snATAC_pancan/Integrate_withRNA_v2.0.R at main · allakarpova/snATAC_pancan · GitHub

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
#Author:Nadezhda Terekhanova
####2020-08-27
####This script performs the cell type annotation, using annotated RNA rds-object
####It will transfer the labels from the "predicted_cell_type" or "cell_type" (change accordingly) meta-data field of RNA RDS-object to the ATAC object, and will save it into the out/
####Upd: we use manually annotated RNA-objects, so the cell type annotation is in the "cell_type" meta.data field
### Adapted by Alla. Added parameters and the metadata will be saved.

library(Signac)
library(Seurat)
library(GenomeInfoDb)
library(ggplot2)

require(magrittr)
require(readr)
require(Matrix)
require(tidyr)
require(dplyr)
set.seed(1234)

library(plyr)
library(dplyr)
library(tibble)
library(reshape)
library(plyr)

library(EnsDb.Hsapiens.v86)

library(optparse)
library(data.table)

option_list = list(
  make_option(c("-r", "--rna.obj"),
              type="character",
              default=NULL,
              help="path to rna RDS object",
              metavar="character"),
  make_option(c("-a", "--atac.obj"),
              type="character",
              default=NULL,
              help="path to atac RDS object",
              metavar="character"),
  make_option(c("-o", "--output"),
              type="character",
              default="./",
              help="output folder path",
              metavar="character"),
  make_option(c("-e", "--extra"),
              type="character",
              default="./",
              help="add unique string identifier for your data",
              metavar="character"),
  make_option(c("-c", "--cell.type.column.name"),
              type="character",
              default="peaks",
              help="the name of the column containing cell type info in the RNA object",
              metavar="character")

);

opt_parser = OptionParser(option_list=option_list);
opt = parse_args(opt_parser);

# read in initial arguments
sample_rna <- opt$rna.obj
sample_atac <- opt$atac.obj
out_path <- opt$output
add_filename <- opt$extra
ct.column <- opt$cell.type.column.name


dir.create(out_path, showWarnings = F)
setwd(out_path)


rna=readRDS(sample_rna)
atac=readRDS(sample_atac)

DefaultAssay(atac) <- 'RNA'

atac <- NormalizeData(
  object = atac,
  assay = 'RNA',
  normalization.method = 'LogNormalize',
  scale.factor = median(atac$nCount_RNA)
)

cat('doing FindTransferAnchors\n')
transfer.anchors <- FindTransferAnchors(
  reference = rna,
  query = atac,
  features = VariableFeatures(object = rna),
  reference.assay = "RNA",
  query.assay = "RNA",
  reduction = 'cca'
)

cat('doing TransferData\n')
predicted.labels <- TransferData(
  anchorset = transfer.anchors,
  reference = rna,
  refdata = ct.column,
  weight.reduction = atac[['lsi']],
  dims = 2:30
)

print(head(predicted.labels))

cat('Adding metadata\n')
atac <- AddMetaData(object = atac, metadata = predicted.labels)


cat('saving shit\n')
saveRDS(atac,paste0(add_filename, "_cellTyped.rds"))
fwrite(atac@meta.data, paste0('Metadata_', add_filename, '_snATAC_cell_type.txt'), row.names = T, sep = '\t')

###Making plots:
cat('plotting plot1\n')
plot1 <- DimPlot(
  object = rna,
  group.by = ct.column,
  label = TRUE,
  repel = TRUE) + ggtitle(paste(add_filename,'snRNA-seq'))

cat('plotting plot2\n')
plot2 <- DimPlot(
  object = atac,
  group.by = 'predicted.id',
  label = TRUE,
  repel = TRUE) + ggtitle(paste(add_filename,'snATAC-seq'))

cat('combining plots\n')
pdf(paste0(add_filename,"_snRNA_snATAC_integrated.pdf"),height=6,width=16)
p=CombinePlots(list(plot1,plot2), ncol = 2)
print(p)
dev.off()