Letter code behaviour of agat_sp_manage_IDs.pl

[SwiftSeal]

I've been trying to rename sequences in a gff file with a prefix with:

agat_sp_manage_IDs.pl --gff gene.gff --prefix cpc54. -o fixed.gff

The result looks like this:

##gff-version 3
chr01   AUGUSTUS        gene    33168   48785   .       -       .       ID=g1
chr01   gmst    mRNA    33168   44206   .       -       .       ID=cpc54.1;Parent=g1
chr01   gmst    exon    33168   33779   312.083742      -       0       ID=cpc54.2;Parent=cpc54.1
chr01   gmst    exon    34560   35239   312.083742      -       2       ID=cpc54.3;Parent=cpc54.1
chr01   gmst    exon    35648   36008   312.083742      -       0       ID=cpc54.4;Parent=cpc54.1
chr01   gmst    exon    36467   36788   312.083742      -       1       ID=cpc54.5;Parent=cpc54.1
chr01   gmst    exon    36883   37385   312.083742      -       0       ID=cpc54.6;Parent=cpc54.1
chr01   gmst    exon    37707   37980   312.083742      -       1       ID=cpc54.7;Parent=cpc54.1
chr01   gmst    exon    38252   38541   312.083742      -       0       ID=cpc54.8;Parent=cpc54.1
chr01   gmst    exon    39456   39878   312.083742      -       0       ID=cpc54.9;Parent=cpc54.1
chr01   gmst    exon    39990   40314   312.083742      -       1       ID=cpc54.10;Parent=cpc54.1
chr01   gmst    exon    43953   44206   312.083742      -       0       ID=cpc54.11;Parent=cpc54.1
chr01   gmst    CDS     33168   33779   312.083742      -       0       ID=cpc54.12;Parent=cpc54.1
chr01   gmst    CDS     34560   35239   312.083742      -       2       ID=cpc54.13;Parent=cpc54.1
chr01   gmst    CDS     35648   36008   312.083742      -       0       ID=cpc54.14;Parent=cpc54.1
chr01   gmst    CDS     36467   36788   312.083742      -       1       ID=cpc54.15;Parent=cpc54.1

Is this normal behaviour? It's not clear to me why the gene IDs have not changed, and why each feature is essentially getting its own number. Ideally I'd like to have cpc54.1.t1 cpc54.1.t2 or similar for each transcript. With the current output. it's causing a lot of problems downstream with software that tries to map transcripts to genes.

Cheers!

[Juke34]

Interesting, I definitely need to look into that.

[SwiftSeal]

I kept trying and eventually found that cpc54_ works, but cpc. doesn't! Maybe some underlying string parsing?

[Juke34]

What version of AGAT do you use?
I gave a try using the last one and I do not have any issue with "cpc."