PyOVERCAST 1.0.0__tar.gz

pyovercast-1.0.0/PKG-INFO

@@ -0,0 +1,86 @@
+Metadata-Version: 2.4
+Name: PyOVERCAST
+Version: 1.0.0
+Summary: A Python package for mining key transcription factors from transcriptome data.
+Author-email: Tinghua Huang <thua45@126.com>
+License-Expression: MIT
+Project-URL: Homepage, https://github.com/thua45/PyOVERCAST
+Project-URL: Documentation, https://github.com/thua45/PyOVERCAST#readme
+Project-URL: Repository, https://github.com/thua45/PyOVERCAST
+Project-URL: Issues, https://github.com/thua45/PyOVERCAST/issues
+Keywords: PyOVERCAST,Transcription Factor,Binding Site,Transcriptome Data
+Classifier: Development Status :: 4 - Beta
+Classifier: Intended Audience :: Developers
+Classifier: Programming Language :: Python :: 3.11
+Classifier: Programming Language :: Python :: 3.12
+Classifier: Programming Language :: Python :: 3.13
+Requires-Python: >=3.11
+Description-Content-Type: text/markdown
+License-File: LICENSE
+Requires-Dist: numpy==1.26; sys_platform == "darwin"
+Requires-Dist: numpy>=2.0; sys_platform == "win32"
+Requires-Dist: numpy>=2.0; sys_platform == "linux"
+Requires-Dist: pandas==3.0; sys_platform == "darwin"
+Requires-Dist: pandas>=3.0; sys_platform == "win32"
+Requires-Dist: pandas>=3.0; sys_platform == "linux"
+Requires-Dist: statsmodels==0.14; sys_platform == "darwin"
+Requires-Dist: statsmodels>=0.14; sys_platform == "win32"
+Requires-Dist: statsmodels>=0.14; sys_platform == "linux"
+Requires-Dist: seaborn==0.13; sys_platform == "darwin"
+Requires-Dist: seaborn>=0.13; sys_platform == "win32"
+Requires-Dist: seaborn>=0.13; sys_platform == "linux"
+Requires-Dist: matplotlib==3.11; sys_platform == "darwin"
+Requires-Dist: matplotlib>=3.11; sys_platform == "win32"
+Requires-Dist: matplotlib>=3.11; sys_platform == "linux"
+Provides-Extra: dev
+Requires-Dist: matplotlib; extra == "dev"
+Provides-Extra: test
+Requires-Dist: matplotlib; extra == "test"
+Dynamic: license-file
+
+# PyOVERCAST
+
+A Python package for mining key transcription factors from transcriptome data.
+
+## Installation
+
+```bash
+pip install numpy pandas statsmodels 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='./PyOVERCAST_data/input_deg-list.txt', win=30, thread_n=16)
+
+	# save result to text file
+	result.to_csv('output.txt', sep='\t', index=False, encoding='utf-8-sig')
+	
+	# 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)
+```

