PerkinElmer Informatics Support Forum
Decrease font size
Increase font size
Topic Title: When using "Copy as MOL", MOL files do not set the valence
Topic Summary: The generated MOL file in incorrect and will not be roundtripped correctly
Created On: 9/28/2017 8:03 AM
Status Post and Reply
Linear : Threading : Single : Branch
Topic Tools Topic Tools
View topic in raw text format. Print this topic.
 9/28/2017 8:03 AM
User is offline View Users Profile Print this message


noeloboyle

Posts: 2
Joined: 9/28/2017

Hi there, I just noticed that ChemDraw 16's MOL file generator is not quite correct and thought I would let you know. Basically, the writer should set the explicit valence when it is different than the MDL valence model.

Here's an example. Let's take the SMILES string, "CIC". If you "Paste as SMILES", ChemDraw inserts H3C-I-CH3 (that is, ChemDraw reads this SMILES string perfectly). It doesn't like the iodine and puts a red box around it, but it reads it in fine.

When you "Copy as MOL text", the line describing iodine is as follows:

" 0.0000 0.2062 0.0000 I 0 0 0 0 0 0 0 0 0 0 0 0"

That's not correct. It should be:

" 0.0000 0.2062 0.0000 I 0 0 0 0 0 2 0 0 0 0 0 0"

That's because the valence model for MOL files is that iodine should be three-valent (in this case), and so a hydrogen is added when ChemDraw (and other software) reads this MOL file. To override this behavior, you need to set the valence explicitly to 2.

- Noel O'Boyle (Open Babel developer, and yes, Open Babel had this problem too)

 11/21/2017 5:20 AM
User is offline View Users Profile Print this message


noeloboyle

Posts: 2
Joined: 9/28/2017

Any update on this? I've been filing bugs with a number of tools regarding (mainly SMILES) valence handling, and I would like to write up a report on the current status of these issues.

FuseTalk Basic Edition v4.0 - © 1999-2018 FuseTalk Inc. All rights reserved.