-
Notifications
You must be signed in to change notification settings - Fork 4
/
Copy pathhelperfunctions.R
323 lines (299 loc) · 11.4 KB
/
helperfunctions.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
272
273
274
275
276
277
278
279
280
281
282
283
284
285
286
287
288
289
290
291
292
293
294
295
296
297
298
299
300
301
302
303
304
305
306
307
308
309
310
311
312
313
314
315
316
317
318
319
320
321
322
323
select.matched <- function(x, y) {
x <- unlist(strsplit(x, split=","))
x <- x[x %in% y]
x <- paste(x, collapse=",")
return(x)
}
v.select.matched <- function(vec1, y) {
sapply(vec1, select.matched, y)
}
#! returns a logical vector, i th element TRUE if one of the strings in vec1[i] is matched by one of the strings in vec2[i]
matchedin.elementwise <- function(vec1, vec2) { # vec1 and vec2 are vectors of comma-separated strings, of same length
N <- length(vec1)
matched <- logical(N)
for(i in 1:N) {
v1 <- unlist(strsplit(vec1[i], split=","))
v2 <- unlist(strsplit(vec2[i], split=","))
matched[i] <- any(v1 %in% v2)
}
return(matched)
}
matchedin <- function(str1, str2) { # returns TRUE if any of comma-separated strings in str1 (possibly a vector of comma-separated strings) are matched by comma-separated strings in str2
v2 <- unlist(strsplit(str2, split=","))
match.element <- function(str1) { # str1 is not a vector
v1 <- unlist(strsplit(str1, split=","))
v1.matched <- paste(unique(v1[v1 %in% v2]), collapse=",")
return(v1.matched)
}
sapply(X=str1, FUN=match.element, simplify=TRUE, USE.NAMES=FALSE)
}
#matchedin(trans.genome.wide.scoresinfo[gene_symbol=="CTLA4", nearby_genes], t1d.genes$nearest)
minus.log10.phnorm.extreme <- function(z, upper=TRUE) {
z.hi <- function(z) {
## https://www.johndcook.com/blog/norm-dist-bounds/
c.upper <- 8 / pi
c.lower <- 4
z.hi <- z[z >= 8.2]
x.upper <- 2 * sqrt(2 / pi) / (z.hi + sqrt(z.hi^2 + c.upper))
x.lower <- 2 * sqrt(2 / pi) / (z.hi + sqrt(z.hi^2 + c.lower))
ln_p.upper = -0.5 * z.hi^2 + log(x.upper)
ln_p.lower = -0.5 * z.hi^2 + log(x.upper)
log10p <- 0.5 * (ln_p.upper + ln_p.lower) / log(10)
return(log10p)
}
z.dt <- data.table(z=abs(z), log10p=NA)
## phnorm gives distribution function for half-normal distribution
z.dt[z < 8.2, log10p := log10(1 - phnorm(z))]
z.dt[z >= 8.2, log10p := z.hi(z)]
return(z.dt[, -log10p])
}
p.threshold <- function(p, fdr=0.01) {
pvalues <- data.table(p)
pvalues[, index := .I]
setorder(pvalues, p)
pvalues[, i := .I]
pvalues[, threshold := fdr * i / .N]
pvalues[, is.signif := p < threshold]
# if no pvalues are below threshold, this returns -Inf
p.max <- suppressWarnings(signif(pvalues[is.signif==TRUE, max(p)], digits=1))
if(is.infinite(p.max)) {
p.max <- min(pvalues$threshold) # defaults to bonferroni threshold
}
return(p.max)
}
pnorm.extreme <- function(z, upper=TRUE) {
## https://www.johndcook.com/blog/norm-dist-bounds/
if(upper) { # upper bound on log10 tail prob
c <- 8 / pi
} else { # lower bound
c = 4
}
x <- 2 * sqrt(2 / pi) / (z + sqrt(z^2 + c))
ln_p = -0.5 * z^2 + log(x)
log10p <- ln_p / log(10)
exponent <- floor(log10p)
coeff <- 10^(log10p - exponent)
string <- paste0(round(coeff), "E", exponent)
return(string)
}
#' reformat for LaTeX p-values that are in scientific notation
format.scinot.pvalue <- function(x, nexp=1) {
x <- toupper(x)
x.split <- as.numeric(unlist(strsplit(as.character(x), "E"))) # split x.split at E
x.split <- signif(as.numeric(x.split, 1))
x.split <- t(matrix(x.split, nrow=2))
roundedto10 <- x.split[, 1]==10
