-
Notifications
You must be signed in to change notification settings - Fork 15
/
Copy path04-CNVanalysis.Rmd
297 lines (193 loc) · 10.1 KB
/
04-CNVanalysis.Rmd
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
61
62
63
64
65
66
67
68
69
70
71
72
73
74
75
76
77
78
79
80
81
82
83
84
85
86
87
88
89
90
91
92
93
94
95
96
97
98
99
100
101
102
103
104
105
106
107
108
109
110
111
112
113
114
115
116
117
118
119
120
121
122
123
124
125
126
127
128
129
130
131
132
133
134
135
136
137
138
139
140
141
142
143
144
145
146
147
148
149
150
151
152
153
154
155
156
157
158
159
160
161
162
163
164
165
166
167
168
169
170
171
172
173
174
175
176
177
178
179
180
181
182
183
184
185
186
187
188
189
190
191
192
193
194
195
196
197
198
199
200
201
202
203
204
205
206
207
208
209
210
211
212
213
214
215
216
217
218
219
220
221
222
223
224
225
226
227
228
229
230
231
232
233
234
235
236
237
238
239
240
241
242
243
244
245
246
247
248
249
250
251
252
253
254
255
256
257
258
259
260
261
262
263
264
265
266
267
268
269
270
271
272
273
274
275
276
277
278
279
280
281
282
283
284
285
286
287
288
289
290
291
292
293
294
# Copy number variation estimation from scRNA-seq { #CNV_analysis }
## Introduction
Copy number variation is a major mutations in many tumors. Recently,
[Minussi et al, 2021](https://www.nature.com/articles/s41586-021-03357-x)
suggested novel evolutionary patterns through analyzing CNV in breast tumors.
There are three types of CNVs:
- Copy gain
- Copy loss
- Loss of heterozygosity
They some may not affect the total copies, e.g., Loss of heterozygosity, while
some may not affect the balance of two allele, e.g., copy gain (2 vs 2).
There are four major methods for CNV analyis in scRNA-seq:
- [inferCNV](https://github.com/broadinstitute/infercnv)
- [CopyKat](https://github.com/navinlabcode/copykat) by [Gao et al 2020](https://www.nature.com/articles/s41587-020-00795-2)
- [CaSpER](https://github.com/akdess/CaSpER) by [Harmanci et al, 2020](https://www.nature.com/articles/s41467-019-13779-x)
- [HoneyBadger](https://jef.works/HoneyBADGER/) by [Fan et al](https://genome.cshlp.org/content/early/2018/06/13/gr.228080.117)
Here, we will illustration how to use the method `inferCNV` for analysing CNV in
single-cell RNA-seq data. We will discuss other methods in the end.
## inferCNV and example
[InferCNV: Inferring copy number alterations from tumor single cell RNA-Seq data](https://github.com/broadinstitute/inferCNV/wiki)
### install inferCNV
**Software Requirements**
- [JAGS](https://mcmc-jags.sourceforge.io/)
- R (>3.6)
In order to run infercnv, JAGS (Just Another Gibbs Sampler) must be installed.
Download JAGS from <https://sourceforge.net/projects/mcmc-jags/files/JAGS/4.x/> and install JAGS in your environment (windows/MAC).
If you use inferCNV on server, install JAGS via `conda install` in your conda environment is recommended.
```
conda install -c conda-forge jags
```
More details refer to [inferCNV wiki page](https://github.com/broadinstitute/inferCNV/wiki)
**Five options for installing inferCNV**
Option A: Install infercnv from BioConductor (preferred)
```
if (!requireNamespace("BiocManager", quietly = TRUE))
install.packages("BiocManager")
BiocManager::install("infercnv")
```
For more other options, refer to [Five options for installing inferCNV](https://github.com/broadinstitute/inferCNV/wiki/Installing-infercnv)
**Data requirements**
- 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
[File-Definitions](https://github.com/broadinstitute/inferCNV/wiki/File-Definitions)
### getting started
If you have installed infercnv from BioConductor, you can run the example data with:
```
library(infercnv)
infercnv_obj = CreateInfercnvObject(raw_counts_matrix=system.file("extdata", "oligodendroglioma_expression_downsampled.counts.matrix.gz", package = "infercnv"),
annotations_file=system.file("extdata", "oligodendroglioma_annotations_downsampled.txt", package = "infercnv"),
delim="\t",
gene_order_file=system.file("extdata", "gencode_downsampled.EXAMPLE_ONLY_DONT_REUSE.txt", package = "infercnv"),
ref_group_names=c("Microglia/Macrophage","Oligodendrocytes (non-malignant)"))
infercnv_obj = infercnv::run(infercnv_obj,
cutoff=1, # cutoff=1 works well for Smart-seq2, and cutoff=0.1 works well for 10x Genomics
out_dir=tempfile(),
cluster_by_groups=TRUE,
denoise=TRUE,
HMM=TRUE)
```
If you can run the getting started part with demo data provided by inferCNV, then it is installed successfully.
**Demo Example Figure**
{width=85%}
## Application on TNBC1
### data description
TNBC1 is a triple negative breast cancer tumor sample of high tumor purity (72.6%) with 796 single tumor cells and 301 normal cells. The dataset is available on NCBI GEO under the accession number GSM4476486.
: Details of TNBC1 dataset (from published articles, copyKAT).