pyovercast-1.0.0/PyOVERCAST/clinks.py

@@ -0,0 +1,177 @@
+# from importlib import resources
+import urllib.request
+import zipfile
+import os
+from pathlib import Path
+
+overcast_data_dir = './OVERCAST_data'
+
+def download_with_progress(url, filename=None):
+	"""
+	带进度显示的下载函数
+	"""
+	def progress_callback(block_num, block_size, total_size):
+		"""
+		进度回调函数
+		
+		Args:
+			block_num: 当前已下载的块数量
+			block_size: 每个块的大小（字节）
+			total_size: 文件总大小（字节）
+		"""
+		downloaded = block_num * block_size
+		percent = downloaded / total_size * 100 if total_size > 0 else 0
+		
+		# 限制显示频率，避免刷新太快
+		if block_num % 100 == 0 or downloaded >= total_size:
+			print(f"\rProgress: {percent:.1f}% ({downloaded}/{total_size} bytes)", end='', flush=True)
+	
+	# 如果没有指定文件名，从URL中提取
+	if filename is None:
+		filename = os.path.basename(url)
+	
+	print(f"start download: {url}")
+	print(f"save as: {filename}")
+	
+	try:
+		urllib.request.urlretrieve(url, filename, progress_callback)
+		print("donwload sucessful")
+	except Exception as e:
+		print(f"donwload failed: {e}")
+		exit(1)
+
+# 使用示例
+# download_with_progress("https://example.com/largefile.zip", "downloaded_file.zip")
+
+def extractall_individual_files(zip_path, extract_path):
+	"""
+	使用 extract() 方法逐个解压文件，确保覆盖
+	"""
+	os.makedirs(extract_path, exist_ok=True)
+	
+	with zipfile.ZipFile(zip_path, 'r') as zip_ref:
+		# 获取所有成员信息
+		members = zip_ref.infolist()
+		
+		print(f"extracting {len(members)} file...")
+		
+		for i, member in enumerate(members, 1):
+			try:
+				# 直接使用 extract() 方法，它会处理覆盖
+				zip_ref.extract(member, extract_path)
+				
+				if i % 10 == 0 or i == len(members):
+					print(f"progress: {i}/{len(members)}")
+					
+			except Exception as e:
+				print(f"extracting failed {member.filename}: {e}")
+				exit(1)
+	
+	print("extracting finished")
+
+# 使用示例
+# extractall_individual_files("example.zip", "extracted_files")
+
+def download_and_extract_zip(url, extract_to):
+	"""
+	下载ZIP文件并解压到指定目录
+	
+	Args:
+		url: ZIP文件的URL
+		extract_to: 解压目标目录
+	"""
+	# 创建目标目录
+	Path(extract_to).mkdir(parents=True, exist_ok=True)
+	
+	# 临时ZIP文件路径
+	zip_path = os.path.join(extract_to, "temp.zip")
+	
+	try:
+		# 下载文件
+		# print(f"downloading: {url}")
+		# urllib.request.urlretrieve(url, zip_path)
+		download_with_progress(url, zip_path)
+		# print("donwload finished")
+		
+		# 解压文件
+		# print(f"extracting: {extract_to}")
+		# with zipfile.ZipFile(zip_path, 'r') as zip_ref:
+		# 	zip_ref.extractall(extract_to, overwrite=True)
+		extractall_individual_files(zip_path, extract_to)
+		# print("extract finished")
+		
+	except Exception as e:
+		print(f"错误: {e}")
+		return False
+	finally:
+		# 清理临时ZIP文件
+		if os.path.exists(zip_path):
+			os.remove(zip_path)
+			print("clearing temp files")
+	
+	return True
+
+def download_data(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(data_dir, 'not exist, creating...')
+		folder_path = Path(data_dir)
+		folder_path.mkdir(exist_ok=True)
+	url = "http://www.thua45.cn/overcast/OVERCAST_data.zip"
+	extract_dir = data_dir
+	# extract_dir = "./downloaded_content"
+	success = download_and_extract_zip(url, extract_dir)
+	if success:
+		print("models installed successful")
+
+def get_sets(species=None, data_dir=None):
+	if data_dir == None:
+		data_dir = overcast_data_dir
+	index_file = data_dir + '/sets_index.txt'
+	if not os.path.exists(index_file):
+		print('sets_index.txt not exist, you may need to run download_data() first')
+		exit(1)
+	header = ''
+	result = []
+	for line in open(index_file, 'r'):
+		if line[0] == "#":
+			header = line.rstrip()
+			continue
+		lblocks = line.rstrip().split('\t')
+		if species != None and lblocks[1] != species:
+			continue
+		result.append(lblocks[0])
+	return result
+
+def get_tfs(set_name=None, data_dir=None):
+	if data_dir == None:
+		data_dir = overcast_data_dir
+	sets_file = data_dir + '/TF-target_sets/' + set_name + '.txt'
+	if not os.path.exists(sets_file):
+		print('sets_file not exist, you may need to run download_data() first')
+		exit(1)
+	result = set()
+	for line in open(sets_file, 'r'):
+		if line[0] == "#":
+			continue
+		lblocks = line.rstrip().split('\t')
+		result.add(lblocks[0])
+	return list(result)
+
+def get_targets(set_name=None, tf=None, data_dir=None):
+	if data_dir == None:
+		data_dir = overcast_data_dir
+	sets_file = data_dir + '/TF-target_sets/' + set_name + '.txt'
+	if not os.path.exists(sets_file):
+		print('sets_file not exist, you may need to run download_data() first')
+		exit(1)
+	result = set()
+	for line in open(sets_file, 'r'):
+		if line[0] == "#":
+			continue
+		lblocks = line.rstrip().split('\t')
+		if tf != None and lblocks[0] != tf:
+			continue
+		result.add(lblocks[1])
+	return list(result)

