So I was wondering, there are a lot of cheminformatics tools (eg Babel) out there, to analyze and convert representations. I also saw some codes which counts certain groups, but ignores the carbon chain, which is obviously not suited for a Joback script. My google research only turned me a "homemade" matlab code found at Reddit, but it is no way general, as the first step of it is to exclude most of the rings.

So is there any open cheminformatic tool for this problem?

  • $\begingroup$ It might be helpful to add some details to your question about what a 'Joback group' is - it seems like en.wikipedia.org/wiki/Joback_method is relevant? In any case, you are probably looking for a /descriptor/ which is simply a value calculated for a structure. $\endgroup$ – gilleain Mar 7 '16 at 11:57
  • $\begingroup$ Yes, by Joback grousps I mean the groups used in Joback's method, but any other group contribution method would do. As to your comment, I am not looking for a calculated descriptor, but for a list of the groups that the molecule contains. $\endgroup$ – Ezze Mar 7 '16 at 12:02
  • $\begingroup$ Right, great - maybe your question could be "is there an algorithm for calculating group contributions" or something. In any case, I would expect the first step for any Jobeck descriptor would be to find the groups. Looks like some kind of simple substructure search would do it. $\endgroup$ – gilleain Mar 7 '16 at 12:05

So it looks like 'Jobeck groups' are some set of functional groups that you want to find in molecules.

Sounds a lot like substructure search, for which there are many cheminformatics tools (like SMSD, for example) however, since you mention OpenBabel:


talks specifically about 'group contribution methods' and gives a simple example of how to get the counts for a set of functional groups.

| improve this answer | |

Your Answer

By clicking “Post Your Answer”, you agree to our terms of service, privacy policy and cookie policy

Not the answer you're looking for? Browse other questions tagged or ask your own question.