Partial Cds Extraction From Incomplete Transcripts

Question

[Raghul]

Hi to all, I have 454 RNA transcripts assembled into Isotigs using Newbler. They are complete, they miss 5' end, they miss 3' end and sometimes they miss both (as revealed by blastx). I want only the CDS (coding sequence) because I want to check the propensity of GC (only in CDS) & hydrophilic amino acids (after translation) in my transcriptome. So do you guys know any tool which does this? So all kind of suggestions are welcome!

thank u raghul

[Neilfws]

Just to note that by definition, an incomplete transcript cannot yield a CDS. I suppose it can yield a "partial CDS".

[Raghul]

I am having the assembled transcripts.

[Raghul]

Thanks to all for answering the questions. I think for protein sequence analysis, I can use OrfPredictor. It takes sequences & blastx output for analysis & give predicted peptide sequences in the FASTA formatI am looking ways to identify (partial)CDS in RNA. This answer was available for another question in Biostar forum for a different question.

[Raghul]

Thanks to all for answering the questions. I think for protein sequence analysis, I can use OrfPredictor. It takes sequences & blastx output for analysis & give predicted peptide sequences in the FASTA format. Now I am looking ways to identify (partial)CDS in RNA. This answer was available for another question in Biostar forum for a different question.

Answer 1

[Damian Kao]

You'll probably want to find the longest ORF and take that as the CDS. You can do a six frame translation and find the longest ORF. Since you might be missing 5' or 3' or both, remember to account for missing stop or start codons.

[Raghul]

The problem here is even after assembled into contigs, many are missing 3 or 5 ends. But still such analysis can be done because no other information is available for this protozoa.

[Michael]

raghul, I think there is not much more you can do about ORFs. use getORFs to find the longest ORF. If the stop codon is missing you will not get the right frame in any case using getORF.

Answer 2

[Michael]

The standard approach for this would be to run a gene prediction software like Glimmer, GeneMark, or Critica on your contigs. Also, try to run a tBLASTx search.

Answer 3

[Jack Min]

You can write a perl script to extract cds region from your input sequences using the information produced by OrfPredictor as in the definition lines of the output containing the frame value, the start and the end position of the cds.

[Jack Min]

I added a function to the OrfPredictor server to generate the CDS file. Please try it out.
Jack Min

[Jack Min]

the server is available at http://proteomics.ysu.edu/tools/OrfPredictor.html

Answer 4

[Chris Miller]

Are these already aligned? If, so you could use bedtools to intersect a bed file containing your coding regions with your bam file:

intersectBed -abam yourbamfile.bam -b cdsfile.bed

[Raghul]

Yes it is already assembled into contigs & it is a protozoa. It has no complete genome sequence available.