-
Notifications
You must be signed in to change notification settings - Fork 4
/
figures2and3_code.R
271 lines (221 loc) · 7.67 KB
/
figures2and3_code.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
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
####
library(sva)
library(limma)
library(scales)
library(RColorBrewer)
# this function regresses SVs out of genomic data
cleaningP = function(y, mod, svaobj, P=ncol(mod)) {
X=cbind(mod,svaobj$sv)
Hat=solve(t(X)%*%X)%*%t(X)
beta=(Hat%*%t(y))
cleany=y-t(as.matrix(X[,-c(1:P)])%*%beta[-c(1:P),])
return(cleany)
}
splitit=function(x) split(seq(along=x),x)
#####
# plotting code, expBySample()
# pp is a vector of expression values, or 1 row, of the expression matrix
# sampleIds is a vector of sample IDs
# conditions: colors the samples by treatment or condition, a vector
expBySample = function(pp, sampleIds, geneName, conditions=NULL,
pal="Dark2",legend=TRUE,cex=0.85,...) {
require(RColorBrewer)
sampleIds = factor(sampleIds)
N=length(levels(sampleIds))
if(!is.null(conditions)) {
conditions = factor(conditions)
cols = as.numeric(conditions)
Ncol = max(cols)
} else {
cols = rep(1,length(pp))
Ncol = 1
}
if(pal %in% rownames(brewer.pal.info)) {
palette(brewer.pal(max(Ncol,3), pal))
} else palette(pal)
xx = jitter(as.numeric(sampleIds),amount=0.25)
plot(pp~xx, pch = 21, bg=cols, cex=cex,
xaxt = "n", ylab = "Expression", xlab="",
main = geneName,...)
abline(v=seq(0.5,N+0.5, by=1),lty=2, lwd = 0.8)
axis(1, at = 1:N, labels = levels(sampleIds), las = 2,...)
if(legend) legend("topright", levels(conditions),
col = 1:Ncol,pch=15, nc =Ncol,bty="n",cex=0.9)
}
## load the data
load("normalizedES_GEO.rda")
p = theDataNorm$E
pd = theDataNorm$targets
map = theDataNorm$genes
####
pd$Treatment = factor(pd$Treatment, levels = c("UNDIFF", "FBS", "KSR"))
mod = model.matrix(~Treatment, data =pd)
svaobj = sva(p,mod, n.sv=27) # do SVA, 27 is default
cleanp = cleaningP(p,mod,svaobj) # cleaned data
######
# find treatment differences w and w/o SVA
# fit linear model without SVA
fit0 = lmFit(p, mod)
eb0 = ebayes(fit0) # empirical bayes for t-stats
# with SVA
fit1 = lmFit(p, cbind(mod, svaobj$sv))
eb1 = ebayes(fit1)
### make a new folder for plots
try(system("mkdir treat_fig"))
## other known genes
neuroGeneSymbols = c("TSPAN31", "BBS10", "TARBP1","C17orf69",
"ABI2", "NRCAM", "SCG5", "SMAD7", "CORO2A", "SLC25A24")
mesenGeneSymbols = c("MBNL2", "ZHX1", "TOM1L2", "HGSNAT", "SSFA2",
"FKBP9", "PARVA",
"LRRC8B", "NPM3", "SNURF", "RTN3", "EIF5A2", "CUTC", "SALL2")
neurIndex = match(neuroGeneSymbols, map$GeneSymbol)
mesenIndex = match(mesenGeneSymbols, map$GeneSymbol)
