search a carbon in either a 5- or 6- member ring

User 8139ea8dbd

06-11-2008 18:08:40

How to draw this query in Marvin?





I either have to modify smarts to get "r5,r6" into atom attributes (which chemists won't know that), or I can only think of using R-group, which is a very tedious way of drawing this query.





Thanks.

ChemAxon 990acf0dec

06-11-2008 20:57:34

What about using a link node?

User 8139ea8dbd

06-11-2008 21:49:26

Looking for simpler answers

ChemAxon a3d59b832c

07-11-2008 12:22:40

Hi Yingyao,





1. Draw isobutane


2. Go to the periodic system window, and click on the Query tab.


3. In the text box at the bottom (under "Custom Property"), type r5,r6 or [r5,r6]


4. The format will be automatically recognized, so the SMARTS button will be highlighted automatically.


5. Click on the appropriate atom.





If for a query property is not correctly recognized in step (4), you can still click on the SMARTS button. (E.g. R1 would create an R-group by default, but selecting SMARTS will make sure it will be the "R1" query property.)





Is this solution OK?





Regards,


Szabolcs

User 8139ea8dbd

07-11-2008 17:27:51

I did some research on ISIS Draw, it turns out Link Node is what it recommends for this situation (see picture). I think it's too tedious. I wonder if Marvin can do better than other vendors. If you can design a GUI component to perform [r4,r5], you can do the same for [R2,R3], [X2,X3], etc. It sounds like a feature general enough to worth considering.





I like the "custom property" solution, however, chemists are not "smart" enough to use smarts, even if it's very simply in our eyes.

ChemAxon 990acf0dec

07-11-2008 20:44:00

Hi Yingyao,





Would a new atom property defining the largest ring size be satisfactory for you? In this way - together with the smallest ring size - you could define the requested ranges.





Best regards,





Akos

User 8139ea8dbd

07-11-2008 21:06:44

r4 does not mean any ring of size 4 and above, it means the size of the smallest ring is 4 (an atom can be part of multiple rings). So we don't even have a way to specify the lower bound of the ring size, introducing a way to specify the upper bound of the ring size won't work.





Maybe "attribute list" is the right thing to have (we have atom list [F,Br,Cl], [r3,r5,r6] seems something similar)? Or maybe one right click on the atom and in the a menu dialog, he can check a few checkboxes on r3, r5, r6. Then it translates into [r3,r4,r5]. Just thinking out loud.

ChemAxon 990acf0dec

07-11-2008 21:20:38

OK, thank you, we will think about a solution.

User 8139ea8dbd

11-11-2008 01:00:40

Okay, after I play with it a bit more, Link Node feature does seem to be convenient enough. I am used to it already, so I guess there is no need to invent a new feature. Thanks.