你真的需要如此多的单细胞亚群注释⼯具吗
Lazy learning methods include CELLBLAST , scmap-cell , CellFishing.jl , and
CellAtlasSearch .
Eager learning methods account for the majority of the automatic methods, including
scHPL , clustifyr , MARS , scPretrain , Superscan , Seurat , , scLearn , scCapsNet ,
ACTINN , CaSTLe , CHETAH , SciBet , scID , scmap-cluster , scPred , SingleCellNet ,欲望春天
SingleR , scVI , scMatch , scClassifR , and Garnett .
Marker learning methods include scTyper , DigitalCellSorter , SCINA , SCSA ,
CellAssign , and scCATCH . MarkerCount
写松树的作文
To facilitate automatic cell-type identification, scLearn, CELLBLAST, SciBet,很简单的手工
SingleCellNet, scMatch, Superscan, and Garnett provide procesd training datats.
Moreover, DigitalCellSorter, SCSA, scTyper, and scCATCH provide canonical cell
markers for certain cell types.
作者开发了⼀个整合这么多⼯具的包（AutomaticCellTypeIdentification），主要是把各个⼯
具分成了3类：
eagersupervid methods include ACTINN, CaSTLe, CHETAH, clustifyr, Garnett,
Markercount, MARS, scClassifR, scHPL, SciBet, scID, scLearn, scmapcluster, scPred,
scVI, Seurat, SingleCellNet and SingleR.lazysupervid methods include CELLBLAST and scmapcell.markersupervid methods include scTyper, Markercount, SCSA,
DigitalCellSorter and SCINA.
⼯作量有点⼤啊！
不过，综述⽂章关于软件⼯具算法测评的思路值得学习：
Fig. 1. Workflow of the traditional and automatic cell-type identification methods.
Fig. 2. Performance of the automatic cell-type identification methods using the Tabula
Muris datats.
中国古建筑博物馆Fig. 3. Performance of the automatic cell-type identification methods using PBMC and
tumor datats.
Fig. 4. Speed of automatic cell-type identification methods.
Fig. 5. Summary of performance of the automatic cell-type identification methods. Bar
graphs of the automatic cell-type identification methods with six evaluation criteria
indicated.
⽂章也提到了⽬前单细胞转录组测序数据都是多个样品了，所以确实存在两个难题（Yet, for integrated datats, there are still two issues to be solved.）：
The first is to try to avoid the influences of different quencing technologies during the
process of data integration, for example, by using MNN , CCA , LIGER , Scanorama , et al.
The cond is to try to unify the currently inconsistent annotation levels in the training
datats, for example, by the joint usage of multiple training datats , or by manual curation of each training datat.
出生的反义词
实际上我做的⼤量肿瘤单细胞数据分析项⽬⾥⾯，⽤不到这些⾃动化注释⼯具，都是⾃⼰⾁眼看，需要有⼀些背景知识哦！⽐如背诵如下所⽰各个细胞亚群⾼表达量基因的列表：
# T Cells (CD3D, CD3E, CD8A), # B cells (CD19, CD79A, MS4A1 [CD20]), # Plasma cells ( IGHG1, MZB1, SDC1, CD79A), # Monocytes and macrophages (CD68, CD163, CD14),# N K Cells (FGFBP2, FCG3RA, CX3CR1), # Photoreceptor cells (RCVRN), # Fibroblasts (FG F7, MME), # Endothelial cells (PECAM1, VWF). # epi or tumor (EPCAM, KRT19, PROM1, A LDH1A1, CD24).# immune (CD45+,PTPRC), epithelial/cancer (EpCAM+,EPCAM), # strom al (CD10+,MME,fibo or CD31+,PECAM1,endo)
最后，摘抄了这个综述⽂章⾥⾯收集整理的各个⼯具的详细GitHub⽹页链接：
Name of method Version URLCELLBLAST v0.3.8 /gao-
lab/Cell_BLASTCellFishing.jl v0.3.2 /bicycle1885/CellFishing.jlscmap-cell v1.6.0 ht
tps:///hemberg-
lab/scmapACTINN master /mafeiyang/ACTINNCaSTLe v1.0.0.2 /yuvallb/CaSTLeCHETAH v1.2.0 /jdekanter/CHETAHGarnett v0.
姜黄的功效1.19 /cole-trapnell-lab/garnettSciBet v0.1.0 /zwj-
隋朝灭亡的原因tina/scibetRscID v2.1 /BatadaLab/scIDscLearn v1.0 /bm2 -lab/scLearnscmap-cluster v1.6.0 /hemberg-
lab/scmapscPred v1.9.0 /powellgenomicslab/scPredscVI v0.4.1 /YofLab/scvi-
toolsSeurat v3.2.2 /satijalab/uratSingleCellNet v0.1.0 / pcahan1/singleCellNetSingleR v1.1.1 /dviraran/SingleRCellAssign v0.99.
21 /Irrationone/cellassignDigitalCellSorter v1.1 /sdoman skyi/DigitalCellSorterSCINA v1.2.0 /jcao89757/SCINASCSA master https: ///bioinfo-ibms-
pumc/SCSAscTyper v0.1.0 /omicsCore/scTyperscHPL V0.0.2 /lcmmichieln/scHPLMARS master /snap-
stanford/marsclustifyr v1.5.0 /rnabioco/clustifyrscClassifR v1.1.1 /grisslab/scClassifRMarkerCount master /combio-
dku/MarkerCount/tree/master
⼊门单细胞数据处理，需要⼀些基础认知，也可以看基础10讲：
赞美老师的歌曲01. 上游分析流程
02.课题多少个样品，测序数据量如何
03. 过滤不合格细胞和基因（数据质控很重要）
04. 过滤线粒体核糖体基因
05. 去除细胞效应和基因效应
06.单细胞转录组数据的降维聚类分群
07.单细胞转录组数据处理之细胞亚群注释
08.把拿到的亚群进⾏更细致的分群
09.单细胞转录组数据处理之细胞亚群⽐例⽐较
最基础的往往是降维聚类分群，参考前⾯的例⼦：⼈⼈都能学会的单细胞聚类分群注释
