5' and 3' untranslated regions for eukaryotic parts

[fxbuson]

I'm transferring the discussion from the SBOL slack channel to here.

I'm collaborating with groups that work with plant synthetic biology, and they developed libraries of parts where 5' and 3' UTRs are key elements of their gene composition. There is currently no glyph assigned to these elements (SO:0000204 and SO:0000205), and I think diagrams for all eukaryotic systems would possibly benefit from the addition of a distinguished glyph for such regions.

iGEM's Mammalian Genetic Design page uses a trapezoid and the ribosome binding site glyph to represent the UTRs:

@jakebeal suggested two options to handle this change:

1. Let trapezoid mean both 5'UTR and 3'UTR, or
2. Add the 3'UTR meaning to ribosome entry site.

Since 5' and 3' UTRs are not interchangeable (they have different biological effects), I would avoid option 1, especially considering I'm trying to represent a modular parts library. Option 2 could be done, although I think it would decrease the specificity that the RBS glyph currently holds, and hurt the readability of past diagrams.

In case we do want a new glyph for the 5' UTR, @Gonza10V suggested we could split the trapezoid leaving the left half for 5' UTR and the right half for the 3' UTR (see below). I think this solution (or a similar one, in case we want different glyphs) is appropriate, since it keeps these elements as generic untranslated regions and implies some connection between both elements. My only worry is that the 3' half-trapezoid would then be too similar to the CDS glyph.

[bbartley]

Thanks Felipe for proposing this interesting symbol idea!What are the “biological effects” of the UTRs in these cases? Are they acting as insulators?  Are they introducing a transcriptional delay to tune the timing of gene expression?  Could we use symbols more representative of their actual functional role in the circuit to distinguish 5’ vs 3’?In the case of the split trapezoid, one minor issue is that it sort of has an implied direction, ie, it looks like it is pointing upstream against the direction of transcription.  For that reason, I slightly prefer the trapezoid, but then I don’t have a helpful suggestion for distinguishing 5’ for 3’!RegardsBryan

[jakebeal]

@bbartley In my own work with plant engineers, the 5'UTR and 3'UTR sequences tend to modulate gene expression levels up or down in a manner that is important for engineering but not yet well understood mechanistically.

One question that I have about the need to differentiate 5'UTR vs. 3'UTR is what happens when there is a polycistronic gene. Is a sequence in between the two genes a 5'UTR sequence, a 3'UTR sequence, or something else entirely?

[Gonza10V]

Well, given the variety of UTRs maybe we should go one step up in the ontology and use the trapezoid to represent UTR (SO:0000203). The tradeoff is covering more uses or being specific, and for now I would target to cover more use cases. This will also remove the directional intuiton provided by the half trapezoid.

[fxbuson]

Jake is correct. The way I understand it, 5' and 3' UTR are generic terms for untranslated regions in mature mRNA, with diverse and not completely understood mechanics.

@jakebeal normally eukaryotic genes won't be polycistronic, and that's why UTRs are separated in the two categories, although there could be an engineered polycistronic gene by use of IRES sequences. For natural prokaryotic genes, there's not normally that much space beween CDSs, but again engineered genes could have.

Even though I'd prefer having two different glyphs for my particular use case, I get that having a single glyph would make more sense for the standard. In that case, wouldn't there be a conflict with Non-Coding RNA Gene (SO:0001263 and SO:0000834)? This might be another issue, but now I'm not sure about when that glyph should/shouldn't be used.

[Gonza10V]

@fxbuson thats a good doubt because UTRs are basically non coding RNA (ncRNA), but the distinction might be in the gene part because that means that what is expressed is an RNA and not a protein therefore ncRNAs like rRNA, tRNA and miRNA goes in this category meanwhile the UTRs that are part of an mRNA not, because they are part of a gene that express a protein. Also a ncRNA gene is a gene product and a UTR is a regulatory element.