)
LOG = logging.getLogger("Crispy")
DPATH = pkg_resources.resource_filename("crispy", "data/")
RPATH = pkg_resources.resource_filename("reports", "eg/")
if __name__ == "__main__":
# Data-sets
#
wes_obj = WES()
mobem_obj = Mobem()
cn_obj = CopyNumber()
prot_obj = Proteomics()
gexp_obj = GeneExpression()
crispr_obj = CRISPR()
drug_obj = DrugResponse()
# Samples
#
samples = set.intersection(set(prot_obj.get_data()))
LOG.info(f"Samples: {len(samples)}")
# Filter data-sets
#
prot = prot_obj.filter(subset=samples)
LOG.info(f"Proteomics: {prot.shape}")
ginfo = ginfo[ginfo["Chromosome/scaffold name"].isin(Utils.CHR_ORDER)]
ginfo_pos = pd.concat([
ginfo.groupby("Gene name")["Chromosome/scaffold name"].first().rename(
"chr"),
ginfo.groupby("Gene name")["mean_pos"].mean().rename("chr_pos"),
],
axis=1)
# Y matrices
#
gexp_obj = GeneExpression()
gexp = gexp_obj.filter()
LOG.info(f"Gexp: {gexp.shape}")
prot_obj = Proteomics()
prot = prot_obj.filter()
prot = prot[prot.count(1) > 300]
LOG.info(f"Prot: {prot.shape}")
# X matrices
#
crispr_obj = CRISPR()
crispr = crispr_obj.filter(dtype="merged")
LOG.info(f"CRISPR: {crispr.shape}")
drespo_obj = DrugResponse()
drespo = drespo_obj.filter()
drespo = drespo[drespo.count(1) > 300]
drespo = drespo[["+" not in i for i in drespo.index]]
drespo.index = [";".join(map(str, i)) for i in drespo.index]
from crispy.LMModels import LMModels, LModel
from sklearn.preprocessing import StandardScaler
from sklearn.linear_model import LinearRegression
from statsmodels.stats.multitest import multipletests
from crispy.DataImporter import Proteomics, CRISPR, GeneExpression, DrugResponse, Sample
LOG = logging.getLogger("Crispy")
RPATH = pkg_resources.resource_filename("reports", "eg/")
if __name__ == "__main__":
# Data-sets
#
gexp_obj = GeneExpression()
prot_obj = Proteomics()
crispr_obj = CRISPR()
drespo_obj = DrugResponse()
# Samples
#
samples_crispr = set.intersection(
set(prot_obj.get_data()), set(crispr_obj.get_data(dtype="merged"))
)
LOG.info(f"CRISPR samples: {len(samples_crispr)}")
samples_drug = set.intersection(
set(prot_obj.get_data()), set(drespo_obj.get_data())
)
LOG.info(f"Drug samples: {len(samples_drug)}")