Skip to content

Commit

Permalink
fix #76 #75 #77 #62
Browse files Browse the repository at this point in the history
  • Loading branch information
@nargesr
nargesr committed Dec 4, 2023
1 parent b941d68 commit 7c781fa
Show file tree
Hide file tree
Showing 3 changed files with 31 additions and 21 deletions.
4 changes: 3 additions & 1 deletion PyWGCNA/utils.py
View file
Original file line number Diff line number Diff line change
@@ -1,6 +1,7 @@
import pickle
import os
import biomart
import pandas as pd
import requests
import matplotlib.pyplot as plt
import networkx as nx
Expand Down Expand Up @@ -123,10 +124,11 @@ def getGeneList(dataset='mmusculus_gene_ensembl',
# Store the data in a dict
for line in data.splitlines():
line = line.split('\t')
tmp = pd.DataFrame(line, index=attributes).T
dict = {}
for i in range(len(attributes)):
dict[attributes[i]] = line[i]
geneInfo = geneInfo.append(dict, ignore_index=True)
geneInfo = pd.concat([geneInfo, tmp], ignore_index=True)

geneInfo.index = geneInfo[attributes[0]]
geneInfo.drop(attributes[0], axis=1, inplace=True)
Expand Down
38 changes: 23 additions & 15 deletions PyWGCNA/wgcna.py
View file
Original file line number Diff line number Diff line change
Expand Up @@ -1741,7 +1741,7 @@ def cutreeHybrid(dendro, distM, cutHeight=None, minClusterSize=20, deepSplit=1,
useObjects = ColorsX in np.unique(labelsOnBranch)
DistSClustClust = distM.iloc[InCluster, useObjects]
MeanDist = DistSClustClust.mean(axis=0)
useColorsFac = pd.Categorical(ColorsX[useObjects])
useColorsFac = ColorsX[useObjects]#pd.Categorical(ColorsX[useObjects])
MeanDist = pd.DataFrame({'MeanDist': MeanDist, 'useColorsFac': useColorsFac})
MeanMeanDist = MeanDist.groupby(
'useColorsFac').mean() # tapply(MeanDist, useColorsFac, mean)
Expand All @@ -1760,7 +1760,7 @@ def cutreeHybrid(dendro, distM, cutHeight=None, minClusterSize=20, deepSplit=1,
InCluster = np.where(SmallLabels == sclust)[0].tolist()
DistSClustClust = distM.iloc[InCluster, useObjects]
MeanDist = DistSClustClust.mean(axis=0)
useColorsFac = pd.Categorical(ColorsX[useObjects])
useColorsFac = ColorsX[useObjects]#pd.Categorical(ColorsX[useObjects])
MeanDist = pd.DataFrame({'MeanDist': MeanDist, 'useColorsFac': useColorsFac})
MeanMeanDist = MeanDist.groupby(
'useColorsFac').mean() # tapply(MeanDist, useColorsFac, mean)
Expand All @@ -1782,7 +1782,7 @@ def cutreeHybrid(dendro, distM, cutHeight=None, minClusterSize=20, deepSplit=1,
basicOnBranch = branch_basicClusters[onBr - 1]
labelsOnBranch = branchLabels[basicOnBranch]
useObjects = ColorsX in np.unique(labelsOnBranch)
useColorsFac = pd.Categorical(ColorsX[useObjects])
useColorsFac = ColorsX[useObjects]#pd.Categorical(ColorsX[useObjects])
UnassdToClustDist = distM.iloc[useObjects, obj].groupby(
'useColorsFac').mean() # tapply(distM[useObjects, obj], useColorsFac, mean)
nearest = UnassdToClustDist.idxmin().astype(int) - 1
Expand All @@ -1794,7 +1794,7 @@ def cutreeHybrid(dendro, distM, cutHeight=None, minClusterSize=20, deepSplit=1,
nPAMed = nPAMed + 1
else:
useObjects = np.where(ColorsX != 0)[0].tolist()
useColorsFac = pd.Categorical(ColorsX[useObjects])
useColorsFac = ColorsX[useObjects]#pd.Categorical(ColorsX[useObjects])
tmp = pd.DataFrame(distM.iloc[useObjects, Unlabeled])
tmp['group'] = useColorsFac
UnassdToClustDist = tmp.groupby(
Expand Down Expand Up @@ -2000,9 +2000,9 @@ def moduleEigengenes(expr, colors, impute=True, nPC=1, align="along average", ex
u = u[:, 0:min(n, p, nPC)]
v = v[0:min(n, p, nPC), :]
tmp = datModule.copy()
tmp = tmp.append(pd.DataFrame(v[0:min(n, p, nVarExplained), :],
columns=tmp.columns.tolist()),
ignore_index=True)
tmp = pd.concat([tmp, pd.DataFrame(v[0:min(n, p, nVarExplained), :],
columns=tmp.columns.tolist())],
ignore_index=True)
veMat = pd.DataFrame(np.corrcoef(tmp.values)).iloc[:-1, -1].T
varExpl.iloc[0:min(n, p, nVarExplained), i] = (veMat ** 2).mean(axis=0)
pc = v[0].tolist()
Expand All @@ -2020,7 +2020,7 @@ def moduleEigengenes(expr, colors, impute=True, nPC=1, align="along average", ex
scaledExpr = pd.DataFrame(scale(datModule.T).T, index=datModule.index,
columns=datModule.columns)
covEx = np.cov(scaledExpr)
covEx[not np.isfinite(covEx)] = 0
covEx[~ np.isfinite(covEx)] = 0
modAdj = np.abs(covEx) ** softPower
kIM = (modAdj.mean(axis=0)) ** 3
if np.max(kIM) > 1:
Expand Down Expand Up @@ -2763,22 +2763,27 @@ def getDatTraits(self, metaData):

