Jarp - Optimizing parameters

Hello all,





I am using Jarp to cluster a library of about 10000 compounds. Practically, what is the best way you suggest to optimise the Rmin parameter in NNeib algorithm ?


Thank you.





Best regards,





IsI

Hi IsI,





I think there's no better way than trial-and-error. The 'optimal' value depends so much on actual data set that I do not believe there is any practical method to predetermine it's value.


Perhaps one could do a very rigorous error analysis, but that's a big effort.





Hopefully other forum readers can give you some good pieces of advice.


Please share your thoughts and findings with the community after you accomplished this clustering exercise.





Thanks


Miklos