pub_soft/rdkit
3
0
Fork 0
mirror of https://github.com/rdkit/rdkit.git synced 2026-06-05 22:04:27 +08:00
Code Issues Packages Projects Releases Wiki Activity

Add multithreaded pattern/fp generator (#2973)

Browse Source 
* Add threaded pattern generator to help speed up library creation

* Add size to FPHolderBase

* Add more complete python wrapper, fix leak in doctest

* Add basic usage test for addPatterns, fix I/O leak

* Add basic test for trusted smiles

* Actually add the new code

* Export addPatterns on windows

* Try fixing the cartridge build

* Change pattern_factory to PatternFactory

* Change boost::ref to std::ref
This commit is contained in:
Brian Kelley
2020-03-04 09:41:03 -05:00
committed by GitHub
parent f616b504e2
commit 4c0a99f2ad
8 changed files with 287 additions and 14 deletions
Download Patch File Download Diff File Expand all files Collapse all files

60
Code/GraphMol/SubstructLibrary/substructLibraryTest.cpp
View File

@@ -37,9 +37,11 @@
#include <GraphMol/RDKitBase.h>
#include <GraphMol/RDKitQueries.h>
#include <GraphMol/SubstructLibrary/SubstructLibrary.h>
#include <GraphMol/SubstructLibrary/PatternFactory.h>
#include <GraphMol/Substruct/SubstructMatch.h>
#include <GraphMol/SmilesParse/SmilesParse.h>
#include <GraphMol/SmilesParse/SmilesWrite.h>
#include <GraphMol/FileParsers/FileParsers.h>
@@ -399,6 +401,63 @@ void ringTest() {
TEST_ASSERT(results.size() == 1);
BOOST_LOG(rdErrorLog) << " Done (C++ ring query tests)" << std::endl;
}
void testAddPatterns() {
BOOST_LOG(rdErrorLog) << "-------------------------------------" << std::endl;
BOOST_LOG(rdErrorLog) << " Add Patterns " << std::endl;
std::vector<std::string> pdb_ligands = {
"CCS(=O)(=O)c1ccc(OC)c(Nc2ncc(-c3cccc(-c4ccccn4)c3)o2)c1",
"COc1ccc(S(=O)(=O)NCC2CC2)cc1Nc1ncc(-c2cccc(-c3cccnc3)c2)o1",
"COc1ccc(-c2oc3ncnc(N)c3c2-c2ccc(NC(=O)Nc3cc(C(F)(F)F)ccc3F)cc2)cc1",
"COC(=O)Nc1nc2ccc(Oc3ccc(NC(=O)Nc4cc(C(F)(F)F)ccc4F)cc3)cc2[nH]1",
"COc1cc(Nc2ncnc(-c3cccnc3Nc3ccccc3)n2)cc(OC)c1OC",
"O=C(Nc1ccc(Oc2ccccc2)cc1)c1cccnc1NCc1ccncc1",
"O=C(Nc1ccc(Oc2ccccc2)cc1)c1cccnc1NCc1ccncc1",
"CNC(=O)c1cc(Oc2ccc3[nH]c(Nc4ccc(Cl)c(C(F)(F)F)c4)nc3c2)ccn1",
"CNC(=O)c1cc(Oc2ccc3oc(Nc4ccc(Cl)c(OCC5CCC[NH+]5C)c4)nc3c2)ccn1",
"CNC(=O)c1cc(Oc2ccc3oc(Nc4ccc(Cl)c(OCC5CCC[NH+]5C)c4)nc3c2)ccn1",
"COc1cc2nccc(Oc3ccc4c(c3)OCCN4C(=O)Nc3ccc(Cl)cc3)c2cc1OC",
"CNC(=O)c1c(C)oc2cc(Oc3cc[nH+]c4cc(OCCN5CCOCC5)ccc34)ccc12",
"COc1cc2[nH+]ccc(Oc3ccc4c(C(=O)Nc5ccc(Cl)cc5)cccc4c3)c2cc1OC",
"COc1cc2[nH+]ccc(Oc3ccc4c(C(=O)Nc5ccc(Cl)cc5)cccc4c3)c2cc1OC",
"COc1cc2[nH+]ccc(Oc3ccc4c(C(=O)NC5CC5)cccc4c3)c2cc1OC",
"COc1cc2[nH+]ccc(Oc3ccc4c(C(=O)NC5CC5)cccc4c3)c2cc1OC",
"Cc1ccc(C(=O)Nc2cc(CCC[NH+](C)C)cc(C(F)(F)F)c2)cc1Nc1ncccc1-c1ccncn1",
"COc1cc(Nc2nccc(Nc3ccc4c(C)n[nH]c4c3)n2)cc(OC)c1OC",
"COc1cc(Nc2nccc(N(C)c3ccc4c(C)n[nH]c4c3)n2)cc(OC)c1OC",
"Cc1ccn(-c2ccc3c(c2)NCC3(C)C)c(=O)c1-c1ccc2nc(N)ncc2c1",
"Cc1ccn(-c2ccc3c(c2)NCC3(C)C)c(=O)c1-c1ccc2nc(N)ncc2c1",
"Cc1ccc(C(=O)NCCC2CCCC2)cc1C(=O)Nc1ccc(N)nc1",
"Cc1ccc(C(=O)NCCC2CCCC2)cc1C(=O)Nc1ccc(N)nc1",
"Cc1ccn(-c2cccc(C(F)(F)F)c2)c(=O)c1-c1ccc2nc(N)ncc2c1",
"Cc1ccn(-c2cccc(C(F)(F)F)c2)c(=O)c1-c1ccc2nc(N)ncc2c1",
"O=C(Nc1cncnc1)c1c(Cl)ccc2c(Nc3cccc(C(F)(F)F)c3)noc12",
"O=C(Nc1cncnc1)c1c(Cl)ccc2c(Nc3cccc(C(F)(F)F)c3)noc12",
"CC1(C)CNc2cc(NC(=O)c3cccnc3NCc3ccncc3)ccc21",
"CC1(C)CNc2cc(NC(=O)c3cccnc3NCc3ccncc3)ccc21"};
boost::shared_ptr<CachedSmilesMolHolder> holder =
boost::make_shared<CachedSmilesMolHolder>();
for(auto s : pdb_ligands) {
holder->addSmiles(s);
}
SubstructLibrary ssslib(holder);
std::vector<int> num_threads = { 1, 0 };
for(auto nthreads : num_threads) {
SubstructLibrary ssslib_with_patterns(holder);
addPatterns(ssslib_with_patterns, nthreads);
for(unsigned int i=0; i<ssslib.size(); ++i) {
TEST_ASSERT( ssslib.countMatches( *ssslib.getMol(i).get() ) ==
ssslib_with_patterns.countMatches( *ssslib.getMol(i).get() ) );
}
}
}
int main() {
@@ -410,6 +469,7 @@ int main() {
test4();
docTest();
ringTest();
testAddPatterns();
#endif
return 0;
}