5' co-ordinate lower than 3' for negative strand ORF - Glimmer

[jlpool]

I have used Glimmer to identify ORFs in my assembled bacterial contigs & after browsing through the .predict file found that the co-ordinates for some negative strand ORFs have a lower 5' than 3' value ie ....

>gi|483984440|ref|NZ_AMSJ01000050.1|
orf00005 793 5501 -2 2.94

As it is on the negative strand the 5' coordinate should be a higher number than the 3'. Am I missing something obvious?? Or is it just an ORF caller error??


Thanks for your help.

[mastal]

5'c o-ordinate lower than 3' for negative strand ORF - Glimmer

What are the column headings for the output, are they just 'start' and 'end' rather than 5' and 3'?

[jlpool]

There are no column headers, I got that information from the glimmer v3.02 release notes (extracted segment below)>>

.predict File
This file has the final gene predictions. It’s format is the fasta-header line of the sequence
followed by one line per gene. Here is a sample of the beginning of such a file:
>gms:3447|cmr:632 chromosome 1 {Mycobacterium smegmatis MC2}
orf00001 499 1692 +1 13.14
orf00004 1721 2614 +2 14.20
orf00006 2624 3778 +2 10.35
orf00009 3775 4359 +1 9.34
The columns are:
13
Column 1 The identifier of the predicted gene. The numeric portion matches the number
in the ID column of the .detail file.
Column 2 The start position of the gene.
Column 3 The end position of the gene. This is the last base of the stop codon, i.e., it
includes the stop codon.
Column 4 The reading frame.
Column 5 The per-base “raw” score of the gene. This is slightly different from the value
in the .detail file, because it includes adjustments for the PWM and startcodon
frequency.