partially resolved structures

Dear,

 

i was wondering if you can add partially resolved structures, meaning i have the structure for a part of the molecule, but for the remaining part i only have the brutoformula. I would like to attach that brutoformula to the molecule so the total MW for the molecule would be calculated. An example, consider i have caffeic acid hexoside, i know the structure for caffeic acid, but i don't know which hexoside is on it, or the place it is attached. Is it possible to attach to the molecule C6H10O5 for example so that the software recognizes the C H and O? 

This is a common issue when resolving structures through mass spec, but you can only add the molecule to the library once you have fully identified it, but very often we can only do this partially, meaning we can never add these to the library. 

 

Is there any solution for this?

 

Thanks,

 

Cheers, Geert

it does work in ACD/chemsketch

in attachment a picture of what i mean :) 

 

just to be clear :)

Hi Geert,

partially defined structures are not handled in Marvin, we only support fully drawn structures and mass calculation from formula in Chemical Terms, but not the mixed representation. 

Best regards,

Peter

Dear Peter,

 

Thanks for the reply, and yes, i know it's not supported, that's why i was asking if it would be possible to do that, i think it would be really useful for many people who deal with the same problem. Is this something that could be developed in the future?

Or another possibility to define the unresolved part of the structure as an R group?

Best Regards, Geert

Dear,

 

I'm really in need of some way to organize partially resolved structures in a database, is there no one that has any suggestion or solution for this?

Cheers, Geert

when you insert a group (for example PO(OMe)2 ) which is a template in Marvin sketch it does recognize the elements and it can calculate the MW, can I make such groups myself?

 

cheers, Geert

Yes, you can create custom abbreviations and custom templates too. 

ElementalAnalyser can calculate with them, but they will have a well-defined  structure behind the abbreviation.

Peter

Ok, i see, so there will be no option to do what i want? you won't consider that option to be included in an update?

 

cheers, geert

is there a way that i can just enter a brutoformula then in the table itself?

cause now it's automatically calculated from the structure..

 

cheers, geert

We're investigating how we could support this feature and how it would affect our search, registration system and other calculations.

I don't think there is a way to change the "Formula" column in IJC, but I have to ask around to get your answer.

BRs

Peter

Dear Geert,

the formula column in IJC can not be changed or modified. It is added automatically to structure tables with "Molecules" or "Any Structures" table types and the values are calculated from the structure field. It is a Chemical Terms Fields.

Regarding the possibilities to insert the bruto formula directly into the table, IJC is limited by the functionality of Marvin, as IJC relies on MarvinSketch as a component to insert molecules into the table, so if you cant use partially defined structures in Marvin, it will be impossible in IJC either.

What you however could do as a partial workaround, is to create 2 "Any Structure" type tables, one containing the structure you described via MS, and the other containing the "unknown" structure part, while on both defining the attachment points. Then, creating a relationship between those 2 tables over some unique ID field and add a Calculated field that would sum Mol Weight fields from both tables.

This solution will however still be limited by Marvin's structural capabilities. And also please note that the molecular weight chemical terms field calculates only with Structure field and cant be used on Formula Field (which is another Chemical Terms field)

Bests

David

Geert, if the only issue is that you need the group to calculate a correct molecular weight, then you can create your own shortcut group for, say, C₆H₁₁O₅, and define it with *one example* of that group, say, OCC1OCC(O)C(O)C1O with the attachment point at atom 5. 

This workaround has implications for database searches, but if you can tolerate false positives or false negatives, it may suit your purposes.