# 安装必要的包importsubprocessimportsys def install_packages(packages): subprocess.check_call([sys.executable,"-m","pip","install", packages])# 检查并且安装必要的包required_packages=["requests","numpy ","pandas","matplotlib.pyplot","seaborn"]forpackageinrequired_packages: try: __import__(package)print(f"Yes!{package} 已经安装")except ImportError: print(f"正在安装{package}...")install_packages(package)print(f"Yes!{package}安装完成")importrequestsimportnumpy as npimportpandas as pdimportseaborn as snsimportmatplotlib.pyplot as plt from sklearn.preprocessingimportStandardScaler# 生成示例数据np.random.seed(42)# 假设有20个基因，10个样本genes=[f'Gene_{i+1}'foriinrange(20)]samples=[f'Sample_{chr(65+i)}'foriinrange(10)]# 生成表达量数据（模拟log2表达量）data=np.random.randn(20,10)*2+5# 添加一些差异表达模式data[:5, :5]+=3# 前5个基因在前5个样本中高表达data[10:15,5:]+=2# 中间5个基因在后5个样本中高表达# 创建DataFramedf=pd.DataFrame(data,index=genes,columns=samples)# 方法1：使用seaborn绘制基础热图plt.figure(figsize=(12,8))# 可以对数据进行标准化（Z-score标准化）df_zscore=pd.DataFrame(StandardScaler().fit_transform(df.T).T,index=df.index,columns=df.columns)# 绘制热图ax=sns.heatmap(df_zscore,cmap='RdBu_r',# 颜色映射，'_r'表示反转center=0,# 中心点对应0值square=False,# 单元格是否设为正方形cbar_kws={"label":"Z-score"},linewidths=0.5,linecolor='gray')plt.title('Gene expression heatmap',fontsize=16,fontweight='bold')plt.xlabel('Sample',fontsize=12)plt.ylabel('Gene',fontsize=12)plt.xticks(rotation=45,ha='right')plt.tight_layout()plt.show()save_path="F:/360MoveData/Users/admin/Desktop/Python_project/heatmap/1.png"plt.savefig(save_path)
![]()
