Skip to content

Instantly share code, notes, and snippets.

@Nireus-lgx
Forked from tdudgeon/conf_gen.py
Created October 18, 2023 02:44
Show Gist options
  • Select an option

  • Save Nireus-lgx/5f4406d5158a816b733000dc3e82dd27 to your computer and use it in GitHub Desktop.

Select an option

Save Nireus-lgx/5f4406d5158a816b733000dc3e82dd27 to your computer and use it in GitHub Desktop.
Conformer generation using RDKit
import sys
from rdkit import Chem
from rdkit.Chem import AllChem, TorsionFingerprints
from rdkit.ML.Cluster import Butina
def gen_conformers(mol, numConfs=100, maxAttempts=1000, pruneRmsThresh=0.1, useExpTorsionAnglePrefs=True, useBasicKnowledge=True, enforceChirality=True):
ids = AllChem.EmbedMultipleConfs(mol, numConfs=numConfs, maxAttempts=maxAttempts, pruneRmsThresh=pruneRmsThresh, useExpTorsionAnglePrefs=useExpTorsionAnglePrefs, useBasicKnowledge=useBasicKnowledge, enforceChirality=enforceChirality, numThreads=0)
return list(ids)
def write_conformers_to_sdf(mol, filename, rmsClusters, conformerPropsDict, minEnergy):
w = Chem.SDWriter(filename)
for cluster in rmsClusters:
for confId in cluster:
for name in mol.GetPropNames():
mol.ClearProp(name)
conformerProps = conformerPropsDict[confId]
mol.SetIntProp("conformer_id", confId + 1)
for key in conformerProps.keys():
mol.SetProp(key, str(conformerProps[key]))
e = conformerProps["energy_abs"]
if e:
mol.SetDoubleProp("energy_delta", e - minEnergy)
w.write(mol, confId=confId)
w.flush()
w.close()
def calc_energy(mol, conformerId, minimizeIts):
ff = AllChem.MMFFGetMoleculeForceField(mol, AllChem.MMFFGetMoleculeProperties(mol), confId=conformerId)
ff.Initialize()
ff.CalcEnergy()
results = {}
if minimizeIts > 0:
results["converged"] = ff.Minimize(maxIts=minimizeIts)
results["energy_abs"] = ff.CalcEnergy()
return results
def cluster_conformers(mol, mode="RMSD", threshold=2.0):
if mode == "TFD":
dmat = TorsionFingerprints.GetTFDMatrix(mol)
else:
dmat = AllChem.GetConformerRMSMatrix(mol, prealigned=False)
rms_clusters = Butina.ClusterData(dmat, mol.GetNumConformers(), threshold, isDistData=True, reordering=True)
return rms_clusters
def align_conformers(mol, clust_ids):
rmslist = []
AllChem.AlignMolConformers(mol, confIds=clust_ids, RMSlist=rmslist)
return rmslist
if len(sys.argv) < 4:
print "Usage: conf_gen.py <sdf input> <num conformers> <max attempts> <prune threshold> <cluster method: (RMSD|TFD) = RMSD> <cluster threshold = 0.2> <minimize iterations: = 0>"
exit()
input_file = sys.argv[1]
numConfs = int(sys.argv[2])
maxAttempts = int(sys.argv[3])
pruneRmsThresh = float(sys.argv[4])
if len(sys.argv) > 5: clusterMethod = sys.argv[5]
else: clusterMethod = "RMSD"
if len(sys.argv) > 6: clusterThreshold = float(sys.argv[6])
else: clusterThreshold = 2.0
if len(sys.argv) > 7: minimizeIterations = int(sys.argv[7])
else: minimizeIterations = 0
suppl = Chem.ForwardSDMolSupplier(input_file)
i=0
for mol in suppl:
i = i+1
if mol is None: continue
m = Chem.AddHs(mol)
# generate the confomers
conformerIds = gen_conformers(m, numConfs, maxAttempts, pruneRmsThresh, True, True, True)
conformerPropsDict = {}
for conformerId in conformerIds:
# energy minimise (optional) and energy calculation
props = calc_energy(m, conformerId, minimizeIterations)
conformerPropsDict[conformerId] = props
# cluster the conformers
rmsClusters = cluster_conformers(m, clusterMethod, clusterThreshold)
print "Molecule", i, ": generated", len(conformerIds), "conformers and", len(rmsClusters), "clusters"
rmsClustersPerCluster = []
clusterNumber = 0
minEnergy = 9999999999999
for cluster in rmsClusters:
clusterNumber = clusterNumber+1
rmsWithinCluster = align_conformers(m, cluster)
for conformerId in cluster:
e = props["energy_abs"]
if e < minEnergy:
minEnergy = e
props = conformerPropsDict[conformerId]
props["cluster_no"] = clusterNumber
props["cluster_centroid"] = cluster[0] + 1
idx = cluster.index(conformerId)
if idx > 0:
props["rms_to_centroid"] = rmsWithinCluster[idx-1]
else:
props["rms_to_centroid"] = 0.0
write_conformers_to_sdf(m, str(i) + ".sdf", rmsClusters, conformerPropsDict, minEnergy)
Sign up for free to join this conversation on GitHub. Already have an account? Sign in to comment