####
# plot treatment effects
oIndex = grep("OLFML1", map$GeneSymbol)
pd$TreatmentColor = factor(pd$Treatment)
pdf("treat_fig/treat_es_noSVA_fig2a_check.pdf")
expBySample(p[oIndex,], pd$SampleID, geneName = "",cex=1,
conditions = as.character(pd$Treatment),
legend = FALSE, cex.axis=1.3, cex.lab=1.5)
dev.off()
pdf("treat_fig/treat_es_SVA_fig2b.pdf")
expBySample(cleanp[oIndex,], pd$SampleID, geneName = "",cex=1,
conditions = as.character(pd$Treatment),
legend = FALSE, cex.axis=1.3, cex.lab=1.5)
legend("bottomright", levels(factor(as.character(pd$Treatment))), bty="n",
col = 1:3, pch = 15, pt.cex = 3,cex=1.6, nc = 3)
dev.off()
plot(eb1$t[neurIndex,3], eb0$t[neurIndex,3],
main = "neuro genes", xlab="SVA", ylab="No SVA",
ylim = c(-20,20), xlim=c(-20,20))
abline(0,1)
plot(rank(eb1$p[,3])[neurIndex], rank(eb0$p[,3])[neurIndex])
abline(0,1)
sort(rank(eb1$p[,3])[neurIndex])
pdf("suppFigure_ranks_neuro.pdf")
plot(rank(eb1$p[,3]), rank(eb0$p[,3]),pch=21,bg="grey",
xlab="Post-SVA Rank", ylab="Pre-SVA Rank",cex=0.7,
cex.axis=1.5,cex.lab=1.5)
points(rank(eb1$p[,3])[neurIndex], cex=2,
rank(eb0$p[,3])[neurIndex],pch=21,bg="red")
abline(0,1,col="blue",lty=2)
dev.off()
pdf("suppFigure_ranks_mesen.pdf")
plot(rank(eb1$p[,2]), rank(eb0$p[,2]),pch=21,bg="grey",
xlab="Post-SVA Rank", ylab="Pre-SVA Rank",cex=0.7)
points(rank(eb1$p[,2])[mesenIndex], cex=2,
rank(eb0$p[,2])[mesenIndex],pch=21,bg="red")
abline(0,1,col="blue",lty=2)
#############
# check pax6
pIndex = grep("PAX6", map$GeneSymbol)
eb0$p[pIndex,2:3]
eb1$p[pIndex,2:3]
pdf("treat_fig/treat_es_noSVA_pax6.pdf")
expBySample(p[pIndex,], pd$SampleID, geneName = "",cex=1,
conditions = pd$Treatment,legend = FALSE, cex.axis=1.3, cex.lab=1.5)
dev.off()
pdf("treat_fig/treat_es_SVA_pax6.pdf")
expBySample(cleanp[pIndex,], pd$SampleID, geneName = "",cex=1,
conditions = pd$Treatment,legend = FALSE, cex.axis=1.3, cex.lab=1.5)
dev.off()
# plot p-value
pdf("treat_fig/overall_pval_new.pdf")
plot(-log10(eb1$p[,2]), -log10(eb0$p[,2]),
xlab="p-value (SVA)",ylab="p-value (No SVA)",
col = alpha("black",0.33),
cex.axis=1.5)
points(-log10(eb1$p[,2])[oIndex], -log10(eb0$p[,2])[oIndex],
col = "red",pch=19,cex=1.4)
points(-log10(eb1$p[,2])[pIndex], -log10(eb0$p[,2])[pIndex],
col = "blue",pch=19,cex=1.4)
abline(0,1,col="blue")
dev.off()
# plot sigma
pdf("treat_fig/overall_sigma.pdf")
plot(fit1$sigma, fit0$sigma,
xlab="sigma (SVA)",ylab="sigma (No SVA)",
col = alpha("black",0.33),
cex.axis=1.5)
abline(0,1,col="blue")
points(fit1$sigma[oIndex], fit0$sigma[oIndex],
col = "red",pch=19,cex=1.4)
points(fit1$sigma[pIndex], fit0$sigma[pIndex],
col = "blue",pch=19,cex=1.4)
dev.off()
# plot betas
pdf("treat_fig/overall_betas.pdf")
