How to build mmpdb with large data set about mmpdb HOT 4 CLOSED

Dear Taka,

indexing chembl28 into a MMP database with the standard mmpdb tool requires a lot of memory and is not likely to be feasible with the standard settings. You can modify some of the settings to reduce the number of transformations and the resolution of the transformations, as discussed here:

#27

I am not sure though whether this will allow you to index chembl28, unless you reduce the DB to transformations of very small fragments. If you only want to index the transformations and not associate any properties with it, I recommend you check out this code which is built for exactly that purpose and derived from mmpdb:

https://github.com/mahendra-awale/medchem_moves

Hope this helps... Bests,
Christian

from mmpdb.

Dear Taka,

I will talk to Mahendra, the repo owner, to see what he can do.

Bests,
Christian

from mmpdb.

Dear Chirstian,

Thanks for your reply.
What I want to do is https://github.com/mahendra-awale/medchem_moves!
I read the code but the repo doesn't provide index_algorithm.py so I couldn't run index_algorithm.findRulesAndEnvs.
If you have any information about it could you please provide suggestions about it?

Kind regards,
Taka

from mmpdb.

Dear Chirstian,
Thanks! That sounds great.

Kind regards,
Taka

from mmpdb.