安装 CellTypist

• 配置环境

conda create -n celltypist -c conda-forge r-base=4.1.2
conda activate celltypist
conda install -c conda-forge r-seurat=4.1.0 -y
conda install -c conda-forge r-irkernel=1.3 -y
IRkernel::installspec(name='celltypist', displayname='r-celltypist')
conda install -c conda-forge scanpy=1.8.2 -y
/opt/conda/envs/celltypist/bin/pip3 install celltypist -i https://pypi.tuna.tsinghua.edu.cn/simple
conda install -c conda-forge r-reticulate=1.24 -y
python3
import celltypist
celltypist.models.download_models(force_update = False)

加载包

Sys.setenv(RETICULATE_PYTHON = "/opt/conda/envs/celltypist/bin/python3.8")
library(reticulate)
scanpy = import("scanpy")
celltypist = import("celltypist")
pandas <- import("pandas")
numpy = import("numpy")
py_config()

加载数据

library(Seurat)
sce <- readRDS("~/upload/yy_zhang_data/scRNA-seq/pca.celltype.rds")
Myeloid <- subset(sce, cell_type=='Myeloid')

seurat 转 scanpy

# 数据矩阵, scanpy与Seurat的行列定义相反
adata.X = numpy$array(t(as.matrix(Myeloid[['RNA']]@counts)))
# 对每个细胞的观察
adata.obs = pandas$DataFrame(Myeloid@meta.data[colnames(Myeloid[['RNA']]@counts),])
# 对基因矩阵的注释
adata.var = pandas$DataFrame(data.frame(gene = rownames(Myeloid[['RNA']]@counts), row.names = rownames(Myeloid[['RNA']]@counts)))
 
# 组装AnnData对象
adata = scanpy$AnnData(X = adata.X, obs=adata.obs, var=adata.var)

进行预测

model = celltypist$models$Model$load(model = 'Immune_All_AddPIP.pkl')
model$cell_types
scanpy$pp$normalize_total(adata, target_sum=1e4)
scanpy$pp$log1p(adata)
predictions = celltypist$annotate(adata, model = 'Immune_All_AddPIP.pkl', majority_voting = T)

预测添加到 seurat 对象

Myeloid = AddMetaData(Myeloid, predictions$predicted_labels)