def module_trait_relationships_heatmap(self,
metaData,
figsize=None,
show=True,
file_name='module-traitRelationships'):
"""
plot topic-trait relationship heatmap
:param metaData: traits you would like to see the relationship with topics (must be column name of datExpr.obs)
:type metaData: list
:param figsize: indicate the size of plot
:type figsize: tuple of float
:param show: indicate if you want to show the plot or not (default: True)
:type show: bool
:param file_name: name and path of the plot use for save (default: topic-traitRelationships)
:type file_name: str
"""
datTraits = self.getDatTraits(metaData)

fig, ax = plt.subplots(figsize=(max(20, int(self.moduleTraitPvalue.shape[0] * 1.5)),
self.moduleTraitPvalue.shape[1] * 1.5), facecolor='white')
if figsize is None:
figsize = (max(20, int(self.moduleTraitPvalue.shape[0] * 1.5)),
self.moduleTraitPvalue.shape[1] * 1.5)
fig, ax = plt.subplots(figsize=figsize, facecolor='white')
# names
xlabels = []
for label in self.MEs.columns:
Expand Down Expand Up @@ -3110,7 +3115,7 @@ def barplotModuleEigenGene(self, moduleName, metadata, combine=True, colorBar=No
else:
return axs

def functional_enrichment_analysis(self, type, moduleName, sets=None, p_value=1, file_name=None):
def functional_enrichment_analysis(self, type, moduleName, sets=None, p_value=1, file_name=None, **kwargs):
"""
Doing functional enrichment analysis including GO, KEGG and REACTOME
Expand All @@ -3124,6 +3129,8 @@ def functional_enrichment_analysis(self, type, moduleName, sets=None, p_value=1,
:type p_value: float
:param file_name: name of the file you want to use to save plot (default is moduleName)
:type file_name: str
:param kwargs: Other keyword arguments are passed through to the underlying gseapy.enrichr() finction
:type kwargs: key, value pairings
"""
if type not in ["GO", "KEGG", "REACTOME"]:
sys.exit("Type is not valid! it should be one of them GO, KEGG, REACTOME")
Expand Down Expand Up @@ -3154,7 +3161,8 @@ def functional_enrichment_analysis(self, type, moduleName, sets=None, p_value=1,
gene_sets=sets,
organism=self.species,
outdir=f"{self.outputPath}figures/{type}/{file_name}",
cutoff=p_value)
cutoff=p_value,
**kwargs)
dotplot(enr.res2d,
title=f"Gene ontology in {moduleName} module",
cmap='viridis_r',
Expand Down Expand Up @@ -3409,9 +3417,9 @@ def CoexpressionModulePlot(self, modules, numGenes=10, numConnections=100, minTO

adj = self.TOM.loc[genes, genes]
adj[adj < minTOM] = 0
adj = adj.where(np.triu(np.ones(adj.shape)).astype(bool))
adj = adj.where(adj.values != np.diag(adj), 0,
adj.where(adj.values != np.flipud(adj).diagonal(0), 0, inplace=True))
adj.where(np.triu(np.ones(adj.shape)).astype(bool), inplace=True)
adj.where(adj.values != np.diag(adj), 0, inplace=True)
adj.where(adj.values != np.flipud(adj).diagonal(0), 0, inplace=True)
adj = adj.stack().nlargest(numConnections)

net = Network()
Expand Down
10 changes: 5 additions & 5 deletions setup.py
View file
Original file line number Diff line number Diff line change
Expand Up @@ -3,17 +3,17 @@
setup(
name='PyWGCNA', # the name of your package
packages=['PyWGCNA'], # same as above
version='1.20.2', # version number
version='2.0.0', # version number
license='MIT', # license type
description='PyWGCNA is a Python package designed to do Weighted correlation network analysis (WGCNA)',
# short description
author='Narges Rezaie', # your name
author_email='[email protected]', # your email
url='https://github.com/mortazavilab/PyWGCNA', # url to your git repo
download_url='https://github.com/mortazavilab/PyWGCNA/archive/refs/tags/v1.20.2.tar.gz', # link to the tar.gz file associated with this release
download_url='https://github.com/mortazavilab/PyWGCNA/archive/refs/tags/v1.20.4.tar.gz', # link to the tar.gz file associated with this release
keywords=['PyWGCNA', 'WGCNA', 'bulk', 'gene clustering', 'network analysis'], #
install_requires=[ # these can also include >, <, == to enforce version compatibility
'pandas==1.4.4', # make sure the packages you put here are those NOT included in the
'pandas>=2.1.0', # make sure the packages you put here are those NOT included in the
'numpy>=1.24.0', # base python distribution
'scipy>=1.9.1',
'scikit-learn>=1.2.2',
Expand All @@ -22,7 +22,7 @@
'seaborn>=0.11.2',
'biomart>=0.9.2',
'gseapy>=1.0.1',
'pyvis>=0.3.1',
'pyvis==0.3.1',
'setuptools>=67.4.0',
'biomart>=0.9.2',
'reactome2py>=3.0.0',
Expand All @@ -38,6 +38,6 @@
'Intended Audience :: Science/Research ',
'Topic :: Scientific/Engineering :: Bio-Informatics',
'License :: OSI Approved :: MIT License',
'Programming Language :: Python :: 3.9',
'Programming Language :: Python :: 3.10',
],
)

0 comments on commit 7c781fa

Please sign in to comment.