forked from douglasgscofield/whyp
-
Notifications
You must be signed in to change notification settings - Fork 0
/
whyp.R
309 lines (282 loc) · 11.5 KB
/
whyp.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
# whyp.R
#
# Calculate parentage using mixed-assay likelihood (Smouse et al. 2012 Journal
# of Heredity).
#
# Copyright (c) 2012 Douglas G. Scofield, Umeå Plant Science Centre, Umeå, Sweden
#
#
# These statistical tools were developed in collaboration with Peter Smouse
# (Rutgers University) and Victoria Sork (UCLA) and were funded by U.S. National
# Science Foundation awards NSF-DEB-0514956 and NSF-DEB-0516529.
#
# Use as you see fit. No warranty regarding this code is implied nor should be
# assumed. Send bug reports etc. to one of the above email addresses.
#
.whypVersion = "0.01"
#
# CHANGELOG
#
# 0.01: First prerelease
z = try(source("readGenalex.R"), silent=TRUE)
if (class(z) == "try-error") {
z = try(source("whyp_functions/readGenalex.R"), silent=TRUE)
if (class(z) == "try-error") {
stop("readGenalex.R must be available, download it at https://github.com/douglasgscofield/popgen")
}
}
source("whyp_functions/whyp_read_data.R")
source("whyp_functions/whyp_alleles.R")
source("whyp_functions/whyp_likelihood.R")
source("whyp_functions/whyp_extract_likelihood.R")
source("whyp_functions/whyp_backward_probability.R")
source("whyp_functions/whyp_check.R")
source("whyp_functions/whyp_utility.R")
# Configuration variables
.max.matches.to.report <- 10
# missing allele indications in raw datafiles
missing.allele <- c("0", "", " ")
genalex.missing.allele <- "0"
# missing allele indication in indexed genotype matrices
missing.allele.index <- -1
# treat single missing allele as homozygote
missing.single.allele.avail.methods <- c("as.missing","as.homozygote","as.is")
#missing.single.allele.method <- "as.missing"
missing.single.allele.method <- match.arg("as.missing",
missing.single.allele.avail.methods)
# any loci to drop from the entire dataset?
.drop.loci <- NULL
# probability of a null genotyping error
.Pr.null <- 0.02
# use the seedling mismatch heuristic
.use.heur.mismatch <- TRUE
.heur.tissue.methods <- c("seedling","pericarp")
.report.heur.mismatch <- FALSE
# probability assigned to a heuristic mismatch
.Pr.heur.mismatch <- 0.01
.heur.methods <- c("if.no.match","any")
# data sources we can use to make a parental assignment
.data.sources <- c("pericarp.priority", "use.all", "pericarp", "seedling")
.poke.data.sources <- c("pericarp.priority", "pericarp", "seedling")
.poke.holes.assay.running <- FALSE
.joint.method <- "pericarp.priority"
# neg-log-likelihood window within the min value for which we report candidates
.L.window <- Inf
# window to use for diagnostics produced by extract.via.L
.included.candidate.window <- c( LOGFUNC(10^2) )
#.included.candidate.window <- c(LOGFUNC(exp(1)^3),
# LOGFUNC(10^1),
# LOGFUNC(10^2),
# LOGFUNC(10^3))
whyp <- function(dat,
recruits=dat$recruit.names,
heur.method="if.no.match",
heur.tissue="seedling")
{
df <- data.frame()
i <- 0
for (recruit in recruits) {
LOD.recruit <- LOD.Xkl(Xkl=Xkl.recruit.single(recruit, dat),
dat=dat,
heur.method=heur.method,
heur.tissue=heur.tissue)
df <- rbind(df, LOD.package(LOD.scores=LOD.recruit))
i <- i + 1
if (i %% 20 == 0)
cat("done with recruit", i, recruit, "...\n")
}
cat("completed whyp analysis for", i, "recruits\n")
####
df
}
extract.samples.with.n.missing.loci <- function(dat, n.missing.loci=NULL)
{
# dat needs to have name-valued rows. shouldn't be an issue
if (! is.null(n.missing.loci)) {
ml <- attr(dat,"n.missing.loci")
ml.indexes <- which(ml == n.missing.loci)
ans <- dat[ml.indexes,]
ml <- ml[ml.indexes]
