-
Notifications
You must be signed in to change notification settings - Fork 1
/
Copy pathcombine.pca.for.umap.r
175 lines (133 loc) · 5.2 KB
/
combine.pca.for.umap.r
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
library(data.table)
library(ggplot2)
library(umap)
setwd("~/SCRATCH/tsne")
# source("./tsne.util.r")
load("./rna/TCGA.rna.filter1.allgene.pca.rda")
pca.rna <- pca
rm(list=c("pca"))
patient.rna <- substr(rownames(pca.rna$x), 1, 12)
load("./mirna/TCGA.mirna.filter1.allgene.pca.rda")
pca.mirna <- pca2
rm(list=c("pca2"))
patient.mirna <- substr(rownames(pca.mirna$x), 1, 12)
load("./meth/TCGA.met.pca.perm100.rda")
pca.met <- pca
rm(list=c("pca"))
patient.met <- substr(rownames(pca.met$x), 1, 12)
load("./cnv/TCGA.cnv.1m.pca.rda")
pca.cnv <- pca2
rm(list=c("pca2"))
patient.cnv <- substr(rownames(pca.cnv$x), 1, 12)
# find common patient
patients <- Reduce(intersect, list(patient.met, patient.rna, patient.mirna, patient.cnv))
rna <- pca.rna$x[match(patients, patient.rna), ]
mirna <- pca.mirna$x[match(patients, patient.mirna), ]
met <- pca.met$x[match(patients, patient.met), ]
cnv <- pca.cnv$x[match(patients, patient.cnv), ]
# merge all PCs
# data <- cbind(rna,
# mirna[match(patients, patient.mirna), ],
# met[match(patients, patient.met), ],
# cnv[match(patients, patient.cnv), ])
# rownames(data) <- patients
# in order to conbine PCs from different dataset, we need to scale the PCs
# we will calculate sqrt(sum of sdev^2), and normalize PCs agaist this factor
sum.var.rna <- sum(apply(rna, 2, sd)^2)
rna <- rna / sqrt(sum.var.rna)
colnames(rna) <- paste0("rna_", colnames(rna))
sum.var.mirna <- sum(apply(mirna, 2, sd)^2)
mirna <- mirna / sqrt(sum.var.mirna)
colnames(mirna) <- paste0("mirna_", colnames(mirna))
sum.var.met <- sum(apply(met, 2, sd)^2)
met <- met / sqrt(sum.var.met)
colnames(met) <- paste0("met_", colnames(met))
sum.var.cnv <- sum(apply(cnv, 2, sd)^2)
cnv <- cnv / sqrt(sum.var.cnv)
colnames(cnv) <- paste0("cnv_", colnames(cnv))
# merge all PCs
data <- cbind(rna, mirna, met, cnv)
rownames(data) <- patients
# UMAP and save results
d <- data.table(patient = patients)
set.seed(1)
umap.rna <- umap(data[, which(grepl("^rna_PC", colnames(data)))])
d$umap.rna.x <- umap.rna$layout[, 1]
d$umap.rna.y <- umap.rna$layout[, 2]
save(umap.rna, file="umap.rna.rda")
set.seed(1)
umap.mirna <- umap(data[, which(grepl("^mirna_PC", colnames(data)))])
d$umap.mirna.x <- umap.mirna$layout[, 1]
d$umap.mirna.y <- umap.mirna$layout[, 2]
save(umap.mirna, file="umap.mirna.rda")
set.seed(1)
umap.met <- umap(data[, which(grepl("^met_PC", colnames(data)))])
d$umap.met.x <- umap.met$layout[, 1]
d$umap.met.y <- umap.met$layout[, 2]
save(umap.met, file="umap.met.rda")
set.seed(1)
umap.cnv <- umap(data[, which(grepl("^cnv_PC", colnames(data)))])
d$umap.cnv.x <- umap.cnv$layout[, 1]
d$umap.cnv.y <- umap.cnv$layout[, 2]
save(umap.cnv, file="umap.cnv.rda")
set.seed(1)
umap.all <- umap(data)
d$umap.all.x <- umap.all$layout[, 1]
d$umap.all.y <- umap.all$layout[, 2]
save(umap.all, file="umap.all.rda")
# add metadata
meta <- fread("all.tsne.data.20170724.txt")
m <- match(d$patient, meta$patient)
d <- cbind(d, meta[m, c("project", "death", "day", "age", "gender", "race")])
write.table(d, "tcga.umap.tsv", col.names=T, row.names=F, sep="\t", quote=F)
png("umap.rna.global.png", width=1280, height=800)
p = ggplot(d, aes(x=umap.rna.x, y=umap.rna.y, col=project, shape = project)) +
geom_point(size = 2) +
scale_shape_manual(values=c(rep(c(0, 2, 8, 16, 17), 6), 0, 2, 8) )
p
dev.off()
png("umap.rna.local.png", width=1280, height=800)
p = ggplot(d[umap.rna.x > -20 & umap.rna.y > -10], aes(x=umap.rna.x, y=umap.rna.y, col=project, shape = project)) +
geom_point(size = 2) +
scale_shape_manual(values=c(rep(c(0, 2, 8, 16, 17), 6), 0, 2, 8) )
p
dev.off()
png("umap.met.global.png", width=1280, height=800)
p = ggplot(d, aes(x=umap.met.x, y=umap.met.y, col=project, shape = project)) +
geom_point(size = 2) +
scale_shape_manual(values=c(rep(c(0, 2, 8, 16, 17), 6), 0, 2, 8) )
p
dev.off()
png("umap.mirna.global.png", width=1280, height=800)
p = ggplot(d, aes(x=umap.mirna.x, y=umap.mirna.y, col=project, shape = project)) +
geom_point(size = 2) +
scale_shape_manual(values=c(rep(c(0, 2, 8, 16, 17), 6), 0, 2, 8) )
p
dev.off()
png("umap.cnv.global.png", width=1280, height=800)
p = ggplot(d, aes(x=umap.cnv.x, y=umap.cnv.y, col=project, shape = project)) +
geom_point(size = 2) +
scale_shape_manual(values=c(rep(c(0, 2, 8, 16, 17), 6), 0, 2, 8) )
p
dev.off()
png("umap.all.global.png", width=1280, height=800)
p = ggplot(d, aes(x=umap.all.x, y=umap.all.y, col=project, shape = project)) +
geom_point(size = 2) +
scale_shape_manual(values=c(rep(c(0, 2, 8, 16, 17), 6), 0, 2, 8) )
p
dev.off()
png("umap.all.hnsc.png", width=1280, height=800)
p = ggplot(d[umap.all.x > -3.5 & umap.all.x < 2.8 & umap.all.y > -4.3 & umap.all.y < 2.4], aes(x=umap.all.x, y=umap.all.y, col=project, shape = project)) +
geom_point(size = 2) +
scale_shape_manual(values=c(rep(c(0, 2, 8, 16, 17), 6), 0, 2, 8) )
p
dev.off()
png("umap.all.hnsc.only.png", width=1280, height=800)
p = ggplot(d[project == "HNSC"], aes(x=umap.all.x, y=umap.all.y, col=project, shape = project)) +
geom_point(size = 2) +
scale_shape_manual(values=c(rep(c(0, 2, 8, 16, 17), 6), 0, 2, 8) )
p
dev.off()
ggplot(d[project == "HNSC"], aes(x=umap.all.x, y=umap.all.y, col=project, shape = project)) +
geom_point(size = 2) +
scale_shape_manual(values=c(rep(c(0, 2, 8, 16, 17), 6), 0, 2, 8))