PyOVERCAST 1.0.3__tar.gz → 1.0.5__tar.gz

{pyovercast-1.0.3 → pyovercast-1.0.5}/PKG-INFO

@@ -1,6 +1,6 @@
 Metadata-Version: 2.4
 Name: PyOVERCAST
-Version: 1.0.3
+Version: 1.0.5
 Summary: A Python package for mining key transcription factors from transcriptome data.
 Author-email: Tinghua Huang <thua45@126.com>
 License-Expression: MIT
@@ -65,25 +65,28 @@ if __name__ == '__main__':
 	print(sets_names)
 
 	# list avaiable TFs
-	tfs_codes = clinks.get_tfs(set_name='human_hocomoco_CLink_wtcoor_1w_0.8')
+	tfs_codes = clinks.get_tfs(set_name='human_hocomoco_CLink_rci4_1w_0.8')
 	print(tfs_codes)
 
 	# get targets
-	targets = clinks.get_targets(set_name='human_hocomoco_CLink_wtcoor_1w_0.8', tf='NFKB1_HUMAN.H11MO.1.B')
+	targets = clinks.get_targets(set_name='human_hocomoco_CLink_rci4_1w_0.8', tf='NFKB1_HUMAN.H11MO.1.B')
 	print(targets)
 	
 	# predict one DEG-list
-	result = predict.olcr(set_names=['human_hocomoco_CLink_wtcoor_1w_0.8', 'human_jaspar_CLink_wtcoor_1w_0.8'], list_file='./PyOVERCAST_data/input_deg-list.txt', win=30, thread_n=16)
+	result = predict.olcr(set_names=['human_jaspar_CLink_rci4_1w_0.8', 'human_hocomoco_CLink_rci4_1w_0.8'], list_file='./PyOVERCAST_data/input_deg-list.txt', win=30, thread_n=16)
 
 	# or predict one DEG-list with bootstrap
-	result = predict.olcr_bootstrap(set_names=['human_hocomoco_CLink_wtcoor_1w_0.8', 'human_jaspar_CLink_wtcoor_1w_0.8'], list_file='./OVERCAST_data/input_deg-list.txt', win=30, bs_n=1000, thread_n=32)
+	result = predict.olcr_bootstrap(set_names=['human_jaspar_CLink_rci4_1w_0.8', 'human_hocomoco_CLink_rci4_1w_0.8'], list_file='./OVERCAST_data/input_deg-list.txt', win=30, bs_n=1000, thread_n=32)
 
 	# save result to text file
 	result.to_csv('output.txt', sep='\t', index=False, encoding='utf-8')
 	
 	# plot OLC matrix
-	predict.plot_olc(set_names=['human_hocomoco_CLink_wtcoor_1w_0.8', 'human_jaspar_CLink_wtcoor_1w_0.8'], list_file='./PyOVERCAST_data/input_deg-list.txt', tf='MA0844.2_XBP1', win=30)
+	predict.plot_olc(set_names=['human_jaspar_CLink_rci4_1w_0.8', 'human_hocomoco_CLink_rci4_1w_0.8'], list_file='./PyOVERCAST_data/input_deg-list.txt', tf='MA0844.2_XBP1', win=30)
+
+	# plot OLC matrix
+	predict.plot_olc(set_names=['human_jaspar_CLink_rci4_1w_0.8', 'human_hocomoco_CLink_rci4_1w_0.8'], list_file='./PyOVERCAST_data/input_deg-list.txt', tf='MA0844.2_XBP1', win=30)
 
