def test_empty_fprints(self):
"""Empty fingerprints cannot be compared.
"""
desc_1 = Descriptor([0])
desc_2 = Descriptor([0])
sim_mes = SimilarityMeasure("tanimoto")
with self.assertRaises(ValueError):
sim_mes(desc_1, desc_2)
def showAllDescriptorsCallback(self):
"""update the descriptors dropdown to show descriptors."""
if "selected" in self.showAllDescriptorsButton.state():
self.molecularDescriptorCombobox[
"values"] = Descriptor.get_all_supported_descriptors()
else:
self.molecularDescriptorCombobox[
"values"] = values = Descriptor.get_supported_fprints()
return
def test_binary_only_metrics(self):
"""Metrics which only allow bit vectors should raise
a value error on other inputs.
"""
desc_1 = Descriptor([1, 2])
desc_2 = Descriptor([3, 4])
for metric in SimilarityMeasure.get_supported_binary_metrics():
sim_mes = SimilarityMeasure(metric)
with self.assertRaises(ValueError):
sim_mes(desc_1, desc_2)
def __init__(
self,
mol_graph=None,
mol_text=None,
mol_property_val=None,
mol_descriptor_val=None,
mol_src=None,
mol_smiles=None,
):
"""Constructor
Args:
mol_graph (RDKIT mol object): Graph-level information of molecule.
Implemented as an RDKIT mol object. Default is None.
mol_text (str): Text identifier of the molecule. Default is None.
Identifiers can be:
------------------
1. Name of the molecule.
2. SMILES string representing the molecule.
mol_property_val (float): Some property associated with the
molecule. This is typically the response being studied.
E.g. Boiling point, Selectivity etc. Default is None.
mol_descriptor_val (numpy ndarray): Descriptor value for the
molecule. Must be numpy array or list. Default is None.
mol_src (str):
Source file or SMILES string to load molecule. Acceptable files:
-> .pdb file
-> .txt file with SMILE string in first column, first row and
(optionally) property in second column, first row.
Default is None.
If provided mol_graph is attempted to be loaded from it.
mol_smiles (str): SMILES string for molecule. If provided, mol_graph
is loaded from it. If mol_text not set in keyword argument,
this string is used to set it.
"""
self.mol_graph = mol_graph
self.mol_text = mol_text
self.mol_property_val = mol_property_val
self.descriptor = (Descriptor() if mol_descriptor_val is None else
Descriptor(value=np.array(mol_descriptor_val)))
if mol_src is not None:
try:
self._set_molecule_from_file(mol_src)
except LoadingError as e:
raise e
if mol_smiles is not None:
try:
self._set_molecule_from_smiles(mol_smiles)
except LoadingError as e:
raise e
def test_vectornorm_length_errors(self):
"""
Vector norm-based similarities should only work with
descriptors of the same length, otherwise it should raise
a value error.
"""
desc_1 = Descriptor([1, 2])
desc_2 = Descriptor([3])
sim_mes_1 = SimilarityMeasure("l0_similarity")
sim_mes_2 = SimilarityMeasure("l1_similarity")
sim_mes_3 = SimilarityMeasure("l2_similarity")
with self.assertRaises(ValueError):
sim_mes_1(desc_1, desc_2)
with self.assertRaises(ValueError):
sim_mes_2(desc_1, desc_2)
with self.assertRaises(ValueError):
sim_mes_3(desc_1, desc_2)
def __init__(
self,
molecule_database_src,
molecule_database_src_type,
is_verbose,
similarity_measure,
n_threads=1,
fingerprint_type=None,
fingerprint_params=None,
sampling_ratio=1.0,
sampling_random_state=42,
):
"""Constructor for the MoleculeSet class.
Args:
sampling_ratio (float): Fraction of the molecules to keep. Useful
for selection subset of dataset for quick computations.
sampling_random_state (int): Random state used for sampling.
Default is 42.