for (att in names(attributes(dat)))
if (! att %in% c("dim","dimnames","names")) # copy all but these
attr(ans,att) <- attr(dat,att)
attr(ans,"n.missing.loci") <- ml # update
dat <- ans
}
dat
}
assemble.whyp.data <- function(mother=NULL,
pericarp=NULL,
seedling=NULL,
file.mother=NULL,
file.pericarp=NULL,
file.seedling=NULL,
report.foreign.recruit=NULL,
report.mismatch.recruit=NULL,
report.null.recruit=NULL,
file.allele.freqs=NULL,
drop.loci=.drop.loci,
write.reports=TRUE)
{
dat <- list() # add it all to a big list
is.mother.info = !is.null(file.mother) || !is.null(mother)
is.seedling.info = !is.null(file.seedling) || !is.null(seedling)
is.pericarp.info = !is.null(file.pericarp) || !is.null(pericarp)
if (!is.mother.info) stop("must provide mother genotypes")
if (is.pericarp.info && !is.seedling.info) {
# for now, work around the inconvenience of having to provide both pericarp
# and seedling genotypes... keep track in the list of which were actuall
# provided
file.seedling = file.pericarp
seedling = pericarp
attr(dat, "data.sources.given") <- c("pericarp")
} else if (is.seeding.info && !is.pericarp.info) {
file.pericarp = file.seedling
pericarp = seedling
attr(dat, "data.sources.given") <- c("seedling")
} else if (!is.seeding.info && !is.pericarp.info) {
stop("must provide pericarp or seedling genotypes, or both")
} else {
attr(dat, "data.sources.given") <- c("pericarp", "seedling")
}
mother <- if (is.null(mother)) read.mother(file.mother) else mother
pericarp <- if (is.null(pericarp)) read.pericarp(file.pericarp) else pericarp
seedling <- if (is.null(seedling)) read.seedling(file.seedling) else seedling
if (! is.null(drop.loci)) {
cat("dropping loci",drop.loci," from mother, pericarp and seedling...\n")
mother <- dropGenalexLoci(mother, drop.loci, quiet=TRUE)
pericarp <- dropGenalexLoci(pericarp, drop.loci, quiet=TRUE)
seedling <- dropGenalexLoci(seedling, drop.loci, quiet=TRUE)
}
if (TRUE) {
# check consistency of loci used
ml <- attr(mother, "locus.names"); nml <- length(ml)
pl <- attr(pericarp, "locus.names"); npl <- length(pl)
sl <- attr(seedling, "locus.names"); nsl <- length(sl)
loci <- auto.drop.loci <- c()
cat("# mother loci =", nml, " # pericarp loci =", npl,
" # seedling loci =", nsl, "\n")
if (nml == npl && all(ml == pl) && nml == nsl && all(ml == sl)) {
cat("all mother, seedling, pericarp loci are identical and in same order\n")
loci <- ml
loci.identical <- TRUE
} else {
loci.identical <- FALSE
if (all(pl %in% ml) && all(ml %in% pl)) {
cat("mother and pericarp locus sets are identical, but not in same order\n")
loci <- sort(ml)
} else {
cat("mother, pericarp, seedling locus sets are not identical\n")
loci <- sort(intersect(ml, intersect(pl, sl)))
auto.drop.loci <- setdiff(union(ml, union(pl, sl)), loci)
}
}
if (length(auto.drop.loci) > 0) {
cat("auto-dropping loci ", auto.drop.loci, " from all datasets\n")
mother <- dropGenalexLoci(mother, auto.drop.loci, quiet=TRUE)
pericarp <- dropGenalexLoci(pericarp, auto.drop.loci, quiet=TRUE)
seedling <- dropGenalexLoci(seedling, auto.drop.loci, quiet=TRUE)
}
mother <- reorderGenalexLoci(mother, loci)
pericarp <- reorderGenalexLoci(pericarp, loci)
seedling <- reorderGenalexLoci(seedling, loci)
}
dat$mother <- mother
cat("*** note: merge.recruit skipped to work around locus column reorder bug\n")
# dat$recruit <- merge.recruit(pericarp=pericarp, seedling=seedling)
dat$recruit <- list(pericarp=pericarp, seedling=seedling)
if ("pericarp" %in% attr(dat, "data.sources.given")) {
dat$recruit.names <- pericarp[,1]
} else if ("seedling" %in% attr(dat, "data.sources.given")) {
dat$recruit.names <- seedling[,1]
} else stop("can't find recruit names")