pyovercast-1.0.0/PyOVERCAST/predict.py

@@ -0,0 +1,631 @@
+#!/usr/bin/env python
+import os
+import math
+import time
+from collections import defaultdict
+from operator import itemgetter
+import numpy as np
+import pandas as pd
+import matplotlib.pyplot as plt
+import statsmodels.api as sm
+# 设置中文字体，避免 DejaVu Sans 缺失中文字符
+plt.rcParams['font.sans-serif'] = ['SimHei', 'Microsoft YaHei', 'PingFang SC', 'Noto Sans CJK SC', 'Arial Unicode MS']  # 指定默认字体
+plt.rcParams['axes.unicode_minus'] = False  # 解决负号显示问题
+import multiprocessing as mp
+import seaborn as sns
+
+overcast_data_dir = './OVERCAST_data'
+
+def read_tfbs_file(file, tfbs):
+	for line in open(file, 'r'):
+		if line[0] == '#':
+			continue
+		lblocks = line.rstrip().split('\t')
+		tfbs[lblocks[0]]['gene'].append(lblocks[1])
+		tfbs[lblocks[0]]['LogFC'].append(float(lblocks[2]))
+
+def read_gene_sets(files):
+	tfbs = defaultdict(lambda: defaultdict(list))
+	for file in files:
+		read_tfbs_file(file, tfbs)
+	tfbs_array = []
+	tfs = list(tfbs.keys())
+	for tf in tfs:
+		df = pd.DataFrame({"gene": tfbs[tf]['gene'], "logFC": tfbs[tf]['LogFC']}, index=tfbs[tf]['gene'])
+		tfbs_array.append([tf, df])
+	return tfbs_array
+
+def read_DEG_list(file_in):
+	genes = []
+	LogFCs = []
+	#file_in = 'XBP1_DEG-list.txt'
+	for line in open(file_in, 'r'):
+		if line[0] == '#':
+			continue
+		lblocks = line.rstrip().split('\t')
+		genes.append(lblocks[0])
+		LogFCs.append(float(lblocks[1]))
+	df = pd.DataFrame({"gene": genes, "logFC": LogFCs}, index=genes)
+	return df
+
+def slide_window_inclusive(lst, w, s):
+	lst_splits = []
+	l_idxs = []
+	n = len(lst)
+	for start in range(0, n, s):
+		lst_splits.append(lst[start:min(start + w, n)])
+		l_idxs.append(start + math.floor(min(w, n - start) / 2.0))
+	return l_idxs, lst_splits
+
+def rrho_matrix_str(list1, list2, window=20, universe=None):
+	# unique genes preserving order
+	list1 = list(dict.fromkeys([str(x) for x in list1]))
+	list2 = list(dict.fromkeys([str(x) for x in list2]))
+
+	l_idxs1, lst_splits1 = slide_window_inclusive(list1, math.floor(len(list1) / window), math.floor(len(list1) / window / 3.0))
+	l_idxs2, lst_splits2 = slide_window_inclusive(list2, math.floor(len(list2) / window), math.floor(len(list2) / window / 3.0))
+	step = math.floor(window / 3)
+
+	mat = np.zeros((len(l_idxs1), len(l_idxs2)), dtype=float)
+	# print(mat.shape)
+	mi = -1
+	for lst_i in lst_splits1:
+		mi += 1
+		mj = -1
+		for lst_j in lst_splits2:
+			mj += 1
+			mat[mi, mj] = len(set(lst_i) & set(lst_j))
+
+	# mat = mat - np.mean(mat)
+	rrho_df = pd.DataFrame(mat, index=l_idxs1, columns=l_idxs2)
+	rrho_df.index.name = "Prefix_list1"
+	rrho_df.columns.name = "Prefix_list2"
+	return rrho_df
+
+def Z_to_frame(Z):
+	# ============================
+	# 1. 生成模拟数据（网格点云）
+	# ============================
+	# 假设正方形边长为2，范围[-1, 1]，生成20x20网格
+	np.random.seed(42)  # 固定随机种子，保证可复现
+	x = np.linspace(-1, 1, Z.shape[1])
+	y = np.linspace(-1, 1, Z.shape[0])
+	X, Y = np.meshgrid(x, y)
+
+	# 展平成一维数组，便于回归
+	df = pd.DataFrame({
+		'x': X.ravel(),
+		'y': Y.ravel(),
+		'z': Z.ravel()
+	})
+
+	return df
+
+def compute_dtci_core(df, x_col='x', y_col='y', z_col='z',
+					  c=1.0, alpha=1.0, beta_w=0.5, gamma=1.0):
+	"""
+	核心DTCI计算 —— 使用 statsmodels.OLS
+