plot(fit1$coef[,2], fit0$coef[,2],
xlab="beta (SVA)",ylab="beta (No SVA)",
col = alpha("black",0.33),
cex.axis=1.5)
abline(0,1,col="blue")
points(fit1$coef[oIndex,2], fit0$coef[oIndex,2],
col = "red",pch=19,cex=1.4)
points(fit1$coef[pIndex,2], fit0$coef[pIndex,2],
col = "blue",pch=19,cex=1.4)
dev.off()
##############################
# gstt1
try(system("mkdir sex_gstt1"))
gIndex = grep("GSTT1", map$GeneSymbol)[2]
ylim = c(6,12.5)
pal = c("black","black")
# take the best hit
pdf("sex_gstt1/gstt1_es_noSVA_fig2d.pdf")
expBySample(p[gIndex,], pd$SampleID, geneName = "",cex=1,ylim=ylim,
conditions = NULL, pal=pal,legend = FALSE, cex.axis=1.3, cex.lab=1.5)
dev.off()
pdf("sex_gstt1/gstt1_es_wrongSVA_fig2e.pdf")
expBySample(cleanp[gIndex,], pd$SampleID, geneName = "",cex=1,
conditions = NULL, pal=pal,legend = FALSE, cex.axis=1.3, cex.lab=1.5,
ylim = ylim)
dev.off()
# determine CNV by group
gstt1exp = tapply(p[gIndex,], pd$SampleID, median)
plot(gstt1exp)
gstt1pred = cut(gstt1exp, c(5,7.5, 10.5,12), labels = F)-1
names(gstt1pred) = names(gstt1exp)
pd$gstt1 = rep(NA)
sIndexes = splitit(pd$SampleID)
for(i in seq(along=sIndexes)) {
ind=sIndexes[[i]]
pd$gstt1[ind] = rep(gstt1pred[names(sIndexes)[i]])
}
mod3 = model.matrix(~gstt1+Treatment, data =pd)
svaobj3 = sva(p,mod3, n.sv=26) # default
cleanp3 = cleaningP(p,mod3,svaobj3)
pdf("sex_gstt1/gstt1_es_rightSVA_fig2f.pdf")
expBySample(cleanp3[gIndex,], pd$SampleID, geneName = "",cex=1,
conditions = NULL, pal=pal,legend = FALSE, cex.axis=1.3, cex.lab=1.5,
ylim = ylim)
dev.off()
#########
# sex
# find sex difference w/o SVA
fit0a = lmFit(p, model.matrix(~pd$Sex + pd$Treatment))
eb0a = ebayes(fit0a)
sIndex = grep("RPS4Y1", map$GeneSymbol)
# take the best hit
pdf("sex_gstt1/sex_es_noSVA_fig2a.pdf")
expBySample(p[sIndex,], pd$SampleID, geneName = map$GeneSymbol[sIndex],
conditions = pd$Sex, legend = FALSE, cex.axis=1.3, cex.lab=1.5,cex=1,
ylim = c(4,17), pal="Set1")
dev.off()
pdf("sex_gstt1/sex_es_wrongSVA_fig2b.pdf")
expBySample(cleanp[sIndex,], pd$SampleID, geneName = map$GeneSymbol[sIndex],
conditions = pd$Sex, legend = FALSE, cex.axis=1.3, cex.lab=1.5,
ylim = c(4,17), cex=1, pal="Set1")
legend("top", levels(factor(pd$Sex)), col = 1:2, pch = 15,
pt.cex=3, nc = 2, cex =1.75,bty="n")
dev.off()
#### adjust for sex in SVA
mod2 = model.matrix(~Sex+Treatment, data =pd)
svaobj2 = sva(p,mod2, n.sv=26)
cleanp2 = cleaningP(p,mod2,svaobj2)
pdf("sex_gstt1/sex_es_rightSVA_fig2c.pdf")
expBySample(cleanp2[sIndex,], pd$SampleID, geneName = map$GeneSymbol[sIndex],
conditions = pd$Sex, legend = FALSE, cex.axis=1.3, cex.lab=1.5,
ylim = c(4,17), cex=1, pal="Set1")
dev.off()