Skip to content
Permalink
Browse files

Update gff3.md

  • Loading branch information...
barrymoore committed Mar 14, 2018
1 parent da5cfd1 commit 3dc0607dc6d6843f468074d4edfaf9a278b6f9e8
Showing with 2 additions and 2 deletions.
  1. +2 −2 gff3.md
@@ -682,8 +682,8 @@ chrX . gene XXXX YYYY . + . ID=gene01;name=my_gene
chrX . mRNA XXXX YYYY . + . ID=tran01;Parent=gene01;Ontology_term=SO:1000069
chrX . exon XXXX YYYY . + . Parent=tran01
chrX . CDS XXXX YYYY . + 0 ID=cds01;Parent=tran01
chrX . CDS YYYY-1 ZZZZ . + 1 ID=cds01;Parent=tran01</pre>
<p>You will also need to adjust the phase field properly so that the CDS translates.</p>
chrX . CDS YYYY-1 ZZZZ . + 0 ID=cds01;Parent=tran01</pre>
<p>The CDS segment that represent the new reading frame will always has a phase of 0 since the ribosome is moving and thus redfining the codon.</p>
<p>It is suggested that the mRNA be tagged with the appropriate SO transcript attributes such as "minus_1_translational_frameshift" (SO:1000069). This will allow all such programmed frameshift mRNAs to be recovered with a query. The accession for "plus_1_translational_frameshift" is SO:1001263.</p>
</dd>
<dt>An operon</dt>

5 comments on commit 3dc0607

@tmgreen

This comment has been minimized.

Copy link

tmgreen replied Mar 15, 2018

Thanks @barrymoore that looks good to me!

@tmgreen

This comment has been minimized.

Copy link

tmgreen replied Mar 15, 2018

Sorry, spoke too soon, I don't think YYYY-1 is right here. First, it implies that it is the same YYYY from the previous line, but that is in general not a safe assumption in the examples, and second, at least from what I've seen in the wild, this value would tend to always be either YYYY or YYYY+2 for a one base shift in either direction. I'm not sure if two-base shifts exist, and that's what YYYY-1 would imply to me. I will say that many times while reading the spec it would have helped me greatly to have some real numbers in place of XXXX and YYYY so I could figure stuff out or check assumptions.

@barrymoore

This comment has been minimized.

Copy link
Contributor Author

barrymoore replied Mar 15, 2018

Well, and then there's my horrible typographic errors - this is what I get for thinking I could get by with a quick fix here! The YYYY-1 works with the way that I'm thinking about this, but I agree that the overall example doesn't really work and a better example that is based an an actual frameshift is needed here. I'll try to scratch together a better example and see if we can get this cleaned up right. One caveat for me right now is that I have several pending deadlines that are going to make me a bit slow on addressing this. I'll fix the typos now, but there is no question that the overall example does need more attention.

@keilbeck

This comment has been minimized.

Copy link

keilbeck replied Mar 15, 2018

Barry - likewise I am dealing with deadlines but suggest that we get together with a 2 large cups of coffee and a white board and hash this out next week. Thanks.

@barrymoore

This comment has been minimized.

Copy link
Contributor Author

barrymoore replied Mar 15, 2018

Please sign in to comment.
You can’t perform that action at this time.