How do I go from a fasta and a chromosome to gtf/gff file?

[Brown_lineage]

Hi all,

I work on a single celled eukaryote whose genome has been sequenced. However the sequencing was not complete as there were several leftover scaffolds that contain more than a hunderd genes. Trying to get this information into cufflinks, I merged all these scaffolds toghether in one articial chromosomes with 4000N spacers in between scaffolds.

Alignement works great but I would like to use a GTF of this 'chromosome' with the original gene models to see if the gene models were substantially altered by cufflinks RABT. Is there a way where I could go from the FASTA files of the gene models to a GTF file for the 'new' chromosome? I could write a script but a pre-existing solution would be great as I am not a perl expert.

[gsgs]

I'm no Perl expert either, maybe a C-program ?

but I don't understand your language with the acronyms

[Brown_lineage]

I know even less about C than perl so writing it in C is not an option

I just think this type of program should already exist as people doing annotation and such would also use this type of tool.

[gsgs]

that's probably right and I could have the program already
or could easily create it with little changes from another program
- if only I understood correctly the details

[Brown_lineage]

Basically I need to get the orientation, start and stop of several small sequences (genes) contained in one big sequence (chromosome). I've run a BLAT to find these things but it looks like not every gene has only one hit and it does not look so trivial to find every exon-intron boundary and combine these into a transcript.
Do you have such a program gsgs?

[gsgs]

no ... but I'm thinking how to write one.
I still don't know what a GTF is.
As I understand you have one big sequence and want to separate the
exons and introns in it.
You could even align it to other existing exons or introns
but it's still difficult to find the correct areas ?
Maybe because there are gaps ?

with what did you align it ?
with a known complete genome of a similar species ?

[Brown_lineage]

Pretty standard stuff in genome annotation. Just a fixed format to describe the position of genes on a chromosome/genome

http://www.ensembl.org/info/website/upload/gff.html

[gsgs]

can I just search for long substrings without stop codon ?
6 reading frames with DNA

[Brown_lineage]

After not looking for a week I found the answer: GMAP has an option to output to GFF3 format. Just wanted to post it here in case anybody else encounters the same problem