1.绘制小提琴图 import scanpy as sc data=sc.read_10x_mtx(‘./input_data/text’,var_names=’gene_symbols’) #读取文件

sc.pp.filter_cells(data, min_genes=200)#最少200个基因表达，否则去除该细胞 sc.pp.filter_genes(data, min_cells=3)

data.var[‘mt’] = data.var_names.str.startswith(‘MT-‘) sc.pp.calculate_qc_metrics(data, qc_vars=[‘mt’], percent_top=None, log1p=False, inplace=True)

sc.pl.violin(data, [‘n_genes_by_counts’, ‘total_counts’, ‘pct_counts_mt’], jitter=0.4, multi_panel=True)

print(data)

2UMP降维

import scanpy as sc

data=sc.read_10x_mtx(‘./input_data/text’,var_names=’gene_symbols’) #读取文件

sc.pp.filter_cells(data, min_genes=200)#最少200个基因表达，否则去除该细胞 sc.pp.filter_genes(data, min_cells=3)

data.var[‘mt’] = data.var_names.str.startswith(‘MT-‘) #除去线粒体

sc.pp.normalize_total(data, target_sum=1e4) #数据标准化

sc.pp.log1p(data) #数据对数化

sc.pp.highly_variable_genes(data, min_mean=0.0125, max_mean=3, min_disp=0.5)

sc.tl.pca(data, svd_solver=’arpack’)

计算Computing the neighborhood graph

sc.pp.neighbors(data, n_neighbors=10, n_pcs=40) sc.tl.umap(data)

做出来的图跟教程有些差别

sc.pl.umap(data, color=[‘CST3’, ‘NKG7’, ‘PPBP’]) sc.pl.umap(data, color=[‘CST3’, ‘NKG7’, ‘PPBP’], use_raw=False)

Clustering the neighborhood graph。聚类并着色

这里需要安装pip3 install leidenalg

sc.tl.leiden(data) sc.pl.umap(data, color=[‘leiden’, ‘CST3’, ‘NKG7’])

3基因表达调控绘制 import scanpy as sc data=sc.read_10x_mtx(‘./input_data/text’,var_names=’gene_symbols’) #读取文件 sc.pl.highest_expr_genes(data, n_top=20, ) print(data)

4 线粒体基因和个基因占比 import scanpy as sc data=sc.read_10x_mtx(‘./input_data/text’,var_names=’gene_symbols’) #读取文件

sc.pp.filter_cells(data, min_genes=200)#最少200个基因表达，否则去除该细胞 sc.pp.filter_genes(data, min_cells=3)

data.var[‘mt’] = data.var_names.str.startswith(‘MT-‘) sc.pp.calculate_qc_metrics(data, qc_vars=[‘mt’], percent_top=None, log1p=False, inplace=True)

sc.pl.scatter(data, x=’total_counts’, y=’pct_counts_mt’) sc.pl.scatter(data, x=’total_counts’, y=’n_genes_by_counts’)

print(data) 5高峰表达与肘击图 import scanpy as sc

data=sc.read_10x_mtx(‘./input_data/text’,var_names=’gene_symbols’) #读取文件

sc.pp.filter_cells(data, min_genes=200)#最少200个基因表达，否则去除该细胞 sc.pp.filter_genes(data, min_cells=3)

data.var[‘mt’] = data.var_names.str.startswith(‘MT-‘) #除去线粒体

sc.pp.normalize_total(data, target_sum=1e4) #数据标准化

sc.pp.log1p(data) #数据对数化

sc.pp.highly_variable_genes(data, min_mean=0.0125, max_mean=3, min_disp=0.5) sc.pl.highly_variable_genes(data)

sc.tl.pca(data, svd_solver=’arpack’) sc.pl.pca(data, color=’CST3’) sc.pl.pca_variance_ratio(data, log=True)

print(data)