+	拟合模型: z = β0 + β1·x + β2·y + β3·(x·y) + ε
+
+	参数:
+		df: DataFrame，包含 x, y, z 列
+		x_col, y_col, z_col: 列名
+		c: 深度调节系数
+		alpha, beta_w, gamma: DI, SI, QI 的权重指数
+
+	返回:
+		DTCI, beta_hat, R2_adj, p_beta3, residuals
+	"""
+	x = df[x_col].values
+	y = df[y_col].values
+	z = df[z_col].values
+	xy = x * y
+
+	# 使用 statsmodels.OLS（自动处理常数项、标准误和p值）
+	X_design = sm.add_constant(pd.DataFrame({'x': x, 'y': y, 'xy': xy}))
+	model = sm.OLS(z, X_design).fit()
+
+	# 提取系数 [const, x, y, xy]
+	beta_hat = model.params.values
+	beta0, beta1, beta2, beta3 = beta_hat
+
+	# 模型质量
+	R2 = model.rsquared
+	R2_adj = model.rsquared_adj
+
+	# 系数显著性（statsmodels 自动计算，更准确）
+	p_values = model.pvalues.values
+
+	# 数据波动
+	z_std = np.std(z, ddof=1)
+
+	# 子指标
+	DI = np.abs(beta3) / (np.abs(beta3) + c * z_std)
+	SI = min(1.0, -np.log10(max(p_values[3], 1e-10)) / 25.0)
+	QI = max(0.0, R2_adj)
+
+	total_weight = alpha + beta_w + gamma
+	DTCI = (DI**alpha * SI**beta_w * QI**gamma) ** (1.0 / total_weight)
+
+	return DTCI, beta_hat, R2_adj, p_values[3], model.resid
+
+def bootstrap_parametric_h0(df, n_bootstrap=2000, random_state=42):
+	"""
+	参数化Bootstrap：在H0（无对角线趋势，即beta3=0）下生成数据
+
+	步骤:
+		1. 在H0下拟合模型：z = beta0 + beta1·x + beta2·y + eps
+		2. 获取拟合值和残差
+		3. 从残差中有放回抽样，构造新数据
+		4. 对新数据计算DTCI
+		5. 比较观测DTCI与零分布，计算p值
+	"""
+	np.random.seed(random_state)
+	n = len(df)
+	x = df['x'].values
+	y = df['y'].values
+	z = df['z'].values
+
+	# 步骤1：在H0下拟合（不含交互项）
+	X_h0 = sm.add_constant(pd.DataFrame({'x': x, 'y': y}))
+	model_h0 = sm.OLS(z, X_h0).fit()
+	z_pred_h0 = model_h0.fittedvalues
+	resid_h0 = model_h0.resid
+
+	# 观测DTCI（全模型）
+	# dtci_obs = dtci_compute_dtci_core(df)[0]
+	dtci_obs, beta_hat, dtci_pval = itemgetter(0, 1, 3)(compute_dtci_core(df))
+
+	# 步骤3-4：生成零分布
+	dtci_null = []
+	for i in range(n_bootstrap):
+		resid_boot = np.random.choice(resid_h0, size=n, replace=True)
+		z_boot = z_pred_h0 + resid_boot
+
+		df_boot = pd.DataFrame({'x': x, 'y': y, 'z': z_boot})
+		try:
+			dtci_i, _, _, _, _ = compute_dtci_core(df_boot)
+			dtci_null.append(dtci_i)
+		except:
+			pass
+
+	dtci_null = np.array(dtci_null)
+
+	# 步骤5：计算p值 (H1: DTCI > 0)
+	p_value = np.mean(dtci_null >= dtci_obs)
+
+	return {
+		'method': '参数化Bootstrap (H0: beta3=0)',
+		'dtci_obs': dtci_obs,
+		'dtci_pval': dtci_pval,
+		'beta3': beta_hat[3],
+		'dtci_null': dtci_null,
+		'p_value': p_value,
+		'se_null': np.std(dtci_null, ddof=1),
+		'mean_null': np.mean(dtci_null),
+		'n_boot': len(dtci_null)
+	}
+
+def rrho_coor(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:
+		return 0.0, 1.0, 0.0
+
+	#beta_xy, pval_xy = td_coor(rrho_df.to_numpy())
+	df = Z_to_frame(rrho_df.to_numpy())
+	DTCI, betas, pval_xy = itemgetter(0, 1, 3)(compute_dtci_core(df))
+
+	return betas[3], pval_xy, DTCI
+
+def plot_rrho(rrho_df, out_file=None, title="RRLO Matrix", cmap="magma"):
+	"""
+	Plot RRHO heatmap.
+	"""
+	plt.figure(figsize=(6, 5), dpi=150)
+	ax = sns.heatmap(
+		rrho_df,
+		cmap=cmap,
+		#cbar_kws={"label": r"$-\log_{10}(p)$"},
+		cbar_kws={"label": r"Overlaps"},
+		xticklabels=max(1, len(rrho_df.columns) // 10),
+		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_title(title, fontsize=10)
+	plt.tight_layout()
+
+	if out_file:
+		plt.savefig(out_file + '.pdf', dpi=300, bbox_inches="tight")
