pushing electrons in MarvinSketch

We had one organic chemistry professor who insisted that reaction mechanisms are based on pushing electrons around. (Philosophically, I'd have to agree with that.)

What he vehemently objected to is that MarvinSketch originates and ends electron-flow arrows on atoms or bonds, which means that you cannot "grab" a lone pair and "push" it around into, e.g., a bond.

I think, conceptually and maybe didactically, he has a point.

Is there any way to change the way MarvinSketch behaves in this respect? If we cannot actually change the behavior of MarvinSketch, can we at least emulate what this professor is asking the application to do: start an electron-flow arrow on a lone pair (or conversely: end an electron-flow arrow on a lone pair)?

Thanks.

 

Hi Michiel,

It is an important missing feature in Marvin, and we intend to implement it, but to be able to do that we have to redesign the way of displaying lone pairs: they have to be graphical objects.

This feature request is already on our list, just had no capacity to implement it yet. Hopefully we will implement it in 5.4 (coming next spring).

Best regards,

Akos

It would be nice for the arrow to start at a lone pair, but it is by no means necessary.  The professor who "vehemently" insists that the arrow must start at a lone pair, not at the elemental symbol, is confusing convention with reality.  

Suggest to him that the base and head of the arrow point to the location of the electrons.  Thus, it is perfectly appropriate for the arrow to begin at the elemental symbol, because that is where the electrons reside.  

Does he also object to the widely used convention wherein the base of the arrow often begins at a negative sign?  

akospapp wrote:

It is an important missing feature in Marvin, and we intend to implement it, but to be able to do that we have to redesign the way of displaying lone pairs: they have to be graphical objects.

I'm not sure you do.  You could just have a new "rule": If an electron-flow arrow begins at an element, then draw in a lone pair for that element and have the arrow appear to originate at the lone pair.  If lone pairs are already being displayed, you have to preserve the total number of lone pairs.

You don't want the internal MRV code to differentiate between arrows originating at elements and arrows originating at lone pairs associated with elements.  That way lies chaos.  

bobgr wrote:

It would be nice for the arrow to start at a lone pair, but it is by no means necessary.  The professor who "vehemently" insists that the arrow must start at a lone pair, not at the elemental symbol, is confusing convention with reality.   Suggest to him that the base and head of the arrow point to the location of the electrons.  Thus, it is perfectly appropriate for the arrow to begin at the elemental symbol, because that is where the electrons reside.   Does he also object to the widely used convention wherein the base of the arrow often begins at a negative sign?

I don't know if you've ever tried to convince an organic chemistry professor that there's more than one way to push electrons around, but I wouldn't generally recommend starting that discussion...

My overriding concern is that it has to be conceptually intuitive for a student to push electrons around that is consistent with what is being taught. Electrons may reside on atoms, but that doesn't make it intuitive to "grab" the atom core itself when you want to push an electron (pair) around.

As for your last comment: I didn't discuss that with him, nor do I intend to. As far as the logic is concerned: it makes sense to "grab" a "negative charge" (representing an excess of electrons) and "push" them towards an area that is electron deficient. I'm not saying this representation is right or wrong, but from a student's perspective this can be readily visualized and understood.

bobgr wrote:

akospapp wrote: It is an important missing feature in Marvin, and we intend to implement it, but to be able to do that we have to redesign the way of displaying lone pairs: they have to be graphical objects. I'm not sure you do.  You could just have a new "rule": If an electron-flow arrow begins at an element, then draw in a lone pair for that element and have the arrow appear to originate at the lone pair.  If lone pairs are already being displayed, you have to preserve the total number of lone pairs. You don't want the internal MRV code to differentiate between arrows originating at elements and arrows originating at lone pairs associated with elements.  That way lies chaos.

I didn't mean to imply that the MRV code has to be rewritten or layered with an extension.

My main concern is that a student needs to be able to "push electrons around" since that is the basis for teaching reactions mechanisms in organic chemistry. In it's current implementation you have to select/"grab" the atom core in order to move electrons from it to another location.

I think I can convince a student that that's how they can get a correct score for their homework/test assignment, but so far I've been unable to convince a professor that it's didactically correct. If bait-and-switch will work with lone pair and radical electrons, that's perfectly fine by me.

I didn't suggest that you suggested that the code be rewritten.  I was simply suggesting to ChemAxon that they could handle your request without changing how they represent lone pairs.

I *am* an organic chemistry professor, I have written a book on drawing mechanisms that has been very widely adopted, and I understand that there is more than one way to draw electron-flow arrows.  I agree with your colleague that it would be pedagogically preferable to originate the arrow at a lone pair instead of at an atom.  I was objecting only to the claim that originating it at the atom is somehow fundamentally *wrong*.  To suggest that this is the case is to confuse the conventions of drawing dots, lines, and arrows with the reality of chemical change.  And you can tell him I said so.  Not that it will make a difference.

bobgr wrote:

I didn't suggest that you suggested that the code be rewritten.  I was simply suggesting to ChemAxon that they could handle your request without changing how they represent lone pairs. I *am* an organic chemistry professor, I have written a book on drawing mechanisms that has been very widely adopted, and I understand that there is more than one way to draw electron-flow arrows.  I agree with your colleague that it would be pedagogically preferable to originate the arrow at a lone pair instead of at an atom.  I was objecting only to the claim that originating it at the atom is somehow fundamentally *wrong*.  To suggest that this is the case is to confuse the conventions of drawing dots, lines, and arrows with the reality of chemical change.  And you can tell him I said so.  Not that it will make a difference.

I don't think telling him you said so is going to earn me any brownie points, so I'll refrain from it, but thanks for sticking to your guns.

I'm a medicinal chemist who suddenly finds himself working with academics so I take push-back from professors very seriously: It helps to keep me honest and ask for improvements where possible. Working in industry with trained chemists has rather spoiled me in terms of the finer interpretations of our art.

 

Thanks Bob for trying to ease our life; your suggestions are very useful for those who would like to represent lone pair transfers in the present form of MarvinSketch.

On the other hand, I completely understand the request from Michiel to enable electron-flow arrows to start on lone pairs. For us the easiest way to implement this is to change the lone pairs to be graphical objects. It will be useful for other reasons, too, e.g. after displaying the lone pairs and automatically arranging them (also when the bonds are rearranged) the user can move them to a better position if he/she doesn't like the automatic arrangement. So i think we have to do this anyway. But as I mentioned, we have no capacity to do this before 5.4.

 

Best regards,

Akos