 	# plot fitted 3D U-surface
-	predict.plot_fit3D(set_names=['human_hocomoco_CLink_wtcoor_1w_0.8', 'human_jaspar_CLink_wtcoor_1w_0.8'], list_file='./PyOVERCAST_data/input_deg-list.txt', tf='MA0844.2_XBP1', win=30)
+	predict.plot_contour(set_names=['human_jaspar_CLink_rci4_1w_0.8', 'human_hocomoco_CLink_rci4_1w_0.8'], list_file='./PyOVERCAST_data/input_deg-list.txt', tf='MA0844.2_XBP1', win=30)
 ```

{pyovercast-1.0.3 → pyovercast-1.0.5}/PyOVERCAST/predict.py

@@ -353,7 +353,7 @@ def plot_rrho(rrho_df, out_file=None, title="RRLO Matrix", cmap="magma"):
 	"""
 	Plot RRHO heatmap.
 	"""
-	plt.figure(figsize=(6, 5), dpi=150)
+	plt.figure(figsize=(8, 6))
 	ax = sns.heatmap(
 		rrho_df,
 		cmap=cmap,
@@ -363,8 +363,8 @@ def plot_rrho(rrho_df, out_file=None, title="RRLO Matrix", cmap="magma"):
 		yticklabels=max(1, len(rrho_df.index) // 10),
 	)
 	ax.invert_yaxis()  # 添加这一行，让 [0,0] 位于左下角
-	ax.set_xlabel("Prefix overlap in TF-target set")
-	ax.set_ylabel("Prefix overlap in DEG list")
+	ax.set_xlabel("Overlap in TF-target set")
+	ax.set_ylabel("Overlap in DEG list")
 	ax.set_title(title, fontsize=10)
 	plt.tight_layout()
 
@@ -762,15 +762,121 @@ def plot_fit3D(set_names=['human_hocomoco_CLink_wtcoor_1w_0.8'], list_file='inpu
 			plot_fit3D_plot(deg_list, tfbs_array[i][1], win)
 			break
 
+def plot_contour_graph(rrho_df, out_file=None):
+	Z_data = rrho_df.to_numpy()
+	row_n, col_n = Z_data.shape
+	#x = np.linspace(-1, 1, col_n)
+	#y = np.linspace(-1, 1, row_n)
+	y = rrho_df.index
+	x = rrho_df.columns
+	X, Y = np.meshgrid(x, y)
+
+	#x = np.linspace(-5, 5, 400)
+	#y = np.linspace(-5, 5, 400)
+	#X, Y = np.meshgrid(x, y)
+	#Z = np.sin(X) * np.cos(Y) * np.exp(-(X**2 + Y**2) / 10)
+
+	def fit_surface(Z, X, Y):
+		df = pd.DataFrame({
+			'x': X.ravel(),
+			'y': Y.ravel(),
+			'z': Z.ravel()
+		})
+		df['xy'] = df['x'] * df['y']  # 交互项
+		X_design = sm.add_constant(df[['x', 'y', 'xy']])
+		y_dep = df['z']
+		model = sm.OLS(y_dep, X_design).fit()
+		beta = model.params
+		# print(beta)
+
+		Z_fit = beta['const'] + beta['x']*X + beta['y']*Y + beta['xy']*X*Y
+		return beta, Z_fit
+
+	beta, Z = fit_surface(Z_data, X, Y)
+
+	fig, ax = plt.subplots(figsize=(8, 6))
+	levels = np.linspace(Z.min(), Z.max(), 16)
+	cs = ax.contour(X, Y, Z, levels=levels, cmap='viridis')
+	ax.clabel(cs, inline=True, fontsize=8, fmt='%.2f')
+	cf = ax.contourf(X, Y, Z, levels=levels, cmap='viridis', alpha=0.35)
+	fig.colorbar(cf, ax=ax, label='z value')
+	ax.set_xlabel('Rank of Gene Set'); ax.set_ylabel('Rank of DEG List')
+	ax.set_title('Contour Graph of RRLO Matrix')
+	# ax.set_aspect('equal')
+	plt.tight_layout()
+	plt.savefig('contour_graph_invF.png', dpi=150)
+	plt.show()
+
+
+def plot_contour_plot(df1, df2, win):
+	gene_col1 = df1.columns[0]
+	score_col1 = df1.columns[1]
+	df1 = df1.sort_values(score_col1, ascending=False)
+	genes1 = df1[gene_col1].astype(str).tolist()
+
+	# remove duplicates while preserving order
+	genes1 = list(dict.fromkeys(genes1))
+
+	gene_col2 = df1.columns[0]