+		plt.savefig(out_file + '.svg', dpi=300, bbox_inches="tight")
+	plt.show()
+
+def plot_olc_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)
+	plot_rrho(rrho_df)
+
+def plot_3D_surface(heatmap_a, out_file=None):
+	row_n, col_n = heatmap_a.shape
+	x = np.linspace(-1, 1, col_n)
+	y = np.linspace(-1, 1, row_n)
+	X, Y = np.meshgrid(x, y)
+	Z_data = heatmap_a
+
+	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 = fit_surface(Z_data.T, X, Y)
+
+	fig = plt.figure(figsize=(6, 4), dpi=150)
+	ax = fig.add_subplot(111, projection='3d')
+
+	color = '#e94560' #['#e94560', '#0f3460', '#0f3460']
+
+	ax.plot_surface(X, Y, Z_fit, cmap='plasma', alpha=0.8)
+
+	ax.tick_params(axis='x', pad=-5)   # x轴标签向外偏移5个点
+	ax.tick_params(axis='y', pad=-5)   # y轴标签向外偏移5个点
+	ax.tick_params(axis='z', pad=0)   # z轴标签向外偏移5个点
+	ax.set_xlabel('x (Rank A)', fontsize=12, labelpad=-5)
+	ax.set_ylabel('y (Rank B)', fontsize=12, labelpad=-5)
+	ax.set_zlabel('z (Overlaps)', fontsize=12, labelpad=0)
+
+	#title = '$z=\\beta_0+\\beta_1x+\\beta_2y+\\beta_3(xy)+\\epsilon$'
+	#ax.set_title(title, fontsize=5, fontweight='bold', color='#16213e', pad=1)
+	idx = 0
+	ax.view_init(elev=20, azim=-55 + idx*12)
+	
+	#beta_text = f'$\\beta_0$={beta['const']:.2f}\n$\\beta_1$={beta['x']:.2f}\n$\\beta_2$={beta['y']:.2f}\n$\\beta_3$={beta['xy']:.2f}'
+	#ax.text2D(-0.13, 0.98, beta_text, transform=ax.transAxes, fontsize=4,
+	#		 verticalalignment='top', bbox=dict(boxstyle='round', facecolor='white', alpha=0.8, edgecolor=color, linewidth=0.4))
+	
+	#ax.set_xlim(-1, 1)
+	#ax.set_ylim(-1, 1)
+	#ax.set_zlim(-3, 8)
+	#ax.tick_params(labelsize=3, pad=0)
+	# 反转 x 轴
+	#ax.invert_yaxis()
+	ax.view_init(elev=20, azim=-145)
+	plt.tight_layout()
+
+	if out_file:
+		plt.savefig(out_file + '.pdf', dpi=300, bbox_inches="tight")
+		plt.savefig(out_file + '.svg', dpi=300, bbox_inches="tight")
+	plt.show()
+
+def plot_fit3D_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_3D_surface(rrho_df.to_numpy())
+
+def rrho_coor_bs(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, 1.0
+	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:
+		return 0.0, 1.0, 0.0, 1.0
+
+	#beta_xy, pval_xy = td_coor(rrho_df.to_numpy())
+	df = Z_to_frame(rrho_df.to_numpy())
+	dtci_res = bootstrap_parametric_h0(df)
+	beta_xy = dtci_res['beta3']
+	pval_xy = dtci_res['dtci_pval']
+	DTCI = dtci_res['dtci_obs']
+	bs_pval = dtci_res['p_value']
+
+	return beta_xy, pval_xy, DTCI, bs_pval
+
+def thread_one(tfbs_array, deg_list, win, shared_list, nn, lock):
+	tf_n = len(tfbs_array)
+	epsilon = 1e-15   # 常用值
+	for i in range(tf_n):
+		# fin_n += 1
+		beta_xy, pval_xy, DTCI = rrho_coor(deg_list, tfbs_array[i][1], win)
+		#score = math.pow(abs(beta_xy), 1.0) * math.log10(pval_xy + epsilon) * -1.0
+		#if math.isnan(score):
+		#	score = 0
+		with lock:
+			shared_list.append([tfbs_array[i][0], beta_xy, pval_xy, DTCI])
+			nn[1] += 1
+			print(str(nn[1]) + ' / ' + str(nn[0]))
+
+def thread_one_bs(tfbs_array, deg_list, win, shared_list, nn, lock):
+	tf_n = len(tfbs_array)
+	epsilon = 1e-15   # 常用值
+	for i in range(tf_n):
+		# fin_n += 1
+		beta_xy, pval_xy, DTCI, bs_pval = rrho_coor_bs(deg_list, tfbs_array[i][1], win)
+		#score = math.pow(abs(beta_xy), 1.0) * math.log10(pval_xy + epsilon) * -1.0
+		#if math.isnan(score):
+		#	score = 0
+		with lock:
