Re: [Rdkit-discuss] adding MORSE & RDF to 3D descriptors

2016-10-13 Thread Greg Landrum
Taking this off list for the moment. On Fri, Oct 14, 2016 at 7:14 AM, Guillaume GODIN < guillaume.go...@firmenich.com> wrote: > Hi Greg, > > > So I fix the compilation issues for python but I have a segmentation fault > error and I cannot find the error in RDF part... > > > Could you take a look

Re: [Rdkit-discuss] adding MORSE & RDF to 3D descriptors

2016-10-13 Thread Guillaume GODIN
Hi Greg, So I fix the compilation issues for python but I have a segmentation fault error and I cannot find the error in RDF part... Could you take a look please ? BR, Dr. Guillaume GODIN Principal Scientist Chemoinformatic & Datamining Innovation CORPORATE R DIVISION DIRECT LINE +41

Re: [Rdkit-discuss] How to find the idx of hydrogens bonded to a specific atom

2016-10-13 Thread Bin Song
Hello Dimitri, Yes, I do have explicit hydrogen throughout entire process. Thank you for the tip. I will check out the link. Bin Original Message Subject: Re: [Rdkit-discuss] How to find the idx of hydrogens bonded to a specific atom Local Time: October 14, 2016 2:26

Re: [Rdkit-discuss] How to find the idx of hydrogens bonded to a specific atom

2016-10-13 Thread Bin Song
Hello Paul, I will take a look at your code. Thank you. Original Message Subject: Re: [Rdkit-discuss] How to find the idx of hydrogens bonded to a specific atom Local Time: October 14, 2016 1:12 AM UTC Time: October 13, 2016 5:12 PM From: pems...@mrc-lmb.cam.ac.uk To:

Re: [Rdkit-discuss] How to find the idx of hydrogens bonded to a specific atom

2016-10-13 Thread Dimitri Maziuk
On 10/13/2016 12:12 PM, Paul Emsley wrote: > Are you sure? I use HasSubstrMatch to match hydrogens. Perhaps this http://www.mail-archive.com/rdkit-discuss@lists.sourceforge.net/msg05897.html may be useful? If OP's looking at crystal structures, they're likely dealing with pdb data model where

Re: [Rdkit-discuss] How to find the idx of hydrogens bonded to a specific atom

2016-10-13 Thread Paul Emsley
On 13/10/2016 05:04, Bin Song wrote: > Hello everyone, > > A new user of Rdkit and new subscriber of the mailing list reporting here. > > I have a question to ask, as stated in the subject. I have googled and > consulted the > documentation of rdkit, but found no answer. Basically, I want to run

Re: [Rdkit-discuss] property of name in smilesMolSupplier

2016-10-13 Thread Greg Landrum
On Thu, Oct 13, 2016 at 2:30 PM, Brian Kelley wrote: > I agree. I tend to think molecules should have GetName/SetName since this > is a common operation. > > I used to argue about this point, but at this point I've either been convinced or just gotten tired of the

Re: [Rdkit-discuss] property of name in smilesMolSupplier

2016-10-13 Thread 杨弘宾
Hi Paolo,?? ? Thank you for your answer. I tried and found an interesting issue: >>> mol = suppl[1] >>> print mol.GetProp('_Name') >>> print list(mol.GetPropNames())---output---1[] I guess that the _Name property is a hidden variable. ?Maybe the user cannot find this property_name

Re: [Rdkit-discuss] property of name in smilesMolSupplier

2016-10-13 Thread Brian Kelley
Paolo, _Name might be considered a private property, you might need to use GetPropNames(True,True) or something like that. Brian Kelley > On Oct 13, 2016, at 6:56 AM, Paolo Tosco wrote: > > Hi Hongbin, > > suppl[0].GetPropNames() is an interable object, so you can

Re: [Rdkit-discuss] property of name in smilesMolSupplier

2016-10-13 Thread Brian Kelley
I agree. I tend to think molecules should have GetName/SetName since this is a common operation. One note though: mol.GetPropsAsDict() Might help with introspection. There are optionally arguments that allow looking at private and computed properties, I.e. mol.GetPropsAsDict(True,True) I

Re: [Rdkit-discuss] property of name in smilesMolSupplier

2016-10-13 Thread Paolo Tosco
Hi Hongbin, suppl[0].GetPropNames() is an interable object, so you can use it in for loops such as: for i in suppl[0].GetPropNames(): print (i) or you may convert it to a list: l = list(suppl[0].GetPropNames()) print (l) Cheers, p. On 10/13/16 11:31, 杨弘宾 wrote: Hi， I spent a lot of

[Rdkit-discuss] property of name in smilesMolSupplier

2016-10-13 Thread 杨弘宾
Hi，? ? I spent a lot of time to explorer "How to get the property of name when using SmilesMolSupplier" I had a smiles file like this:===c1c1\t1\n...===where 1 means that it is positve So I wanted to read this smiles file via SMilesMolSupplier and I knew that the second column is the

Re: [Rdkit-discuss] adding MORSE & RDF to 3D descriptors

2016-10-13 Thread Guillaume GODIN
?Ho yes, sorry I didn't see that! thanks Dr. Guillaume GODIN Principal Scientist Chemoinformatic & Datamining Innovation CORPORATE R DIVISION DIRECT LINE +41 (0)22 780 3645 MOBILE +41 (0)79 536 1039 Firmenich SA RUE DES JEUNES 1 | CASE POSTALE 239 | CH-1211 GENEVE 8

Re: [Rdkit-discuss] adding MORSE & RDF to 3D descriptors

2016-10-13 Thread Greg Landrum
ha! that was a fun one. The compile guards at in RDF.h: https://github.com/thegodone/rdkit/blob/master/Code/GraphMol/Descriptors/RDF.h#L44 and MORSE.h: https://github.com/thegodone/rdkit/blob/master/Code/GraphMol/Descriptors/MORSE.h#L44 are wrong. By the way, the syntax: std::vector

[Rdkit-discuss] How to find the idx of hydrogens bonded to a specific atom

2016-10-13 Thread Bin Song
Hello everyone, A new user of Rdkit and new subscriber of the mailing list reporting here. I have a question to ask, as stated in the subject. I have googled and consulted the documentation of rdkit, but found no answer. Basically, I want to run a search with a smarts string on my molecule. I

Re: [Rdkit-discuss] adding MORSE & RDF to 3D descriptors

2016-10-13 Thread Guillaume GODIN
Dear Greg, Thanks for the tip, but same error after adding the if statement. All the cpp & cpp test works fine, but I have not clue why python wrap failed. BR, Guillaume Dr. Guillaume GODIN Principal Scientist Chemoinformatic & Datamining Innovation CORPORATE R DIVISION DIRECT LINE +41

Re: [Rdkit-discuss] adding MORSE & RDF to 3D descriptors

2016-10-13 Thread Greg Landrum
Guillaume, I'm not sure if this is everything, but you definitely need to put an#ifdef RDK_BUILD_DESCRIPTORS3Dblock around the definition of MORSEs and RDFs in the Python wrapper code. -greg On Thu, Oct 13, 2016 at 10:25 AM +0200, "Guillaume GODIN" wrote:

[Rdkit-discuss] adding MORSE & RDF to 3D descriptors

2016-10-13 Thread Guillaume GODIN
Dear All, I trying to expose in python two vector of 3D descriptors but when I compile have got this error message [ 96%] Built target rdMolHash [ 96%] Building CXX object Code/GraphMol/Descriptors/Wrap/CMakeFiles/rdMolDescriptors.dir/rdMolDescriptors.cpp.o [ 98%] Built target