"""
self.is_verbose = is_verbose
self.molecule_database = None
self.descriptor = Descriptor()
self.molecule_database, features = self._get_molecule_database(
molecule_database_src, molecule_database_src_type)
if features is not None:
self._set_descriptor(arbitrary_descriptor_vals=features)
if 0.0 < sampling_ratio < 1.0:
if self.is_verbose:
print(f"Using {int(sampling_ratio * 100)}% of the database...")
self._subsample_database(sampling_ratio=sampling_ratio,
random_state=sampling_random_state)
if fingerprint_type is not None:
# overrides if descriptor set in self._set_molecule_database
self._set_descriptor(fingerprint_type=fingerprint_type,
fingerprint_params=fingerprint_params)
self.similarity_measure = SimilarityMeasure(similarity_measure)
if n_threads == 'auto':
def speedup_eqn(n_mols, n_procs):
return 1.8505e-4 * n_mols + 2.235e-1 * n_procs + 7.082e-2
n_cores = psutil.cpu_count(logical=False)
n_mols = len(self.molecule_database)
if speedup_eqn(n_mols, n_cores) > 1.0:
self.n_threads = n_cores
elif speedup_eqn(n_mols, n_cores // 2) > 1.0:
self.n_threads = n_cores // 2
else:
self.n_threads = n_cores
else:
self.n_threads = n_threads
self.similarity_matrix = None
self._set_similarity_matrix()
def __init__(self, master=None):
"""Constructor for AIMSim.
Args:
master (tk, optional): tk window. Defaults to None.
"""
# build ui
self.window = tk.Tk() if master is None else tk.Toplevel(master)
self.window.title("AIMSim")
resource_path = pkg_resources.resource_filename(
__name__,
"AIMSim-logo.png",
)
self.window.iconphoto(False, tk.PhotoImage(file=resource_path))
self.databaseFile = tk.StringVar(self.window)
self.targetMolecule = tk.StringVar(self.window)
self.similarityMeasure = tk.StringVar(self.window)
self.molecularDescriptor = tk.StringVar(self.window)
self.titleLabel = ttk.Label(self.window)
self.titleLabel.configure(font="TkDefaultFont 14 bold",
text="AI Molecular Similarity")
self.titleLabel.place(anchor="center",
relx="0.5",
rely="0.05",
x="0",
y="0")
self.mainframe = ttk.Frame(self.window)
self.verboseCheckbutton = ttk.Checkbutton(self.mainframe)
self.verboseCheckbutton.configure(compound="top",
cursor="arrow",
offvalue="False",
onvalue="True")
self.verboseCheckbutton.configure(state="normal", text="Verbose")
self.verboseCheckbutton.place(anchor="center",
relx="0.1",
rely="0.95",
x="0",
y="0")
self.databaseFileEntry = ttk.Entry(self.mainframe,
textvariable=self.databaseFile)
_text_ = """smiles_responses.txt"""
self.databaseFileEntry.delete("0", "end")
self.databaseFileEntry.insert("0", _text_)
self.databaseFileEntry.place(anchor="center",
relx="0.5",
rely="0.03",
x="0",
y="0")
self.databaseFileLabel = ttk.Label(self.mainframe)
self.databaseFileLabel.configure(text="Database File:")
self.databaseFileLabel.place(anchor="center",
relx=".2",
rely="0.03",
x="0",
y="0")
self.targetMoleculeEntry = ttk.Entry(self.mainframe,
textvariable=self.targetMolecule)
_text_ = """CO"""
self.targetMoleculeEntry.delete("0", "end")
self.targetMoleculeEntry.insert("0", _text_)
self.targetMoleculeEntry.place(anchor="center",
relx="0.5",
rely="0.27",
x="0",
y="0")
self.targetMoleculeLabel = ttk.Label(self.mainframe)
self.targetMoleculeLabel.configure(text="Target Molecule:")
self.targetMoleculeLabel.place(anchor="center",
relx="0.17",
rely="0.27",
x="0",
y="0")
self.similarityPlotsCheckbutton = ttk.Checkbutton(self.mainframe)
self.similarityPlotsCheckbutton.configure(text="Similarity Plots")
self.similarityPlotsCheckbutton.place(anchor="center",
relx="0.5",
rely="0.1",
x="0",
y="0")
self.propertySimilarityCheckbutton = ttk.Checkbutton(self.mainframe)
self.propertySimilarityCheckbutton.configure(
text="Property Similarity Plot")
self.propertySimilarityCheckbutton.place(anchor="center",
relx="0.5",
rely="0.15",
x="0",
y="0")