x.split[, 1][roundedto10] <- 1
x.split[, 2][roundedto10] <- x.split[, 2][roundedto10] - 1
p.latex <- sprintf("\\ensuremath{%.*f \\times 10^{%0*d}}", 0, x.split[, 1], nexp, x.split[, 2])
return(p.latex)
}
#' generate formatted p-values from z value
#' global variables: sigfig, neglogp.threshold.scinot, neglogp.threshold
format.z.aspvalue <- function(z) {
if(!exists("sigfig")) {
sigfig <- 1
} else {
if(sigfig==0) {
sigfig <- 1
}
}
## neglogp.threshold.scinot is threshold for using scientific notation
if(!exists("neglogp.threshold.scinot")) {
neglogp.threshold.scinot <- 3
} else {
if(neglogp.threshold.scinot==0) {
neglogp.threshold.scinot <- 3
}
}
p <- signif(2 * pnorm(-abs(z)), sigfig)
p.char <- toupper(as.character(p))
## pnorm.extreme returns a character string of form "NE-NNN"
p.char[!is.na(p.char) & p.char=="0"] <- pnorm.extreme(z[!is.na(p.char) & p.char=="0"]) # where R outputs 0
sci.revert <- grepl("E", p.char) & p > 10^-neglogp.threshold.scinot
p.char[sci.revert] <- format(p[sci.revert], scientific=FALSE)
if(exists("neglogp.threshold")) {
if(neglogp.threshold > 0) { # thresholding of p values
p.char[p < 10^-neglogp.threshold] <- paste0("<",
format(10^-neglogp.threshold,
scientific=FALSE))
}
} else {
}
p.char[grep("E", p.char)] <- format.scinot.pvalue(p.char[grep("E", p.char)])
return(p.char)
}
format.pvalue <- function(z, pvalue=NULL) {
format.z.aspvalue(z)
}
## format a vector of pvalues in LaTeX and return a vector of mode character
pvalue.latex <- function(x, n=1, nexp=1, s.threshold=s.threshold, p.threshold=p.threshold) {
## this function has to be able to handle x whether numeric or character
pvalue <- sapply(x, function(z) { # sapply returns a vector applying FUN to each element of x
if (is.na(z) | is.nan(z)) {
return(NA)
} else if(as.numeric(z) >= 10^s.threshold) {
## return character string to one sig fig, not in scientific notation
return(as.character(signif(as.numeric(z), 1)))
} else {
if(is.numeric(z)) {
## rounds to 1 sig fig and convert to character string
z <- sprintf("%.*E", 0, signif(z, n)) # default is 1 sig fig
} else {
z <- toupper(z)
}
z <- as.numeric(unlist(strsplit(as.character(z), "E"))) # split z at E
sprintf("\\ensuremath{%.*f\\times 10^{%0*d}}", 0, z[1], nexp, z[2])
}
}
)
pvalue <- as.character(pvalue)
if(threshold==-4) {
pvalue[grep("\\\\times", pvalue)] <- "<0.0001" # fix for thresholding at 0.0001
}
return(pvalue)
}
dist2order = function(corr, method, ...) {
d_corr = as.dist(1 - corr)
s = seriation::seriate(d_corr, method = method, ...)
i = seriation::get_order(s)
return(i)
}
diversity <- function(x, q) {
p <- x / sum(x)
if(q != 1) {
diversity <- (sum(p^q))^(1 / (1 - q))
} else {
diversity <- 2^sum(-p * log2(p))
}
return(diversity)
}
minus.log10.phnorm.extreme <- function(z, upper=TRUE) {
z.hi <- function(z) {
## https://www.johndcook.com/blog/norm-dist-bounds/
c.upper <- 8 / pi
c.lower <- 4
z.hi <- z[z >= 8.2]
x.upper <- 2 * sqrt(2 / pi) / (z.hi + sqrt(z.hi^2 + c.upper))
x.lower <- 2 * sqrt(2 / pi) / (z.hi + sqrt(z.hi^2 + c.lower))
ln_p.upper = -0.5 * z.hi^2 + log(x.upper)
ln_p.lower = -0.5 * z.hi^2 + log(x.upper)
log10p <- 0.5 * (ln_p.upper + ln_p.lower) / log(10)
return(log10p)
}
log10p <- numeric(length(z))
z <- abs(z)
log10p[z < 8.2] <- log10(1 - phnorm(z[z < 8.2]))
log10p[z >=8.2] <- z.hi(z)
return(-log10p)
}
eqtls.gene <- function(gsymbol, qtl.info) {
eqtls <- qtl.info[gene_symbol==gsymbol,
.(qtl_type, CHR=chrom_min, startpos,
