如何按顺序绘制DataFrame指定列对的散点图及Pearson相关系数

核心代码实现 #

假设你已导入所需库，直接用以下代码遍历对应列对完成绘图和系数计算：

import pandas as pd
import matplotlib.pyplot as plt
from scipy.stats import pearsonr

# 定义要配对的列组
npm_col = ["snv", "het-hom", "ti-tv"]
drg_col = ["snv-drg", "het-hom-drg", "ti-tv-drg"]

# 遍历每一组对应列
for npm_col_name, drg_col_name in zip(npm_col, drg_col):
    # 提取数据并处理缺失值（避免计算和绘图报错）
    x_data = merged_dft[npm_col_name].dropna()
    y_data = merged_dft[drg_col_name].dropna()
    
    # 确保两组数据长度一致（同步删除缺失值）
    common_indices = x_data.index.intersection(y_data.index)
    x = x_data.loc[common_indices]
    y = y_data.loc[common_indices]
    
    # 计算Pearson相关系数和p值
    corr_coef, p_value = pearsonr(x, y)
    
    # 绘制散点图
    plt.figure(figsize=(8, 6))
    plt.scatter(x, y, alpha=0.6, color="#1f77b4")
    plt.title(f"{npm_col_name} vs {drg_col_name}\nPearson相关系数: {corr_coef:.3f}, p值: {p_value:.3e}")
    plt.xlabel(npm_col_name)
    plt.ylabel(drg_col_name)
    plt.grid(True, linestyle="--", alpha=0.7)
    plt.show()

常见问题修正点 #

1. 列配对错误：必须用zip()函数将npm_col和drg_col按顺序一一配对，避免列对应关系混乱
2. 缺失值处理：如果数据包含NaN，必须同步清理两组数据的缺失值（用index.intersection确保数据对齐），否则pearsonr会返回无效结果
3. KeyError：检查merged_dft中是否存在所有指定的列名，确保列名拼写完全一致

可选：多子图布局 #

如果需要将三个散点图放在同一画布上，用以下代码：

fig, axes = plt.subplots(1, 3, figsize=(22, 6))

for idx, (npm_col_name, drg_col_name) in enumerate(zip(npm_col, drg_col)):
    x_data = merged_dft[npm_col_name].dropna()
    y_data = merged_dft[drg_col_name].dropna()
    common_indices = x_data.index.intersection(y_data.index)
    x = x_data.loc[common_indices]
    y = y_data.loc[common_indices]
    
    corr_coef, p_value = pearsonr(x, y)
    
    ax = axes[idx]
    ax.scatter(x, y, alpha=0.6, color="#ff7f0e")
    ax.set_title(f"{npm_col_name}\nvs {drg_col_name}\nCorr: {corr_coef:.3f}")
    ax.set_xlabel(npm_col_name)
    ax.set_ylabel(drg_col_name)
    ax.grid(True, linestyle="--", alpha=0.7)

plt.tight_layout()
plt.show()

内容的提问来源于stack exchange，提问作者user3214212