self.similarityPlotCheckbutton = ttk.Checkbutton(self.mainframe)
self.similarityPlotCheckbutton.configure(text="Similarity Plot")
self.similarityPlotCheckbutton.place(anchor="center",
relx="0.5",
rely="0.35",
x="0",
y="0")
self.similarityMeasureCombobox = ttk.Combobox(
self.mainframe,
textvariable=self.similarityMeasure,
state="readonly")
self.similarityMeasureCombobox.configure(
takefocus=False, values=SimilarityMeasure.get_supported_metrics())
self.similarityMeasureCombobox.current(0)
self.similarityMeasureCombobox.place(anchor="center",
relx="0.55",
rely="0.45",
x="0",
y="0")
self.similarityMeasureLabel = ttk.Label(self.mainframe)
self.similarityMeasureLabel.configure(text="Similarity Measure:")
self.similarityMeasureLabel.place(anchor="center",
relx="0.2",
rely="0.45",
x="0",
y="0")
self.molecularDescriptorLabel = ttk.Label(self.mainframe)
self.molecularDescriptorLabel.configure(text="Molecular Descriptor:")
self.molecularDescriptorLabel.place(anchor="center",
relx="0.18",
rely="0.55",
x="0",
y="0")
self.molecularDescriptorCombobox = ttk.Combobox(
self.mainframe,
textvariable=self.molecularDescriptor,
state="readonly")
self.molecularDescriptorCombobox.configure(
cursor="arrow",
justify="left",
takefocus=False,
# values=Descriptor.get_all_supported_descriptors(),
values=Descriptor.get_supported_fprints(),
)
# define the callback for the descriptor
def updateCompatibleMetricsListener(event):
"""Show only compatible metrics, given a descriptor."""
self.similarityMeasureCombobox[
"values"] = SimilarityMeasure.get_compatible_metrics().get(
self.molecularDescriptor.get(), "Error")
self.similarityMeasureCombobox.current(0)
return
# bind this listener to the combobox
self.molecularDescriptorCombobox.bind("<<ComboboxSelected>>",
updateCompatibleMetricsListener)
self.molecularDescriptorCombobox.place(anchor="center",
relx="0.55",
rely="0.55",
x="0",
y="0")
self.molecularDescriptorCombobox.current(0)
self.runButton = ttk.Button(self.mainframe)
self.runButton.configure(text="Run")
self.runButton.place(anchor="center",
relx="0.5",
rely="0.75",
x="0",
y="0")
self.runButton.configure(command=self.runCallback)
self.openConfigButton = ttk.Button(self.mainframe)
self.openConfigButton.configure(text="Open Config")
self.openConfigButton.place(anchor="center",
relx="0.5",
rely="0.85",
x="0",
y="0")
self.openConfigButton.configure(command=self.openConfigCallback)
self.showAllDescriptorsButton = ttk.Checkbutton(self.mainframe)
self.showAllDescriptorsButton.configure(
compound="top",
cursor="arrow",
offvalue="False",
onvalue="True",
command=self.showAllDescriptorsCallback,
)
self.showAllDescriptorsButton.configure(
state="normal", text="Show experimental descriptors")
self.showAllDescriptorsButton.place(anchor="center",
relx="0.45",
rely="0.65",
x="0",
y="0")
self.multiprocessingCheckbutton = ttk.Checkbutton(self.mainframe)
self.multiprocessingCheckbutton.configure(compound="top",
cursor="arrow",
offvalue="False",
onvalue="True")
self.multiprocessingCheckbutton.configure(
state="normal", text="Enable Multiple Workers")
self.multiprocessingCheckbutton.place(anchor="center",
relx="0.78",
rely="0.95",
x="0",
y="0")
self.identifyOutliersCheckbutton = ttk.Checkbutton(self.mainframe)
self.identifyOutliersCheckbutton.configure(compound="top",
cursor="arrow",
offvalue="False",
onvalue="True")
self.identifyOutliersCheckbutton.configure(state="normal",
text="Outlier Check")
self.identifyOutliersCheckbutton.place(anchor="center",
relx="0.4",
rely="0.95",
x="0",
y="0")
self.mainframe.configure(height="400", width="400")
self.mainframe.place(anchor="nw",
relheight="0.9",
rely="0.1",
x="0",
y="0")
self.window.configure(cursor="arrow",
height="400",
relief="flat",
takefocus=False)
self.window.configure(width="400")
# Main widget
self.mainwindow = self.window