openmolecules.org

 
Home » DataWarrior » Cheminformatics » Free Wilson Analysis
Free Wilson Analysis [message #1873] Fri, 17 March 2023 17:44 Go to next message
amorrison
Messages: 38
Registered: March 2016
Member
Hi Thomas,

I was wondering if you think it would possible to do a Free-Wilson analysis within Datawarrior? Not necessarily a feature you might want to incorporate but if you think it would be achievable and could give me some hints I'd much appreciate it.

I appreciate there are ML models already available in DW, but as a medicinal chemist it's much easier for me to relate to the Free-Wilson models and I've used them successfully in the past. I do have other tools that I can use for this but being able to do it in DW would be marvellous.

Thanks,

Angus
Re: Free Wilson Analysis [message #1876 is a reply to message #1873] Sat, 18 March 2023 18:25 Go to previous messageGo to next message
thomas is currently offline  thomas
Messages: 715
Registered: June 2014
Senior Member
Hi Angus,

this cannot be done currently, but it is not the first time to here the request. I may consider...

A way would be (similar to automatic SAR) determining core fragents, e.g. Murcko, and then for all molecules with matching core fragments analyse changing substituents. Then create a result table with these columns:

1 - core fragment with R1,r2,... for every chaging substituent position
2 - substituent position (R1 or R2 or ...)
3 - Structure of substituent at this position
4 - mean contribution of this substituent at this position to activity
5 - count of this substituent found at this position
6 - StdDev of constribution value (this substituent at this position)

Is it this that you have in mind?

Thomas

[Updated on: Sat, 18 March 2023 18:25]

Report message to a moderator

Re: Free Wilson Analysis [message #1877 is a reply to message #1876] Mon, 20 March 2023 08:41 Go to previous message
amorrison
Messages: 38
Registered: March 2016
Member
Hi Thomas,
Thanks for the response. Yes that would be perfect, maybe an option to define your own core, similar to the core based SAR analysis and the outputs you suggest would be great.

Many thanks for considering.

Angus
Previous Topic: Displaying change in r value from a selected sample of a dataset
Next Topic: peediction missing values- descriptors
Goto Forum:
  


Current Time: Thu Nov 21 13:31:32 CET 2024

Total time taken to generate the page: 0.02885 seconds