# do we read allele frequencies from a file, or calculate them from input data?
if (! is.null(file.allele.freqs)) {
dat$master.alist <- file.allele.freqs
} else {
# create allele frequencies from scratch
cat("calculating allele frequencies from scratch...\n")
mother.alist <- create.allele.list(dat$mother)
pericarp.alist <- create.allele.list(dat$recruit$pericarp)
seedling.alist <- create.allele.list(dat$recruit$seedling)
recruit.alist <- merge.allele.list(pericarp.alist, seedling.alist)
dat$foreign.alist <- find.foreign.alleles(mother.alist,
foreign=recruit.alist,
method="actual.counts")
dat$master.alist <- merge.allele.list(mother.alist, dat$foreign.alist)
}
# create allele table and index genotypes using it
cat("creating allele table and using it to index genotypes...\n")
dat$allele.table <- create.allele.table(dat$master.alist)
dat$imother <- index.genotypes(dat$mother, dat$allele.table)
dat$ipericarp <- index.genotypes(dat$recruit$pericarp, dat$allele.table)
dat$iseedling <- index.genotypes(dat$recruit$seedling, dat$allele.table)
# accuracies of these depend on original and indexed versions sharing row order
attr(dat$mother,"n.missing.loci") <- attr(dat$imother,"n.missing.loci")
for (a in c("n.missing.loci","name.missing.loci")) {
attr(dat$recruit$pericarp,a) <- attr(dat$ipericarp,a)
attr(dat$recruit$seedling,a) <- attr(dat$iseedling,a)
}
if (! is.null(report.foreign.recruit) && ! is.null(dat$foreign.alist)
&& write.reports) {
cat("writing report to file", report.foreign.recruit, "\n")
sink(file=report.foreign.recruit)
report.foreign.alleles(dat)
sink()
}
# add names for check.recruit.mismatch()
cat("checking for mismatches between recruit pericarp and seedling...\n")
if (! is.null(report.mismatch.recruit)) {
if (write.reports) {
cat("writing report to file", report.mismatch.recruit, "\n")
sink(file=report.mismatch.recruit)
dat <- check.recruit.mismatch(dat, report=TRUE)
sink()
}
} else {
dat <- check.recruit.mismatch(dat, report=FALSE)
}
cat("editing mismatched genotypes...\n")
dat <- edit.recruit.mismatch(dat, method="set.all.missing", update.attributes=TRUE)
cat("checking for potential null genotyping errors between recruit pericarps and all mothers...\n")
if (! is.null(report.null.recruit)) {
if (write.reports) {
cat("writing report to file", report.null.recruit, "\n")
sink(file=report.null.recruit)
dat <- check.recruit.null(dat, report=TRUE)
sink()
}
} else {
dat <- check.recruit.null(dat, report=FALSE)
}
#
#
dat
}
# old wrapper functions
full.mixed.assay <- function(read.freqs=FALSE)
{
dat <- assemble.whyp.data(read.freqs=read.freqs)
report.foreign.alleles(dat)
Ljk.r <- mixed.assay(dat)
print.mixed.assay(Ljk.r)
invisible(Ljk.r)
}
mixed.assay <- function(dat, Pr.null=.Pr.null)
{
# calculate recruit X mother X locus likelihoods
cat("mixed.assay Pr.null =",Pr.null,"\n")
Xjkl.p <- Xjkl.pericarp(dat$ipericarp, dat$imother,
dat$allele.table, Pr.null=Pr.null)
Xjkl.s <- Xjkl.seedling(dat$iseedling, dat$imother,
dat$allele.table)
# assemble recruit X mother likelihoods
Ljk.r <- Ljk.recruit(Xjkl.pericarp=Xjkl.p, Xjkl.seedling=Xjkl.s)
attr(Ljk.r, "data.sources.given") <- attr(dat, "data.sources.given")
attr(Ljk.r, "data") <- dat
####
####
Ljk.r
}