+			shared_list.append([tfbs_array[i][0], beta_xy, pval_xy, DTCI, bs_pval])
+			nn[1] += 1
+			print(str(nn[1]) + ' / ' + str(nn[0]))
+
+def split_into_n_even(lst, n):
+	nn = len(lst)
+	base = nn // n		  # 每份基础大小
+	remainder = nn % n	  # 余数，前remainder份各多1个
+	
+	result = []
+	start = 0
+	for i in range(n):
+		size = base + (1 if i < remainder else 0)
+		result.append(lst[start:start+size])
+		start += size
+	return result
+
+def olcr(set_names=None, list_file=None, win=30, thread_n=16, 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)
+	
+	manager = mp.Manager()
+	shared_list = manager.list([])  # 支持任意类型
+	nn = manager.list([len(tfbs_array), 0])  # 支持任意类型
+	lock = manager.Lock()
+
+	tfbs_array_splits = split_into_n_even(tfbs_array, thread_n)
+	task_args = []
+	for ti in range(thread_n):
+		task_args.append((tfbs_array_splits[ti], deg_list, win, shared_list, nn, lock))
+	
+	# 创建进程池
+	with mp.Pool(processes=thread_n) as pool:
+		print(f"thrend_n:", thread_n)
+		print(f"total {len(task_args)} tasks")
+		# starmap 会自动分配任务到进程池
+		start_time = time.time()
+		results = pool.starmap(thread_one, task_args)
+		elapsed = time.time() - start_time
+	# 打印结果
+	print("time consumed:", elapsed)
+
+	sorted_results = sorted(shared_list, key=itemgetter(-1), reverse=True)
+	results = []
+	ln = len(sorted_results)
+	for li in range(len(sorted_results)):
+		ratio = (li + 1) / ln
+		rline = sorted_results[li] + [li + 1, ratio]
+		results.append(rline)
+	col_names = ['TF_motif', 'Beta3', 'Beta3_Pvalue', 'DTCI', 'Rank', 'Rank / N']
+	df = pd.DataFrame(results, columns=col_names)
+
+	return df
+
+def olcr_bootstrap(set_names=None, list_file=None, win=30, thread_n=16, 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)
+	
+	manager = mp.Manager()
+	shared_list = manager.list([])  # 支持任意类型
+	nn = manager.list([len(tfbs_array), 0])  # 支持任意类型
+	lock = manager.Lock()
+
+	tfbs_array_splits = split_into_n_even(tfbs_array, thread_n)
+	task_args = []
+	for ti in range(thread_n):
+		task_args.append((tfbs_array_splits[ti], deg_list, win, shared_list, nn, lock))
+	
+	# 创建进程池
+	with mp.Pool(processes=thread_n) as pool:
+		print(f"thrend_n:", thread_n)
+		print(f"total {len(task_args)} tasks")
+		# starmap 会自动分配任务到进程池
+		start_time = time.time()
+		results = pool.starmap(thread_one_bs, task_args)
+		elapsed = time.time() - start_time
+	# 打印结果
+	print("time consumed:", elapsed)
+
+	sorted_results = sorted(shared_list, key=lambda x: (x[-1], -x[-2]), reverse=False) # lambda x: (x[4], -x[1]), reverse=False
+	results = []
+	ln = len(sorted_results)
+	for li in range(len(sorted_results)):
+		ratio = (li + 1) / ln
+		rline = sorted_results[li] + [li + 1, ratio]
+		results.append(rline)
+	col_names = ['TF_motif', 'Beta3', 'Beta3_Pvalue', 'DTCI', 'DTCI_Pvalue', 'Rank', 'Rank / N']
+	df = pd.DataFrame(results, columns=col_names)
+
+	return df
+
+def plot_olc(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_olc_plot(deg_list, tfbs_array[i][1], win)
+			break
+
+def plot_fit3D(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_fit3D_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.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.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_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)