+	score_col2 = df2.columns[1]
+	df2 = df2.sort_values(score_col2, ascending=False)
+	genes2 = df2[gene_col2].astype(str).tolist()
+	# remove duplicates while preserving order
+	genes2 = list(dict.fromkeys(genes2))
+	
+	window = win
+	if (len(genes1) < window*3 or len(genes2) < window*3):
+		return 0.0, 1.0, 0.0
+	rrho_df = rrho_matrix_str(genes1, genes2, window=window)
+
+	if rrho_df.shape[0] == 0 or rrho_df.shape[1] == 0 or (rrho_df > 0.0).sum().sum() == 0:
+		print('no overlaps found')
+		exit(1)
+
+	#beta_xy, pval_xy = td_coor(rrho_df.to_numpy())
+	# df = Z_to_frame(rrho_df.to_numpy())
+	plot_contour_graph(rrho_df)
+
+def plot_contour(set_names=['human_hocomoco_CLink_wtcoor_1w_0.8'], list_file='input_deg-list.txt', tf='NFKB1_HUMAN.H11MO.1.B', win=30, data_dir=None):
+	if data_dir == None:
+		data_dir = overcast_data_dir
+	if not (os.path.exists(data_dir) and os.path.isdir(data_dir)):
+		print("OVERCAST_data folder can not found!")
+		exit(1)
+	sets_files = []
+	for set1 in set_names:
+		sets_file = data_dir + '/TF-target_sets/' + set1 + '.txt'
+		if not os.path.exists(sets_file):
+			print(sets_file, 'not exist!')
+			exit(1)
+		else:
+			sets_files.append(sets_file)
+	degl_file = list_file
+	if not os.path.exists(degl_file):
+		print(degl_file, 'not exist!')
+		exit(1)
+
+	tfbs_array = read_gene_sets(sets_files)
+	deg_list = read_DEG_list(degl_file)
+
+	tf_n = len(tfbs_array)
+	epsilon = 1e-15   # 常用值
+	for i in range(tf_n):
+		if tfbs_array[i][0] == tf:
+			plot_contour_plot(deg_list, tfbs_array[i][1], win)
+			break
+
 if __name__ == '__main__':
 	'''
-	result = olcr(set_names=['human_hocomoco_CLink_wtcoor_1w_0.8', 'human_jaspar_CLink_wtcoor_1w_0.8'], list_file='OVERCAST_data/input_deg-list.txt', win=30, thread_n=16)
+	result = olcr(set_names=['human_jaspar_CLink_rci4_1w_0.8', 'human_hocomoco_CLink_rci4_1w_0.8'], list_file='OVERCAST_data/input_deg-list.txt', win=30, thread_n=16)
 	result.to_csv('output.txt', sep='\t', index=False, encoding='utf-8')
 	
-	result = olcr_bootstrap(set_names=['human_hocomoco_CLink_wtcoor_1w_0.8', 'human_jaspar_CLink_wtcoor_1w_0.8'], list_file='OVERCAST_data/input_deg-list.txt', win=30, thread_n=16)
+	result = olcr_bootstrap(set_names=['human_jaspar_CLink_rci4_1w_0.8', 'human_hocomoco_CLink_rci4_1w_0.8'], list_file='OVERCAST_data/input_deg-list.txt', win=30, thread_n=16)
 	result.to_csv('output_bootstrap.txt', sep='\t', index=False, encoding='utf-8')
 	'''
 
-	#plot_olc(set_names=['human_hocomoco_CLink_wtcoor_1w_0.8', 'human_jaspar_CLink_wtcoor_1w_0.8'], list_file='OVERCAST_data/input_deg-list.txt', tf='MA0844.2_XBP1', win=30)
+	#plot_olc(set_names=['human_jaspar_CLink_rci4_1w_0.8', 'human_hocomoco_CLink_rci4_1w_0.8'], list_file='OVERCAST_data/input_deg-list.txt', tf='MA0844.2_XBP1', win=30)
+
+	#plot_contour(set_names=['human_jaspar_CLink_rci4_1w_0.8', 'human_hocomoco_CLink_rci4_1w_0.8'], list_file='OVERCAST_data/input_deg-list.txt', tf='MA0844.2_XBP1', win=30)
 
