def load_resources():
"""Known resources:
SMILES,
STRUCTURES: containing Mol_b64 column,
BATCH, CONTAINER, DATA"""
mf_config = mft.load_config("config")
global SMILES
SMILES = mf_config["Paths"]["SmilesPath"]
global STRUCT
STRUCT = mf_config["Paths"]["StructPath"]
global DATA
DATA = mf_config["Paths"]["ContainerDataPath"]
global CONTAINER
CONTAINER = mf_config["Paths"]["ContainerPath"]
global BATCH
BATCH = mf_config["Paths"]["BatchPath"]
def load_resource(resource, limit_cols=True):
"""Known resources:
SMILES,
STRUCTURES: containing Mol_b64 column,
BATCH, CONTAINER"""
mf_config = mft.load_config("config")
res = resource.lower()
glbls = globals()
if "smi" in res:
if "SMILES" not in glbls:
# except NameError:
print("- loading resource: (SMILES)")
result = dd.read_csv(mf_config["Paths"]["SmilesPath"], sep="\t")
if isinstance(limit_cols, list):
result = result[limit_cols]
elif limit_cols is True and len(
mf_config["Paths"]["SmilesCols"]) > 0:
result = result[mf_config["Paths"]["SmilesCols"]]
# result = result.apply(pd.to_numeric, errors='ignore', axis=1)
global SMILES
SMILES = MolFrame()
SMILES.data = result
elif "struct" in res:
if "STRUCTURES" not in glbls:
# except NameError:
print("- loading resource: (STRUCTURES)")
result = dd.read_csv(mf_config["Paths"]["SmilesPath"], sep="\t")
if isinstance(limit_cols, list):
result = result[limit_cols]
elif limit_cols is True and len(
mf_config["Paths"]["StructCols"]) > 0:
result = result[mf_config["Paths"]["StructCols"]]
# result.data = result.data.apply(pd.to_numeric, errors='ignore', axis=1)
global STRUCT
STRUCT = result
elif "data" in res:
if "DATA" not in glbls:
print("- loading resource: (DATA)")
result = dd.read_csv(mf_config["Paths"]["ContainerDataPath"],
sep="\t",
compression="gzip")
if isinstance(limit_cols, list):
result = result[limit_cols]
elif limit_cols is True and len(
mf_config["Paths"]["ContainerDataCols"]) > 0:
result = result[mf_config["Paths"]["ContainerDataCols"]]
# result = result.apply(pd.to_numeric, errors='ignore', axis=1)
global DATA
DATA = MolFrame()
DATA.data = result
elif "cont" in res:
if "CONTAINER" not in glbls:
print("- loading resource: (CONTAINER)")
result = dd.read_csv(mf_config["Paths"]["ContainerPath"], sep="\t")
if isinstance(limit_cols, list):
result = result[limit_cols]
elif limit_cols is True and len(
mf_config["Paths"]["ContainerCols"]) > 0:
result = result[mf_config["Paths"]["ContainerCols"]]
# result = result.apply(pd.to_numeric, errors='ignore', axis=1)
global CONTAINER
CONTAINER = MolFrame()
CONTAINER.data = result
elif "batch" in res:
if "BATCH" not in glbls:
print("- loading resource: (BATCH)")
result = dd.read_csv(mf_config["Paths"]["BatchPath"], sep="\t")
if isinstance(limit_cols, list):
result = result[limit_cols]
elif limit_cols is True and len(
mf_config["Paths"]["BatchCols"]) > 0:
result = result[mf_config["Paths"]["BatchCols"]]
# result = result.apply(pd.to_numeric, errors='ignore', axis=1)
global BATCH
BATCH = MolFrame()
BATCH.data = result
else:
raise FileNotFoundError("# unknown resource: {}".format(resource))
from PIL import Image, ImageChops
from rdkit.Chem import AllChem as Chem
from rdkit.Chem.rdCoordGen import AddCoords # New coord. generation
from rdkit.Chem import Draw
from mol_frame import tools as mft
try:
Draw.DrawingOptions.atomLabelFontFace = "DejaVu Sans"
Draw.DrawingOptions.atomLabelFontSize = 18
except KeyError: # Font "DejaVu Sans" is not available
pass
config = mft.load_config()
USE_RDKIT_NEW_COORD = config["Options"].get("UseNewRdkitCoord", True)
# try:
# # Try to import Avalon so it can be used for generation of 2d coordinates.
# from rdkit.Avalon import pyAvalonTools as pyAv
# USE_AVALON_2D = True
# except ImportError:
# print("* Avalon not available. Using RDKit for 2d coordinate generation.")
# USE_AVALON_2D = False
def rescale(mol, f=1.4):
tm = np.zeros((4, 4), np.double)
for i in range(3):