def load_dataframe(self, file_resources, npartitions=None):
"""
Args:
file_resources:
npartitions:
"""
go_terms = pd.read_table(
file_resources["rnacentral_rfam_annotations.tsv"],
low_memory=True,
header=None,
names=["RNAcentral id", "GO terms", "Rfams"])
go_terms["RNAcentral id"] = go_terms["RNAcentral id"].str.split(
"_", expand=True, n=2)[0]
gene_ids = []
for file in file_resources:
if "database_mappings" in file:
if npartitions:
id_mapping = dd.read_table(file_resources[file],
header=None,
names=[
"RNAcentral id", "database",
"external id", "species",
"RNA type", "gene symbol"
])
else:
id_mapping = pd.read_table(file_resources[file],
low_memory=True,
header=None,
names=[
"RNAcentral id", "database",
"external id", "species",
"RNA type", "gene symbol"
])
gene_ids.append(id_mapping)
if npartitions:
gene_ids = dd.concat(gene_ids, join="inner")
else:
gene_ids = pd.concat(gene_ids, join="inner")
gene_ids["species"] = gene_ids["species"].astype("O")
if self.species is not None:
gene_ids = gene_ids[gene_ids["species"] == self.species]
lnc_go_terms = go_terms[go_terms["RNAcentral id"].isin(
gene_ids["RNAcentral id"])].groupby("RNAcentral id")[
"GO terms"].apply(lambda x: "|".join(x.unique()))
lnc_rfams = go_terms[go_terms["RNAcentral id"].isin(
gene_ids["RNAcentral id"])].groupby(
"RNAcentral id")["Rfams"].apply(lambda x: "|".join(x.unique()))
gene_ids["GO terms"] = gene_ids["RNAcentral id"].map(lnc_go_terms)
gene_ids["Rfams"] = gene_ids["RNAcentral id"].map(lnc_rfams)
gene_ids = gene_ids[gene_ids["GO terms"].notnull()
| gene_ids["Rfams"].notnull()]
return gene_ids
def add_drug_response_data(
self,
file_path="nationwidechildrens.org_clinical_drug.txt",
patient_column="bcr_patient_barcode",
columns=[
'bcr_patient_barcode', 'pharmaceutical_therapy_drug_name',
'pharmaceutical_therapy_type', 'treatment_best_response'
],
drug_name_col=None,
response_column=None):
"""
Args:
file_path:
patient_column:
columns:
drug_name_col:
response_column:
"""
if not os.path.exists(file_path):
raise FileNotFoundError(file_path)
self.drug_name_col = drug_name_col
self.response_column = response_column
self.drugs = pd.read_table(
file_path,
sep="\t",
skiprows=[1, 2],
na_values=["[Not Available]", "[Unknown]", "[Not Applicable]"],
usecols=columns)
self.drugs.set_index(patient_column, inplace=True)
def load_dataframe(self, file_resources, npartitions=None):
"""
Args:
file_resources:
npartitions:
"""
source_df = pd.read_table(file_resources["NONCODEv5_source"],
header=None)
source_df.columns = ["NONCODE Transcript ID", "name type", "Gene ID"]
transcript2gene_df = pd.read_table(
file_resources["NONCODEv5_Transcript2Gene"], header=None)
transcript2gene_df.columns = [
"NONCODE Transcript ID", "NONCODE Gene ID"
]
if npartitions:
self.noncode_func_df = dd.read_table(
file_resources["NONCODEv5_human.func"], header=None)
else:
self.noncode_func_df = pd.read_table(
file_resources["NONCODEv5_human.func"], header=None)
self.noncode_func_df.columns = ["NONCODE Gene ID", "GO terms"]
self.noncode_func_df.set_index("NONCODE Gene ID", inplace=True)
# Convert to NONCODE transcript ID for the functional annotation data
self.noncode_func_df[
"NONCODE Transcript ID"] = self.noncode_func_df.index.map(
pd.Series(
transcript2gene_df['NONCODE Transcript ID'].values,
index=transcript2gene_df['NONCODE Gene ID']).to_dict())
# Convert NONCODE transcript ID to gene names
source_gene_names_df = source_df[source_df["name type"] ==
"NAME"].copy()
self.noncode_func_df["Gene Name"] = self.noncode_func_df[
"NONCODE Transcript ID"].map(
pd.Series(source_gene_names_df['Gene ID'].values,
index=source_gene_names_df['NONCODE Transcript ID']).
