openmolecules.org Forum: Functionality » Similarity analysis using "find similar compounds..."

Home » DataWarrior » Functionality » Similarity analysis using "find similar compounds..." - slow analysis of libraries (Similarity analysis)

Show: Today's Messages :: Polls :: Message Navigator

Fri, 27 November 2020 01:35

SM2020
Messages: 13
Registered: June 2020
Location: UK

Junior Member

Thanks very much Thomas!! - that will be a really useful update. Looking forward to that rolling out soon.

Getting a distribution of molecules within a single file would be great, but comparing 2 discrete files - and plotting a resulting similarity distribution - is more what I was thinking about. In my mind they are slightly different (but I could be mistaken).

Maybe I'm misunderstanding, but using your suggestion, could I just put all the cpds into a single file, run the analysis you suggest and then (assuming there is a unique identifier for the cpds - e.g. vendor), use DW to highlight by vendor and then assess the distribution in the resulting histogram? Would this compound selection step be possible (like in other graphical displays in DW)?

I was essentially thinking to keep one library constant (reference) and compare against "other" libraries thereby generating a binned histogram of closet neighbour similarities for each "other" library compared to the reference - one measure of similarity between "other" libraries and a reference library is then gauged by the apparent shift of distributions between (near) 0 and 1.

Either way, I'd like to try both analyses. It's great that DW (will) allow it.

Is there a hard upper limit for the no. of compounds analysed in these ways (i.e. file size, > 16K cpds?? Wink

or is it driven by available computational resources to carry out the task?

Thanks again for your effort.

Report message to a moderator

[Message index]

		Similarity analysis using "find similar compounds..." - slow analysis of libraries By: SM2020 on Tue, 24 November 2020 12:35
		Re: Similarity analysis using "find similar compounds..." - slow analysis of libraries By: SM2020 on Tue, 24 November 2020 21:33
		Re: Similarity analysis using "find similar compounds..." - slow analysis of libraries By: thomas on Thu, 26 November 2020 18:16
		Re: Similarity analysis using "find similar compounds..." - slow analysis of libraries By: SM2020 on Fri, 27 November 2020 01:35
		Re: Similarity analysis using "find similar compounds..." - slow analysis of libraries By: thomas on Sun, 29 November 2020 00:05
		Re: Similarity analysis using "find similar compounds..." - slow analysis of libraries By: SM2020 on Sun, 29 November 2020 15:07
		Re: Similarity analysis using "find similar compounds..." - slow analysis of libraries By: SM2020 on Thu, 04 February 2021 11:11
		Re: Similarity analysis using "find similar compounds..." - slow analysis of libraries By: thomas on Thu, 11 February 2021 10:55
		Re: Similarity analysis using "find similar compounds..." - slow analysis of libraries By: SM2020 on Mon, 01 March 2021 01:00
		Re: Similarity analysis using "find similar compounds..." - slow analysis of libraries By: SM2020 on Tue, 02 March 2021 23:40

Previous Topic:	Confidentiality of compound structures
Next Topic:	reaction enumeration reagent connection

Goto Forum:

-=] Back to Top [=-

[ Syndicate this forum (XML) ] [

] [

]

Current Time: Wed May 21 06:00:08 CEST 2025

Total time taken to generate the page: 0.03407 seconds