Reference

Install

Install
一定要先装JAGS，我在conda环境中安装了（conda install r-rjags）再BiocManager::install("infercnv")

Input （Data requirements）

具体说明在：(https://github.com/broadinstitute/inferCNV/wiki/File-Definitions).
我用了注释过的seuratobject。

a raw counts matrix of single-cell RNA-Seq expression
an annotations file which indicates which cells are tumor vs. normal.
a gene/chromosome positions file :

#https://github.com/broadinstitute/inferCNV/tree/master/scripts
#[infercnv/scripts/gtf_to_position_file.py at master · broadinstitute/infercnv (github.com)](https://github.com/broadinstitute/infercnv/blob/master/scripts/gtf_to_position_file.py)
#[data.broadinstitute.org/Trinity/CTAT/cnv/hg38_gencode_v27.txt](https://data.broadinstitute.org/Trinity/CTAT/cnv/hg38_gencode_v27.txt)

# By Default use gene_id as the name of your feature
python ./scripts/gtf_to_position_file.py your_reference.gtf your_gen_pos.txt

# You can change what gtf attribute key is used, here transcript_id is used.
python ./scripts/gtf_to_position_file.py --attribute_name transcript_id your_reference.gtf your_gen_pos.txt

Usage

两步：先输入三个文件构建对象CreateInfercnvObject；再运行infercnv::run(infercnv_obj,...)。

# create the infercnv object
infercnv_obj = CreateInfercnvObject(raw_counts_matrix="singleCell.counts.matrix",
                                    annotations_file="cellAnnotations.txt",
                                    delim="\t",
                                    gene_order_file="gene_ordering_file.txt",
                                    ref_group_names=c("normal"))

# perform infercnv operations to reveal cnv signal
infercnv_obj = infercnv::run(infercnv_obj,
                             cutoff=1,  # use 1 for smart-seq, 0.1 for 10x-genomics
                             out_dir="output_dir",  # dir is auto-created for storing outputs
                             cluster_by_groups=T,   # cluster
                             denoise=T,
                             HMM=T
                             )

从SeuratObject 开始构建对象

刚开始用自己的数据集一直卡死，服务器内存不够，还是subset细胞类型了。

library(Seurat)
library(infercnv)

#查看数据
seurat_object <- readRDS('seurat_object.rds')
levels(seurat_object)
#筛选分析需要的细胞类型
seurat_object <- subset(seurat_object, idents=c('Epithelial cells', 'Myeloid cells', 'T cells'))
#抽样，仅用于该教程
#seurat_object <- subset(seurat_object, downsample=200)
counts <- GetAssayData(seurat_object, y= 'counts')
anno <- data.frame(Idents(seurat_object))

infercnv_obj = CreateInfercnvObject(raw_counts_matrix = counts,
                                        annotations_file = anno,
                                        delim="\t",
                                        gene_order_file = gene_order,
                                        min_max_counts_per_cell = c(100, +Inf),
                                        ref_group_names = c("Myeloid cells", "T cells"))
。。。