to_dict())
def load_dataframe(self, file_resources, npartitions=None):
"""
Args:
file_resources:
npartitions:
"""
if npartitions:
df = dd.read_table(file_resources["proteinatlas.tsv"])
else:
df = pd.read_table(file_resources["proteinatlas.tsv"])
return df
def get_expressions(self, genes_index):
"""Preprocess LNCRNA expression file obtained from TANRIC MDAnderson,
and replace ENSEMBL gene ID to HUGO gene names (HGNC). This function
overwrites the GenomicData.process_expression_table() function which
processes TCGA-Assembler data. TANRIC LNCRNA expression values are log2
transformed
Args:
genes_index:
"""
df = pd.read_table(self.file_resources["TCGA-LUAD-rnaexpr.tsv"])
df[genes_index] = df[genes_index].str.replace(
"[.].*", "") # Removing .# ENGS gene version number at the end
df = df[~df[genes_index].duplicated(
keep='first')] # Remove duplicate genes
# Drop NA gene rows
df.dropna(axis=0, inplace=True)
# Transpose matrix to patients rows and genes columns
df.index = df[genes_index]
df = df.T.iloc[1:, :]
# Change index string to bcr_sample_barcode standard
def change_patient_barcode(s):
if "Normal" in s:
return s[s.find('TCGA'):] + "-11A"
elif "Tumor" in s:
return s[s.find('TCGA'):] + "-01A"
else:
return s
df.index = df.index.map(change_patient_barcode)
df.index.name = "gene_id"
return df
def add_biospecimen_data(
self,
file_path="genome.wustl.edu_biospecimen_sample.txt",
patient_col_name="bcr_patient_barcode",
columns=['bcr_sample_barcode', 'sample_type']):
"""
Args:
file_path:
patient_col_name:
columns:
"""
if not os.path.exists(file_path):
raise FileNotFoundError(file_path)
self.biospecimen = pd.read_table(
file_path,
sep="\t",
skiprows=[
1,
],
na_values=["[Not Available]", "[Unknown]", "[Not Applicable]"],
usecols=columns)
self.sample_barcodes = self.biospecimen[patient_col_name].tolist()
self.biospecimen.set_index(patient_col_name, inplace=True)
def __init__(self,
patients_file,
patient_id_col="bcr_patient_barcode",
columns=None):
"""This class manages the clinical data tables to handle the phenotype,
treatment, and sample data associated to a patient.
Args:
patients_file (str): path to the patients clinical data file
patient_id_col (str): the patient's ID column name
columns (list): default None. Specifies the columns to import, if
None, then import all columns. Example: ['bcr_patient_barcode',
'gender', 'race', 'histologic_diagnosis', 'tumor_status',
'death_days_to',
'ajcc_pathologic_tumor_stage']
"""
# self.cohort_name = cohort_name
self.patient_column = patient_id_col
if columns and patient_id_col not in columns:
columns.append(patient_id_col)
if isinstance(patients_file, io.StringIO):
patients_file.seek(
0
) # Needed since the file was previous read to extract columns information
self.patient = pd.read_table(patients_file,
skiprows=[1, 2],
na_values=[
"[Not Available]", "[Unknown]",
"[Not Applicable]",
"[Discrepancy]"
],
usecols=columns)
elif type(patients_file) == str and os.path.exists(patients_file):
self.patient = pd.read_table(patients_file,
skiprows=[1, 2],
na_values=[
"[Not Available]", "[Unknown]",
"[Not Applicable]",
"[Discrepancy]"
],
usecols=columns)
else:
raise IOError(patients_file)
self.patient_barcodes = self.patient[patient_id_col].tolist()
self.patient.set_index(patient_id_col, inplace=True)
# Rename columns
self.patient.rename(
{
"ajcc_pathologic_tumor_stage": PATHOLOGIC_STAGE,
"histological_type": HISTOLOGIC_SUBTYPE,
"histologic_diagnosis.1": HISTOLOGIC_SUBTYPE
},
axis=1,
inplace=True)
self.patient.replace(
{PATHOLOGIC_STAGE: ClinicalData.pathologic_stage_map},
inplace=True)