Marvin Sketch 5.8 import: Group attachment point issue

When opening a ChemDraw .cdx file or an ISIS/Draw .skc file containing structures with abbreviated groups, the attachment point of these groups is rendered incorrectly - it points to the center of the group, instead of pointing to the attachment atom (see attached screenshot). Are there any plans to fix this?

Regards,
Stefan

Dear Stefan,

Thank you for highlighting this problem. It is really a bug
and we are working on it.

Best Regards,

Efi

Unfortunately there's still no fix for this really disturbing display issue in the newly released Marvin 5.9 . Are there any plans to fix this at all?

Regards, Stefan

We got your request right before our commit deadline, and had no time to fix it for 5.9. Yes, we still want to fix it, hopefully we can do it for 5.10 release . Until then, you can do it manually:

1., After opening your cdx or skc file, choose menu: Edit>Source>View>"ChemAxon Marvin Documents"

2., Add leftName="EtOOC" (or reverse of the name of your group) after the title of your group (see attached picture)

3., Choose: File>Import as>"Import as Recognized(..)"

Regards,

Efi

After this correction:

Thanks Efi. Since I am using Marvin integrated in my ELN application, I don't think that it is feasible to propose my users to perform this workaround every time they paste a ChemDraw sketch. So let's hope it will be fixed in 5.10.

Regards, Stefan

Hopefully this fix still is planned for 5.10. I've meanwhile seen some bugfix releases of Marvin 5.9, also fixing import issues, but none of them addressing the current attachment bug. From an import perspective this issue, reported way back on February 2 (i.e. almost 3 month ago), is not cosmetic but in fact a show-stopper which simply will deter many users from importing at all. Therefore some priority for fixing it would be highly appreciated ...  Thanks.

Regards, Stefan    

Our core team is working on this bug; and it is still planed for 5.10.

Best Regards,

Efi

Is there any news on this?
Behaviour seems to be the same, still in 6.0.2 !

It's really annoying... 

It works for me with our test files.

Could you send me non confidential test files, please, with which I can reproduce this bug.

Thank you,

Efi

This is what I'm doing:

1.) Draw Triphenylphosphine (one phosphorous  atom with 3 phenyl systems attached to it)

2.) Create Superatom -PPh2 (select two phenyl systems and the phosphorous, goup them)

3.) Contract this Superatom --> the bond will point to the h of the abbreviation and not to the P

Hi,

First of all I have opened the file you posted here in MarvinSketch 6.0.2 and please see the results :



I think this should be the desired behaviour, you are speaking about.

On the other hand, you could spare some extra work creating a group of of PPh2 (and contracting it) because Marvin has such a predefined abbreviated group, which could be reached only by typing "pph2". :)

However if you need to create your special abbreviated groups, and you wish to differentiate how it should be connected from the left or from the right, then please see  the "User defined abbreviated groups" section in the documentation:
http://www.chemaxon.com/marvin/help/sketch/sketch-basic.html#User_defined_abbreviated_groups />and
http://www.chemaxon.com/marvin/help/formats/abbrevgroup-doc.html

I hope this helps.

viktorw

 

 

This is strange!

I reopened the file and it worked like expected. But it didn't when it was "freshly drawn" :-(.

I did some testing and had sometimes the "good" and sometimes the "bad" behaviour. Couldn't find a reason, though...It somehow depended on wether the -PPh2 was bonded to a carbon or another (oxygen, e.g.) atom....

 

Here's another example which I didn't get working - this time the problematic atom is the tin (which was also my original problem):

I have "freshly drawn" your example and edited the source to insert the "leftName=Ph3Sn" as Efi already explained here in March of 2012. (Please find below!)
And I have saved the modified mrv file, where the abbreviated groups have a defined "leftName":

I hope this helps.

Regards,
viktorw

Beeing on version 6.1.0 now, I wonder why I still do need to adjust the linking atom manually?

The information is obviously included in the drawing - expanded groups have the correct atom linked. Why is it so hard to implement a correctly functioning contraction?

Another thing arised: In the attached file, I don't know how to convince MarvinSketch to display the contracted group as Ph3P instead of PPh3. All efforts in manually editing the source file yielded in groups attached via Ph.

Hasn't there been a function to "bond from the left" or "bond from the right", resp.? I can't find it anymore...

 

Thanks!

And another issue.

If I take the before attached file and complete it with a reaction arrow, select the "Ph3P"-group and copy-paste it to appear after the reaction arrow, complete it with e.g. a double bond to give PPh3=CH2 (which includes the above problem of not resulting in the chemically correct Ph3P=CH2) and export this to PDF, I get a ugly formatting problem. The subscripted "3" looses the right spacing and gets attached to the "h".

Interestingly, this behaviour seems to depend on the order of actions. If the reaction arrow is inserted after the copy-paste+extending, formatting seems to be stable.

I attached the corresponding files for reference.

[edith] I just observed this behaviour without reaction arrows in a more complicated case. So unfortunately no easy error tracking :-( [/edith]

Dear Guntram,

 

regarding the PDF export problem, this is, indeed, a bug. We have put it on our TO DO list. Thank you for reporting it to us.

As for displaying the PPh₃ contracted group as Ph₃P, you have to define the left name of the group as Ph₃P, similarly to the previous cases. I attach the modfied version of your reaction file, where this left name has been set for both PPh3 group. Defining the left name is necessary in these cases in Marvin, and we do not plan to change this behavior.

Best regards,

Dóra

I am happy to inform you that this pdf export problem will be solved in 6.2 version.

Best Regards,

Efi