+-------------------------------+--------------------+-----------------------+------------------------------------+
| TNBC1 | Number of clones | Number of tumor clones| Tumor clone-specific copy gain |
+===============================+====================+=======================+====================================+
| Triple negative breast cancer | 3 | 2 | - C1: 4p, 7q, 9, 17q |
| | | | - C2: 3p, 6q, 7p, 11q, X |
+-------------------------------+--------------------+-----------------------+------------------------------------+
- Expression
Clone A Clone B Normal
------ ------ ------
488 307 302
Table: Subclusters of TNBC1 dataset (from gene expression analysis-Seurat).
Notes: 488,307, 247, 55
{width=50%}
- B Allele Frenquency (BAF)
{width=120%}
{width=120%}
{width=120%}
- BAF V.S. Expression
{width=50%}
### run inferCNV
[data_download](https://sourceforge.net/projects/sgcellworkshop/files/CNV_analysis/CNV_analysis_data.zip/download)
[demo1_log_file](https://github.com/Rongtingting/SingleCell-Workshop-2021/blob/master/CNV_analysis_file/demo1_inferCNV_TNBC1.log)
[demo2_log_file](https://github.com/Rongtingting/SingleCell-Workshop-2021/blob/master/CNV_analysis_file/demo2_inferCNV_TNBC1.log)
[output_files](https://github.com/Rongtingting/SingleCell-Workshop-2021/blob/master/CNV_analysis_file/demo2_outputfiles.lst)
```{r}
# library(infercnv)
# library(utils)
# library (BiocGenerics)
## DEMO1
#
# tnbc <- read.delim("C://Users/Rongting/Documents/GitHub_repos/combinedTNBC1.txt")
# anno <- tnbc[2,]
# anno <- t(anno)
# anno <- as.data.frame(anno)
#
# gex <- tnbc[-c(1:2),]
# gex <- type.convert(gex)
#
# gene_file <- "C://Users/Rongting/Documents/GitHub_repos/gene_note_noheader_unique.txt"
#
# infercnv_obj = CreateInfercnvObject(raw_counts_matrix=gex,
# annotations_file=anno,
# delim='\t',
# gene_order_file=gene_file,
# ref_group_names= "N")
# output = "C://Users/Rongting/Documents/GitHub_repos/tnbc1_demo"
# infercnv_obj = infercnv::run(infercnv_obj,
# cutoff=0.1,
# out_dir= output ,
# cluster_by_groups=T,
# denoise=T,
# HMM=T)
## DEMO2
# gene_file <- "C://Users/Rongting/Documents/GitHub_repos/gene_note_noheader_unique.txt"
#
# anno_file <- 'C://Users/Rongting/Documents/GitHub_repos/tnbc-3cluster-id.txt'
#
# infercnv_obj2 = CreateInfercnvObject(raw_counts_matrix=gex,
# annotations_file=anno_file,
# delim='\t',
# gene_order_file=gene_file,
# ref_group_names= "Normal")
#
# output = "C://Users/Rongting/Documents/GitHub_repos/tnbc1_demo2"
#
# infercnv_obj2 = infercnv::run(infercnv_obj2,
# cutoff=0.1,
# out_dir= output,
# cluster_by_groups=T,
# denoise=T,
# HMM=T)
```
```
#################
##Notes
#################
## load the package
library(Seurat)
library(infercnv)
## prepare the data (cellranger output)
### load count matrix (example)
matrix_path <- "../cellranger/xxxx/count_xxxxx/outs/filtered_gene_bc_matrices/GRCh38/"
### read count matrix
gex_mtx <- Seurat::Read10X(data.dir = matrix_path)
### run inferCNV with loop
celltype = c('CloneA', 'CloneB', 'Normal')
for (i in celltype){
infercnv_obj1 = CreateInfercnvObject(raw_counts_matrix=gex_mtx,
annotations_file=anno_file,
delim='\t',
gene_order_file=gene_file,
ref_group_names=c(i))
output <- paste0('/groups/cgsd/rthuang/processed_data/inferCNV/xxxx/','xxxx_', i)
infercnv_obj1 = infercnv::run(infercnv_obj1,
cutoff=0.1,
out_dir= output ,
cluster_by_groups=T,
denoise=T,
HMM=T)
}
```
### inferCNV result
- demo1
{width=85%}
{width=85%}
- demo2
{width=85%}
{width=85%}
## Last notes
There are four major methods for CNV analyis in scRNA-seq:
[inferCNV](https://github.com/broadinstitute/infercnv),
[CopyKat](https://github.com/navinlabcode/copykat),
[CaSpER](https://github.com/akdess/CaSpER), and
[HoneyBadger](https://jef.works/HoneyBADGER/)
However, the two BAF-supporting methods HoneyBadger and Casper works less
accurately from our experience. The expression-only method InferCNV and CopyKat
generally works fine, though it has no capability to detect loss of
heterozygosity. In general, we suggest perform a pseudo-bulk BAF on the
identified clones or cell clusters with copy number variations.
Feel free to get in touch ([email protected]) if you need advice on CNV analysis
on your own data / project.
**ref**
https://www.r-bloggers.com/2012/04/getting-started-with-jags-rjags-and-bayesian-modelling/
```
bookdown::render_book("index.Rmd", "bookdown::gitbook")
```