-	#plot_fit3D(set_names=['human_hocomoco_CLink_wtcoor_1w_0.8', 'human_jaspar_CLink_wtcoor_1w_0.8'], list_file='OVERCAST_data/input_deg-list.txt', tf='MA0844.2_XBP1', win=30)
+	#plot_fit3D(set_names=['human_jaspar_CLink_rci4_1w_0.8', 'human_hocomoco_CLink_rci4_1w_0.8'], list_file='OVERCAST_data/input_deg-list.txt', tf='MA0844.2_XBP1', win=30)

{pyovercast-1.0.3 → pyovercast-1.0.5}/PyOVERCAST.egg-info/PKG-INFO

@@ -1,6 +1,6 @@
 Metadata-Version: 2.4
 Name: PyOVERCAST
-Version: 1.0.3
+Version: 1.0.5
 Summary: A Python package for mining key transcription factors from transcriptome data.
 Author-email: Tinghua Huang <thua45@126.com>
 License-Expression: MIT
@@ -65,25 +65,28 @@ if __name__ == '__main__':
 	print(sets_names)
 
 	# list avaiable TFs
-	tfs_codes = clinks.get_tfs(set_name='human_hocomoco_CLink_wtcoor_1w_0.8')
+	tfs_codes = clinks.get_tfs(set_name='human_hocomoco_CLink_rci4_1w_0.8')
 	print(tfs_codes)
 
 	# get targets
-	targets = clinks.get_targets(set_name='human_hocomoco_CLink_wtcoor_1w_0.8', tf='NFKB1_HUMAN.H11MO.1.B')
+	targets = clinks.get_targets(set_name='human_hocomoco_CLink_rci4_1w_0.8', tf='NFKB1_HUMAN.H11MO.1.B')
 	print(targets)
 	
 	# predict one DEG-list
-	result = predict.olcr(set_names=['human_hocomoco_CLink_wtcoor_1w_0.8', 'human_jaspar_CLink_wtcoor_1w_0.8'], list_file='./PyOVERCAST_data/input_deg-list.txt', win=30, thread_n=16)
+	result = predict.olcr(set_names=['human_jaspar_CLink_rci4_1w_0.8', 'human_hocomoco_CLink_rci4_1w_0.8'], list_file='./PyOVERCAST_data/input_deg-list.txt', win=30, thread_n=16)
 
 	# or predict one DEG-list with bootstrap
-	result = predict.olcr_bootstrap(set_names=['human_hocomoco_CLink_wtcoor_1w_0.8', 'human_jaspar_CLink_wtcoor_1w_0.8'], list_file='./OVERCAST_data/input_deg-list.txt', win=30, bs_n=1000, thread_n=32)
+	result = predict.olcr_bootstrap(set_names=['human_jaspar_CLink_rci4_1w_0.8', 'human_hocomoco_CLink_rci4_1w_0.8'], list_file='./OVERCAST_data/input_deg-list.txt', win=30, bs_n=1000, thread_n=32)
 
 	# save result to text file
 	result.to_csv('output.txt', sep='\t', index=False, encoding='utf-8')
 	
 	# plot OLC matrix
-	predict.plot_olc(set_names=['human_hocomoco_CLink_wtcoor_1w_0.8', 'human_jaspar_CLink_wtcoor_1w_0.8'], list_file='./PyOVERCAST_data/input_deg-list.txt', tf='MA0844.2_XBP1', win=30)
+	predict.plot_olc(set_names=['human_jaspar_CLink_rci4_1w_0.8', 'human_hocomoco_CLink_rci4_1w_0.8'], list_file='./PyOVERCAST_data/input_deg-list.txt', tf='MA0844.2_XBP1', win=30)
+
+	# plot OLC matrix
+	predict.plot_olc(set_names=['human_jaspar_CLink_rci4_1w_0.8', 'human_hocomoco_CLink_rci4_1w_0.8'], list_file='./PyOVERCAST_data/input_deg-list.txt', tf='MA0844.2_XBP1', win=30)
 
 	# plot fitted 3D U-surface
-	predict.plot_fit3D(set_names=['human_hocomoco_CLink_wtcoor_1w_0.8', 'human_jaspar_CLink_wtcoor_1w_0.8'], list_file='./PyOVERCAST_data/input_deg-list.txt', tf='MA0844.2_XBP1', win=30)
+	predict.plot_contour(set_names=['human_jaspar_CLink_rci4_1w_0.8', 'human_hocomoco_CLink_rci4_1w_0.8'], list_file='./PyOVERCAST_data/input_deg-list.txt', tf='MA0844.2_XBP1', win=30)
 ```

