Do you have any more examples outside of RHEA? I think they just didn’t have the proper arrow types and it does not seem correct to me

How many of these (#/%) are these in Rhea?
Can you also give more context on "net reaction" it's not something I'm familiar with.
I somewhat feel like perhaps a single line might be better, or make the '=' shorter by default. Note Rhea use "=>" which would normally be retrosynthetic.

Orthogonal to this we probably need a BIDIRECTIONAL_UNBALANCED, MULTISTEP. The second doesn't really fit as a "direction" but would be useful.

For reference here are ChemDraw's arrows:

Hello,

In Rhea, there are separate forward, backward, bidirectional and undirected reactions for each transformation. So, for ex:

https://www.rhea-db.org/rhea/23884 (undirected)
https://www.rhea-db.org/rhea/23885 (forward)
https://www.rhea-db.org/rhea/23886 (backward)
https://www.rhea-db.org/rhea/23887 (bi)

All these talk about the same transformation but directions are different.

• I also prefer a shorter "=". Probably can play with scale to render an acceptable length with CDK.
• I agree => should be replaced with -> in Rhea and that's why looking into CDK for a better visualization.
• As of now, Rhea only uses => <= = <=>.
• To be clear, the aim here is to only introduce "=" sign for undirected reactions only.
• As per Net reaction, I think a multi-step reaction is cumulatively written with an equal to sign in Literature. So there is a precedence to use "=" in writing chemical reactions.
  https://chemistry.stackexchange.com/questions/87086/what-does-an-equal-sign-mean-in-chemical-equations

However, it is useful in kinetics to distinguish this from a net reaction, which is written either with two one-sided arrows or an "equal" sign.

I think the two one sided arrow (EQUILIBRIUM) is now preferred there. (I'm on the working group for the new IUPAC reaction standards).

If you make it a bit smaller (i.e. like an equal sign) then I will merge. You should be able to do this by just not drawing the lines to full length.

Hi John,

neither Parit nor I are chemists, so we may not always use the correct terminology, sorry for that. This isn't about step reactions, overall/net reactions, etc.

Let me try to explain where we are coming from: We are both working on the Rhea and UniProt databases. The Rhea database was originally created to structure the textual representations of the reactions described by the IUBMB (https://iubmb.qmul.ac.uk/), here is an example where you can see that the IUBMB uses the '=' sign: https://iubmb.qmul.ac.uk/enzyme/EC1/1/1/1.html

We have used these IUBMB text representations for many years to annotate enzymes in UniProt, until in 2018 we replaced everything we could with reactions from the Rhea database: https://www.uniprot.org/release-notes/2018-12-05-release. Our interpretation of '=' is not equilibrium, because we do not make a statement about kinetics, we merely want to say that enzyme X catalyzes A + B = C. When we have experimental evidence for the net flow in vivo, we will add that the enzyme catalyzes the forward ('->') or reverse ('<-') direction (or both depending on conditions), but this information is not always available. So for annotating enzymes in UniProt we need '=', '->' and '<-'. For historic reasons unknown to us, Rhea has also what we call bi-directional, but we do not use these in UniProt.

Rhea has currently ~15,000 reactions (https://www.rhea-db.org/) and these are used to annotate enzymes in over 26 mio UniProt records: https://www.uniprot.org/uniprotkb?query=(cc_catalytic_activity:%22rhea:*%22). Look for "Catalytic activity" in https://www.uniprot.org/uniprotkb/Q9BV79/entry to see how this currently loooks. Parit is trying to use CDK to improve the current visualization of these reactions.

I hope this helps to understand our use case.

Regards,
Nicole

Yes I understand, I did my PhD in ChEBI :-). I was mere commenting on the linked page you have the following which shows an equilibrium arrow and '=' used interchangeably:

• https://chemistry.stackexchange.com/questions/87086/what-does-an-equal-sign-mean-in-chemical-equations

I understand sometimes you don't know which direction it goes and that's useful for Rhea to differentiate that.

Like I said - make it smaller/prettier and I'll merge.

Yes I understand, I did my PhD in ChEBI :-)

I thought I had heard your name before :-)

Like I said - make it smaller/prettier and I'll merge.

OK, got it. Thanks!

@johnmay Hi, Sorry for the delay. I made some corrections. Please look at the generated image.

No worries, looks good. A lot changed recently in the ReactionDepiction so I need to rebase. Will try and find time today.

Alas the recent changes to the reaction drawing to support multi-step depictions were too much to rebase. I've redone the changes on the current main in #1009 and also tweaked the method a little to be a bit cleaner (avoiding the 0.08 fudge that was in there for sizing).

Done in #1009