pyovercast-1.0.0/PyOVERCAST.egg-info/PKG-INFO

@@ -0,0 +1,86 @@
+Metadata-Version: 2.4
+Name: PyOVERCAST
+Version: 1.0.0
+Summary: A Python package for mining key transcription factors from transcriptome data.
+Author-email: Tinghua Huang <thua45@126.com>
+License-Expression: MIT
+Project-URL: Homepage, https://github.com/thua45/PyOVERCAST
+Project-URL: Documentation, https://github.com/thua45/PyOVERCAST#readme
+Project-URL: Repository, https://github.com/thua45/PyOVERCAST
+Project-URL: Issues, https://github.com/thua45/PyOVERCAST/issues
+Keywords: PyOVERCAST,Transcription Factor,Binding Site,Transcriptome Data
+Classifier: Development Status :: 4 - Beta
+Classifier: Intended Audience :: Developers
+Classifier: Programming Language :: Python :: 3.11
+Classifier: Programming Language :: Python :: 3.12
+Classifier: Programming Language :: Python :: 3.13
+Requires-Python: >=3.11
+Description-Content-Type: text/markdown
+License-File: LICENSE
+Requires-Dist: numpy==1.26; sys_platform == "darwin"
+Requires-Dist: numpy>=2.0; sys_platform == "win32"
+Requires-Dist: numpy>=2.0; sys_platform == "linux"
+Requires-Dist: pandas==3.0; sys_platform == "darwin"
+Requires-Dist: pandas>=3.0; sys_platform == "win32"
+Requires-Dist: pandas>=3.0; sys_platform == "linux"
+Requires-Dist: statsmodels==0.14; sys_platform == "darwin"
+Requires-Dist: statsmodels>=0.14; sys_platform == "win32"
+Requires-Dist: statsmodels>=0.14; sys_platform == "linux"
+Requires-Dist: seaborn==0.13; sys_platform == "darwin"
+Requires-Dist: seaborn>=0.13; sys_platform == "win32"
+Requires-Dist: seaborn>=0.13; sys_platform == "linux"
+Requires-Dist: matplotlib==3.11; sys_platform == "darwin"
+Requires-Dist: matplotlib>=3.11; sys_platform == "win32"
+Requires-Dist: matplotlib>=3.11; sys_platform == "linux"
+Provides-Extra: dev
+Requires-Dist: matplotlib; extra == "dev"
+Provides-Extra: test
+Requires-Dist: matplotlib; extra == "test"
+Dynamic: license-file
+
+# PyOVERCAST
+
+A Python package for mining key transcription factors from transcriptome data.
+
+## Installation
+
+```bash
+pip install numpy pandas statsmodels 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='./PyOVERCAST_data/input_deg-list.txt', win=30, thread_n=16)
+
+	# save result to text file
+	result.to_csv('output.txt', sep='\t', index=False, encoding='utf-8-sig')
+	
+	# 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)
+```

pyovercast-1.0.0/PyOVERCAST.egg-info/SOURCES.txt

@@ -0,0 +1,12 @@
+LICENSE
+README.md
+pyproject.toml
+PyOVERCAST/__init__.py
+PyOVERCAST/clinks.py
+PyOVERCAST/predict.py
+PyOVERCAST.egg-info/PKG-INFO
+PyOVERCAST.egg-info/SOURCES.txt
+PyOVERCAST.egg-info/dependency_links.txt
+PyOVERCAST.egg-info/entry_points.txt
+PyOVERCAST.egg-info/requires.txt
+PyOVERCAST.egg-info/top_level.txt

pyovercast-1.0.0/PyOVERCAST.egg-info/dependency_links.txt

@@ -0,0 +1 @@
+

pyovercast-1.0.0/PyOVERCAST.egg-info/entry_points.txt

@@ -0,0 +1,3 @@
+[console_scripts]
+PyOVERCAST = PyOVERCAST.cli:main
+PyOVERCAST-cleanup = PyOVERCAST.uninstall:cleanup

