迪斯科工具包

DISCO 工具包 1.1.0（2024 年 6 月 11 日）

DISCOtoolkit 是一个 R 包，提供对 DISCO 数据库中的数据和工具的访问。其功能包括：

• 根据样本元数据和指定的细胞类型过滤和下载 DISCO 数据
• CELLiD：细胞类型注释
• scEnrichment：使用 DISCO DEG 进行基因组富集

要求

DISCOtoolkit 依赖于以下软件包：

• R (>= 4.0.0)
• PBMC申请
• 纵梁
• jsonlite
• 进步

安装

 devtools :: install_github( " [email protected]:JinmiaoChenLab/DISCOtoolkit.git " )

基本用法

过滤并下载 DISCO 数据

library( DISCOtoolkit )

# find samples from normal lung tissue and sequenced by 10X Genomics 5' platform
# retain samples containing more than 100 Macrophages(or its children)
metadata = FilterDiscoMetadata(
  sample_id = NULL ,
  project_id = NULL ,
  tissue = " lung " ,
  disease = NULL ,
  platform = c( " 10x5' " ),
  sample_type = c( " control " , " adjacent normal " ),
  cell_type = " Macrophage " , 
  cell_type_confidence = " high " , 
  include_cell_type_children = T , 
  min_cell_per_sample = 100
)
# ## print information ###
# Fetching sample metadata
# Filtering sample
# Fetching cell type information
# Fetching ontology from DISCO database
# 18 samples and 64592 cells were found

# download filtered data into 'disco_data' folder
DownloadDiscoData( metadata , output_dir = " disco_data " )

赛利德

library( DISCOtoolkit )
library( Seurat )

metadata = FilterDiscoMetadata(
  sample_id = " ERX2757110 "
)

DownloadDiscoData( metadata , output_dir = " disco_data " )

rna = readRDS( " disco_data/ERX2757110.rds " )
rna = CreateSeuratObject( rna )
rna = NormalizeData( rna )
rna = FindVariableFeatures( rna )
rna = ScaleData( rna )
rna = RunPCA( rna )
rna = FindNeighbors( rna , dims = 1 : 10 )
rna = FindClusters( rna )

rna_average = AverageExpression( rna )
predict_ct = CELLiDCluster( rna = as.matrix( rna_average $ RNA ))


# It will download reference data and differential expression gene (DEG) data from DISCO and save them in the 'DISCOtmp' folder by default. You can reuse this data for subsequent CELLiD analyses as follow:

ref_data = readRDS( " DISCOtmp/ref_data.rds " )
ref_deg = readRDS( " DISCOtmp/ref_deg.rds " )
predict_ct = CELLiDCluster( rna = as.matrix( rna_average $ RNA ), ref_data = ref_data , ref_deg = ref_deg )


rna $ cell_type = predict_ct $ predict_cell_type_1 [as.numeric( rna $ seurat_clusters )]
rna = RunUMAP( rna , dims = 1 : 10 )
DimPlot( rna , group.by = " cell_type " , label = T )

科学富集

 markers = FindMarkers( rna , ident.1 = 0 , only.pos = T , logfc.threshold = 0.5 )
cellid_input = data.frame ( gene = rownames( markers ), logFC = markers $ avg_log2FC )
cellid_res = CELLiDEnrichment( cellid_input )

# also it will download 'ref_geneset.rds' in 'DISCOtmp' folder by default,
# You can reuse this data for subsequent CELLiDEnrichment analyses as follow:
ref = readRDS( " DISCOtmp/ref_geneset.rds " )
cellid_res = CELLiDEnrichment( cellid_input , reference = ref )