endpos, variance, near.t1dgenes, nearby_genes)]
#eqtls[, CHR := factor(CHR, levels=c(1:22, "X", "Y"))]
setkey(eqtls, CHR)
setkey(chr.lengths, CHR)
eqtls <- chr.lengths[eqtls]
eqtls[, xstart := 0.5 * (startpos + endpos) + cumbp]
eqtls[, xend := eqtls[qtl_type=="cis", xstart]]
return(eqtls)
}
chr.lengths <- fread(text="CHR,length
1, 248956422
2, 242193529
3,198295559
4,190214555
5,181538259
6,170805979
7,159345973
8,145138636
9,138394717
10,133797422
11,135086622
12,133275309
13,114364328
14,107043718
15,101991189
16,90338345
17,83257441
18,80373285
19,58617616
20,64444167
21,46709983
22,50818468
X,156040895
Y,57227415")
chr.lengths[, CHR := factor(CHR, levels=c(1:22, "X", "Y"))]
chr.lengths[, cumbp := c(0, cumsum(as.numeric(length)[-.N]))]
chr.lengths[, midpoint.chr := 0.5 * length]
manhattan <- function(results, point.size=1) {
N <- nrow(results)
## thin data points
keep.rows <- c(results[, sample(.N, ceiling(0.01 * .N))],
results[minuslog10p > 1, sample(.N, ceiling(0.1 * .N))],
results[minuslog10p > 1.5, which=TRUE])
keep.rows <- sort(unique(keep.rows))
results.pruned <- results[keep.rows]
chrom.labels <- as.character(1:22)
chrom.labels[c(19, 21)] <- " "
chrom.midpoints <- position.absolute(chr.lengths$CHR, chr.lengths$midpoint.chr)[1:22]
chrom.breaks <- chr.lengths[1:22, cumbp]
chrom.ends <- position.absolute(chr.lengths$CHR, chr.lengths$length)
p.manhattan <- ggplot(results.pruned,
aes(x=x, y=minuslog10p, colour=CHR)) +
geom_point(size=point.size) +
scale_x_continuous(expand=c(0, 0),
breaks=chrom.midpoints, labels=chrom.labels,
minor_breaks = chrom.breaks) +
theme(panel.grid.major.x = element_blank()) +
scale_y_continuous(expand=expansion(add = c(0, 1)),
breaks=c(0, 5, 10, 15, 20)) +
labs(y = expression(paste(-log[10], " ", italic(p))),
x = "Chromosome") +
theme(legend.position="none") +
theme(axis.title.x = element_text(size = 14),
axis.text.x = element_text(size = 12),
axis.title.y = element_text(size = 14),
axis.text.y = element_text(size = 12)) +
scale_color_manual(values=rep(c("#AAAAAA","#666666"), length(breaks.chr)))
return(p.manhattan)
}
#! convert chr bp to absolute position x, using hg38 to set length of chr
position.absolute <- function(CHR, BP, padding=0) {
positions <- data.table(CHR, BP)
positions[, rownum := .I]
setkey(positions, CHR)
setkey(chr.lengths, CHR)
positions <- chr.lengths[, .(CHR, cumbp)][positions]
positions[, x := BP + cumbp + padding]
setorder(positions, rownum)
return(positions[, x])
}
## lift positions from hg19 (GRCh37) to hg38
liftover.positions <- function(CHR, BP, data.dir) {
positions37.bed <- file.path(data.dir, "positions37.bed")
positions38.bed <- file.path(data.dir, "positions38.bed")
chain.file <- file.path(data.dir, "hg19ToHg38.over.chain")
unmapped <- file.path(data.dir, "unmapped")
positions37 <- data.table(CHR, BP=as.integer(BP))
positions37[, rowname := sprintf("row%09d", .I)]
positions37[, CHR := paste0("chr", CHR)]
write.table(positions37[, .(chrom=CHR,
chromStart=BP-1,
chromEnd=BP,
name=rowname)],
row.names=FALSE, col.names=FALSE, quote=FALSE, sep="\t",
file=positions37.bed)
cmd <- sprintf("liftOver %s %s %s %s",
positions37.bed, chain.file, positions38.bed, unmapped)
system(cmd)
positions38 <- fread(positions38.bed)
positions38 <- positions38[, c(4, 2)]
colnames(positions38) <- c("rowname", "pos.hg38")
setkey(positions38, rowname)
setkey(positions37, rowname)
positions <- positions38[positions37]
setorder(positions, rowname)
return(positions[, pos.hg38])
}