pyovercast-1.0.0/PyOVERCAST.egg-info/requires.txt

@@ -0,0 +1,27 @@
+
+[:sys_platform == "darwin"]
+numpy==1.26
+pandas==3.0
+statsmodels==0.14
+seaborn==0.13
+matplotlib==3.11
+
+[:sys_platform == "linux"]
+numpy>=2.0
+pandas>=3.0
+statsmodels>=0.14
+seaborn>=0.13
+matplotlib>=3.11
+
+[:sys_platform == "win32"]
+numpy>=2.0
+pandas>=3.0
+statsmodels>=0.14
+seaborn>=0.13
+matplotlib>=3.11
+
+[dev]
+matplotlib
+
+[test]
+matplotlib

pyovercast-1.0.0/PyOVERCAST.egg-info/top_level.txt

@@ -0,0 +1 @@
+PyOVERCAST

pyovercast-1.0.0/README.md

@@ -0,0 +1,46 @@
+# PyOVERCAST
+
+A Python package for mining key transcription factors from transcriptome data.
+
+## Installation
+
+```bash
+pip install numpy pandas statsmodels 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='./PyOVERCAST_data/input_deg-list.txt', win=30, thread_n=16)
+
+	# save result to text file
+	result.to_csv('output.txt', sep='\t', index=False, encoding='utf-8-sig')
+	
+	# 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)
+```

pyovercast-1.0.0/pyproject.toml

@@ -0,0 +1,62 @@
+[build-system]
+requires = ["setuptools>=61.0", "wheel"]
+build-backend = "setuptools.build_meta"
+
+[project]
+name = "PyOVERCAST"
+version = "1.0.0"
+license = "MIT"  # SPDX expression
+description = "A Python package for mining key transcription factors from transcriptome data."
+readme = "README.md"
+requires-python = ">=3.11"
+authors = [
+    {name = "Tinghua Huang", email = "thua45@126.com"}
+]
+keywords = ["PyOVERCAST", "Transcription Factor", "Binding Site", 'Transcriptome Data']
+classifiers = [
+    "Development Status :: 4 - Beta",
+    "Intended Audience :: Developers",
+    "Programming Language :: Python :: 3.11",
+    "Programming Language :: Python :: 3.12",
+    "Programming Language :: Python :: 3.13",
+]
+dependencies = [
+    "numpy==1.26; sys_platform == 'darwin'",
+    "numpy>=2.0; sys_platform == 'win32'",
+    "numpy>=2.0; sys_platform == 'linux'",
+    "pandas==3.0; sys_platform == 'darwin'",
+    "pandas>=3.0; sys_platform == 'win32'",
+    "pandas>=3.0; sys_platform == 'linux'",
+    "statsmodels==0.14; sys_platform == 'darwin'",
+    "statsmodels>=0.14; sys_platform == 'win32'",
+    "statsmodels>=0.14; sys_platform == 'linux'",
+    "seaborn==0.13; sys_platform == 'darwin'",
+    "seaborn>=0.13; sys_platform == 'win32'",
+    "seaborn>=0.13; sys_platform == 'linux'",
+    "matplotlib==3.11; sys_platform == 'darwin'",
+    "matplotlib>=3.11; sys_platform == 'win32'",
+    "matplotlib>=3.11; sys_platform == 'linux'",
+]
+
+[project.urls]
+Homepage = "https://github.com/thua45/PyOVERCAST"
+Documentation = "https://github.com/thua45/PyOVERCAST#readme"
+Repository = "https://github.com/thua45/PyOVERCAST"
+Issues = "https://github.com/thua45/PyOVERCAST/issues"
+
+[project.optional-dependencies]
+dev = [
+    "matplotlib",
+]
+test = [
+    "matplotlib",
+]
+
+[project.scripts]
+PyOVERCAST = "PyOVERCAST.cli:main"
+PyOVERCAST-cleanup = "PyOVERCAST.uninstall:cleanup"
+
+[tool.setuptools]
+packages = ["PyOVERCAST"]
+package-data = {"*" = ["*.txt", "*.md", "*.rst", "data/*"]}
+

pyovercast-1.0.0/setup.cfg

@@ -0,0 +1,4 @@
+[egg_info]
+tag_build = 
+tag_date = 0
+