一步一个坑：单细胞数据的h5ad格式转换成R可读取对象

library(SeuratDisk)
错误: package or namespace load failed for ‘SeuratDisk’ in loadNamespace(j <- i[[1L]], c(lib.loc, .libPaths()), versionCheck = vI[[j]]):
 不存在叫‘hdf5r’这个名字的程辑包

## 使用西湖大学的 Bioconductor镜像
options(BioC_mirror="https://mirrors.westlake.edu.cn/bioconductor")
options("repos"=c(CRAN="https://mirrors.westlake.edu.cn/CRAN/"))

# 安装
devtools::install_github("hhoeflin/hdf5r")

# h5ad读取方式
# 自己安装  mojaveazure/seurat-disk 这个GitHub包：
# remotes::install_github("mojaveazure/seurat-disk")
library(SeuratDisk)
library(patchwork)
#～～～～～开始读数据～～～～～
## h5ad是python的Scanpy读取文件格式，需要转换
#～～～～读取adipose～～～～

Convert('./GSE222427/GSM6923183_MC_scRNA.h5ad', "h5seurat", overwrite = TRUE,assay = "RNA")

Warning: Unknown file type: h5ad
Creating h5Seurat file for version 3.1.5.9900
Adding X as data
Adding X as counts
Adding meta.features from var
Adding X_pca as cell embeddings for pca
Adding X_pca_harmony as cell embeddings for pca_harmony
Adding X_umap as cell embeddings for umap
Adding miscellaneous information for pca
Adding standard deviations for pca
Adding miscellaneous information for umap
Adding Major_cluster_colors to miscellaneous data
Adding Phenotype_colors to miscellaneous data
Adding donor_id_colors to miscellaneous data
Adding hvg to miscellaneous data
Adding leiden to miscellaneous data
Adding leiden_colors to miscellaneous data
Adding log1p to miscellaneous data
Adding majority_voting_colors to miscellaneous data
Adding predicted_labels_NC2022_colors to miscellaneous data
Adding predicted_labels_colors to miscellaneous data
Adding rank_genes_groups to miscellaneous data
Adding layer ambiguous as data in assay ambiguous
Adding layer ambiguous as counts in assay ambiguous
Adding layer counts as data in assay counts
Adding layer counts as counts in assay counts
Adding layer matrix as data in assay matrix
Adding layer matrix as counts in assay matrix
Adding layer spliced as data in assay spliced
Adding layer spliced as counts in assay spliced
Adding layer unspliced as data in assay unspliced
Adding layer unspliced as counts in assay unspliced

sce.all <- LoadH5Seurat("./GSE222427/GSM6923183_MC_scRNA.h5seurat")

Validating h5Seurat file
错误: Ambiguous assays

Validating h5Seurat file
Initializing RNA with data
Adding counts for RNA
Adding feature-level metadata for RNA
错误: Missing required datasets 'levels' and 'values'

library("Seurat")
library("anndata")
print("Convert from Scanpy to Seurat...")
data <- read_h5ad("example.hd5ad")
data <- CreateSeuratObject(counts = t(data$X), meta.data = data$obs)
print(str(data))

# 其他方法：R代码
# install.packages("anndata")
library(sceasy)
library(reticulate)
library(Seurat)
# v4
packageVersion("Seurat")
# [1] ‘4.4.0’
library(BiocParallel)
register(MulticoreParam(workers = 4, progressbar = TRUE)) 
use_condaenv(condaenv = '/nas2/zhangj/biosoft/miniconda3/envs/sc')
# h5ad转为Seurat
sceasy::convertFormat(obj = "GSE222427/GSM6923183_MC_scRNA.h5ad", from="anndata",to="seurat",outFile = 'scRNA.rds')

# 
# X -> counts
# An object of class Seurat
# 20320 features across 53748 samples within 1 assay
# Active assay: RNA (20320 features, 0 variable features)
# 2 layers present: counts, data
# 3 dimensional reductions calculated: pca, pca_harmony, umap

sce.all <- readRDS("scRNA.rds")
sce.all

# 查看特征
as.data.frame(sce.all@assays$RNA$counts[1:10, 1:2])
head(sce.all@meta.data, 10)
colnames(sce.all@meta.data)
table(sce.all$donor_id) 
table(sce.all$batch) 
sce.all$orig.ident <- sce.all$donor_id

An object of class Seurat 
20320 features across 53748 samples within 1 assay 
Active assay: RNA (20320 features, 0 variable features)
 2 layers present: counts, data
 3 dimensional reductions calculated: pca, pca_harmony, umap

# 变成v5
packageVersion("Seurat")
# [1] ‘5.1.0’
sce.all <- readRDS("scRNA.rds")
sce.all

sce.all_v5 <- CreateSeuratObject(counts = sce.all@assays$RNA@counts, meta.data = sce.all@meta.data)
sce.all_v5
sce.all_v5$orig.ident <- sce.all_v5$donor_id
head(sce.all_v5@meta.data)

library(qs)
qsave(sce.all_v5, file="GSE222427/sce.all_v5.qs")

# python中导出数据
import scipy.sparse as sparse
import scipy.io as sio
import scipy.stats as stats
import numpy as np
import scanpy as sc
import os

all_data = sc.read_h5ad("./GSE222427/GSM6923183_MC_scRNA.h5ad")
all_data

all_data.var.head()
cellinfo = all_data.obs
cellinfo = all_data.obs
geneinfo = all_data.var
mtx=all_data.X

# 保存
cellinfo.to_csv("cellinfo.csv")
geneinfo.to_csv("geneinfo.csv")
sio.mmwrite("sparse_matrix.mtx",mtx)

library(Seurat)
library(clustree)
library(cowplot)
library(data.table)
library(ggplot2)
library(patchwork)
library(stringr)
library(qs)
library(Matrix)

mtx <- readMM( "sparse_matrix.mtx" )
mtx[1:4,1:4]
dim(mtx)

cl <- fread( "cellinfo.csv", header = T,data.table = F ) 
head(cl)
dim(cl)

rl <- fread( "geneinfo.csv", header = T,data.table = F ) 
head(rl)
dim(rl)

rownames(mtx) <- cl$V1
colnames(mtx) <- rl$V1 #paste0('c',cl$V2)
mtx[1:4,1:4]
mtx <- t(mtx)
dim(mtx)

meta <- cl
rownames(meta) <- cl$V1
identical(rownames(meta),colnames(mtx))

library(Seurat)
sce.all <- CreateSeuratObject(counts = mtx,  meta.data = meta, min.cells = 5)
as.data.frame(sce.all@assays$RNA$counts[1:10, 1:2])
head(sce.all@meta.data, 10)