pyovercast-1.0.5/README.md

@@ -0,0 +1,49 @@
+# PyOVERCAST
+
+A Python package for mining key transcription factors from transcriptome data.
+
+## Installation
+
+```bash
+pip install numpy pandas statsmodels scipy seaborn matplotlib PyOVERCAST
+```
+
+## Usage
+
+```python
+from PyOVERCAST import clinks, predict
+
+if __name__ == '__main__':
+	# download TF-target set, only need to run once!!!
+	clinks.download_data()
+
+	# list available TF-target set
+	sets_names = clinks.get_sets(species='Homo sapiens')
+	print(sets_names)
+
+	# list avaiable TFs
+	tfs_codes = clinks.get_tfs(set_name='human_hocomoco_CLink_rci4_1w_0.8')
+	print(tfs_codes)
+
+	# get targets
+	targets = clinks.get_targets(set_name='human_hocomoco_CLink_rci4_1w_0.8', tf='NFKB1_HUMAN.H11MO.1.B')
+	print(targets)
+	
+	# predict one DEG-list
+	result = predict.olcr(set_names=['human_jaspar_CLink_rci4_1w_0.8', 'human_hocomoco_CLink_rci4_1w_0.8'], list_file='./PyOVERCAST_data/input_deg-list.txt', win=30, thread_n=16)
+
+	# or predict one DEG-list with bootstrap
+	result = predict.olcr_bootstrap(set_names=['human_jaspar_CLink_rci4_1w_0.8', 'human_hocomoco_CLink_rci4_1w_0.8'], list_file='./OVERCAST_data/input_deg-list.txt', win=30, bs_n=1000, thread_n=32)
+
+	# save result to text file
+	result.to_csv('output.txt', sep='\t', index=False, encoding='utf-8')
+	
+	# plot OLC matrix
+	predict.plot_olc(set_names=['human_jaspar_CLink_rci4_1w_0.8', 'human_hocomoco_CLink_rci4_1w_0.8'], list_file='./PyOVERCAST_data/input_deg-list.txt', tf='MA0844.2_XBP1', win=30)
+
+	# plot OLC matrix
+	predict.plot_olc(set_names=['human_jaspar_CLink_rci4_1w_0.8', 'human_hocomoco_CLink_rci4_1w_0.8'], list_file='./PyOVERCAST_data/input_deg-list.txt', tf='MA0844.2_XBP1', win=30)
+
+	# plot fitted 3D U-surface
+	predict.plot_contour(set_names=['human_jaspar_CLink_rci4_1w_0.8', 'human_hocomoco_CLink_rci4_1w_0.8'], list_file='./PyOVERCAST_data/input_deg-list.txt', tf='MA0844.2_XBP1', win=30)
+```

{pyovercast-1.0.3 → pyovercast-1.0.5}/pyproject.toml

@@ -4,7 +4,7 @@ build-backend = "setuptools.build_meta"
 
 [project]
 name = "PyOVERCAST"
-version = "1.0.3"
+version = "1.0.5"
 license = "MIT"  # SPDX expression
 description = "A Python package for mining key transcription factors from transcriptome data."
 readme = "README.md"

pyovercast-1.0.3/README.md

@@ -1,46 +0,0 @@
-# PyOVERCAST
-
-A Python package for mining key transcription factors from transcriptome data.
-
-## Installation
-
-```bash
-pip install numpy pandas statsmodels scipy seaborn matplotlib PyOVERCAST
-```
-
-## Usage
-
-```python
-from PyOVERCAST import clinks, predict
-
-if __name__ == '__main__':
-	# download TF-target set, only need to run once!!!
-	clinks.download_data()
-
-	# list available TF-target set
-	sets_names = clinks.get_sets(species='Homo sapiens')
-	print(sets_names)
-
-	# list avaiable TFs
-	tfs_codes = clinks.get_tfs(set_name='human_hocomoco_CLink_wtcoor_1w_0.8')
-	print(tfs_codes)
-
-	# get targets
-	targets = clinks.get_targets(set_name='human_hocomoco_CLink_wtcoor_1w_0.8', tf='NFKB1_HUMAN.H11MO.1.B')
-	print(targets)
-	
-	# predict one DEG-list
-	result = predict.olcr(set_names=['human_hocomoco_CLink_wtcoor_1w_0.8', 'human_jaspar_CLink_wtcoor_1w_0.8'], list_file='./PyOVERCAST_data/input_deg-list.txt', win=30, thread_n=16)
-
-	# or predict one DEG-list with bootstrap
-	result = predict.olcr_bootstrap(set_names=['human_hocomoco_CLink_wtcoor_1w_0.8', 'human_jaspar_CLink_wtcoor_1w_0.8'], list_file='./OVERCAST_data/input_deg-list.txt', win=30, bs_n=1000, thread_n=32)
-
-	# save result to text file
-	result.to_csv('output.txt', sep='\t', index=False, encoding='utf-8')
-	
-	# plot OLC matrix
-	predict.plot_olc(set_names=['human_hocomoco_CLink_wtcoor_1w_0.8', 'human_jaspar_CLink_wtcoor_1w_0.8'], list_file='./PyOVERCAST_data/input_deg-list.txt', tf='MA0844.2_XBP1', win=30)
-
-	# plot fitted 3D U-surface
-	predict.plot_fit3D(set_names=['human_hocomoco_CLink_wtcoor_1w_0.8', 'human_jaspar_CLink_wtcoor_1w_0.8'], list_file='./PyOVERCAST_data/input_deg-list.txt', tf='MA0844.2_XBP1', win=30)
-```