[maker-devel] Question about phase for CDS with start codon

[Carson Holt]

I know there was an incorrect phase issue on a previous maker version that
is now fixed, but I really doubt that is the issue causing your error.
What are you using to convert from GFF3 to tbl format before using
tbl2asn?  I'd start there.

we can send you a GFF3 to tbl converter if that will help.

--Carson




On 9/11/13 4:23 PM, "Brian Hall" <bhall7 at hawaii.edu> wrote:

>Aloha,
>
>I'm working with a gff produced by maker. (I didn't run the program
>myself, but I believe it was version 2.24.) Here are the lines in
>question:
>
>scaffold00033    maker    CDS    729494    729949    .    -    2
>ID=107343;Name=BDOR_005037-RC:cds:250;Parent=107334
>scaffold00033    maker    start_codon    729947    729949    . -    .
>ID=107349;Name=BDOR_005037-RB:start1;Parent=107334
>
>If I understand correctly, the start codon in this reverse-strand CDS is
>from position 729949 to 729947 -- the first three bases in the CDS.
>However, the phase value for the CDS is 2, which essentially skips the
>start codon. Downstream software (tbl2asn) is kicking up a "missing
>start codon" error.
>
>I have several hundred such issues in the gff for a single genome. They
>generally only occur on reverse-strand CDSs. Any ideas?
>
>Sincerest apologies if this is a duplicate question or if I've provided
>incomplete information. I am new at this. Thanks for your help!
>
>--Brian
>
>_______________________________________________
>maker-devel mailing list
>maker-devel at box290.bluehost.com
>http://box290.bluehost.com/mailman/listinfo/maker-devel_yandell-lab.org