to me I'll post it.

[kmcarr]

Thanks I will check that out. A day or two ago someone suggested an alignment tool called Spicey and then they promptly left town. I can't find a reference or website for it but when they get back to me I'll post it.

They probably meant 'Spidey'

[mycorr10]

Oh that does look likely - thanks.

[mycorr10]

getting a usable alignment out of exonerate

Though I posted a separate thread for this I've gotten no response there so I'm asking here. As of now I've found exonerate to be useful in aligning ests to genomic dna and the output is supposed to be easy to parse but I'm not that great at writing my own code. Does anyone have existing scripts they'd ve willing to share to parse and combine the alignments from exonerate to create a consensus sequence?