Re: [Rdkit-discuss] RDkit cartridge on Amazon RDS

2015-05-25 Thread Tim Dudgeon
I followed up with AWS on this, and they they stated: For the extension to be available we need to test it and white-list the extension. I have put in a feature request for the module to added in the future. Please note that this may take some time and there is no definite deadline at this

[Rdkit-discuss] Molecular dis / similarity using fingerprints

2015-05-25 Thread JP
RDKitters, I have a partial RDKit / partial Methodology question. I hope this email isn't much of the how long is a piece of string nature. I have a set of molecules (~30,000) which I would like to get a structural diversity index for. So I thought easy - generate some fingerprint I fancy

Re: [Rdkit-discuss] Molecular dis / similarity using fingerprints

2015-05-25 Thread Tim Dudgeon
ECFP, like all fingerprints are notoriously sensitive to trivial structural changes so your example is no surprise (the extra carbonyl group also affects aromaticity depending on your view of aromaticity). But I don't think there is anything simple and fast that's better. Maybe a clustering

Re: [Rdkit-discuss] Molecular dis / similarity using fingerprints

2015-05-25 Thread Greg Landrum
Hi JP. On Mon, May 25, 2015 at 4:10 PM, JP jeanpaul.ebe...@inhibox.com wrote: I have a set of molecules (~30,000) which I would like to get a structural diversity index for. So I thought easy - generate some fingerprint I fancy (ECFP-like, rad 2), take a threshold I fancy (0.7), select a