root / src / R / nucleominer.R @ 1d833b97
Historique | Voir | Annoter | Télécharger (85,06 ko)
| 1 |
# get_content = function(# Get content from cached |
|---|---|
| 2 |
# ### Acces to the cached content of a file via the global variable NM_CACHE. Load content in NM_CACHE if needed. |
| 3 |
# obj, ##<< The object that we want its content. |
| 4 |
# ... ##<< Parameters that will be passed to my_read |
| 5 |
# ) {
|
| 6 |
# UseMethod("get_content", obj)
|
| 7 |
# ### Returns the cached content of the current object. |
| 8 |
# } |
| 9 |
# |
| 10 |
# get_content.default = structure( function(# Get content from cached |
| 11 |
# ### Acces to the cached content of a file via the global variable NM_CACHE. Load content in NM_CACHE if needed. |
| 12 |
# obj, ##<< The object that we want its content. |
| 13 |
# ... ##<< Parameters that will be passed to my_read |
| 14 |
# ) {
|
| 15 |
# if (inherits(try(NM_CACHE,TRUE), "try-error") || is.null(NM_CACHE)) {
|
| 16 |
# NM_CACHE <<- list() |
| 17 |
# } |
| 18 |
# if (is.null(NM_CACHE[[obj$filename]])) {
|
| 19 |
# print(paste("Loading file ",obj$filename, sep=""))
|
| 20 |
# tmp_content = my_read(obj, ...) |
| 21 |
# print("affect it...")
|
| 22 |
# NM_CACHE[[obj$filename]] <<- tmp_content |
| 23 |
# print("done.")
|
| 24 |
# } |
| 25 |
# return(NM_CACHE[[obj$filename]]) |
| 26 |
# ### Returns the cached content of the current object. |
| 27 |
# }, ex=function(){
|
| 28 |
# # Create a dataframe |
| 29 |
# df = NULL |
| 30 |
# df = dfadd(df, list(key1 = "value1", key2 = "value2")) |
| 31 |
# df = dfadd(df, list(key1 = "value1'", key2 = "value3'")) |
| 32 |
# # Dump it into tmp file |
| 33 |
# write.table(df, file="/tmp/tmp.dump.table.tmp") |
| 34 |
# # Load it into cache using feature. |
| 35 |
# cached_content_object = list(filename="/tmp/tmp.dump.table.tmp") |
| 36 |
# class(cached_content_object) = "table" |
| 37 |
# # First time it will be load into cache |
| 38 |
# print(get_content(cached_content_object)) |
| 39 |
# # Second time not |
| 40 |
# print(get_content(cached_content_object)) |
| 41 |
# |
| 42 |
# }) |
| 43 |
# |
| 44 |
# my_read = function( |
| 45 |
# ### Abstract my_read function. |
| 46 |
# obj, ...) {
|
| 47 |
# UseMethod("my_read", obj)
|
| 48 |
# } |
| 49 |
# |
| 50 |
# my_read.default = function( |
| 51 |
# ### Default my_read function. |
| 52 |
# obj, ...){
|
| 53 |
# stop(paste("ERROR, my_read is not defined for any Objects (file ", obj$filename," )", sep=""))
|
| 54 |
# } |
| 55 |
# |
| 56 |
# my_read.fasta = function( |
| 57 |
# ### my_read function for fasta files. |
| 58 |
# obj, ...){
|
| 59 |
# # require(seqinr) |
| 60 |
# return(read.fasta(obj$filename, ...)) |
| 61 |
# } |
| 62 |
# |
| 63 |
# my_read.table = function( |
| 64 |
# ### my_read function for table files. |
| 65 |
# obj, ...){
|
| 66 |
# if (rev(unlist(strsplit(obj$filename, ".", fixed=TRUE)))[1] == "gz") {
|
| 67 |
# return(read.table(file=gzfile(obj$filename), ...)) |
| 68 |
# } else {
|
| 69 |
# return(read.table(file=obj$filename, ...)) |
| 70 |
# } |
| 71 |
# } |
| 72 |
# |
| 73 |
# my_read.cvs = function( |
| 74 |
# ### my_read function for cvs files. |
| 75 |
# obj, ...){
|
| 76 |
# return(read.csv(file=obj$filename, ...)) |
| 77 |
# } |
| 78 |
|
| 79 |
|
| 80 |
|
| 81 |
|
| 82 |
|
| 83 |
|
| 84 |
|
| 85 |
|
| 86 |
|
| 87 |
FDR = structure(function# False Discovery Rate |
| 88 |
### From a vector x of independent p-values, extract the cutoff corresponding to the specified FDR. See Benjamini & Hochberg 1995 paper |
| 89 |
##author<< Gael Yvert, |
| 90 |
( |
| 91 |
x, ##<< A vector x of independent p-values. |
| 92 |
FDR ##<< The specified FDR. |
| 93 |
) {
|
| 94 |
x <- sort(na.omit(x)) |
| 95 |
N = length(x) |
| 96 |
i = 1; |
| 97 |
while(N*x[i]/i < FDR & i <= N) i = i + 1; # we search for the highest i where Nrandom / Nobserved < FDR |
| 98 |
if (i == 1) |
| 99 |
return (NA) |
| 100 |
else |
| 101 |
return( x[i-1] ) |
| 102 |
### Return the the corresponding cutoff. |
| 103 |
}, ex=function(){
|
| 104 |
print("example")
|
| 105 |
}) |
| 106 |
|
| 107 |
|
| 108 |
|
| 109 |
lod_score_vecs = structure(function # Likelihood ratio |
| 110 |
### Compute the likelihood log of two set of value from two models Vs. a unique model. |
| 111 |
( |
| 112 |
x ,##<< First vector. |
| 113 |
y ##<< Second vector. |
| 114 |
) {
|
| 115 |
if (length(x) <=1 | length(y) <= 1) {
|
| 116 |
return(NA) |
| 117 |
} |
| 118 |
meanX = mean(x) |
| 119 |
sdX = sd(x) |
| 120 |
meanY = mean(y) |
| 121 |
sdY = sd(y) |
| 122 |
meanXY = mean(c(x,y)) |
| 123 |
sdXY = sd(c(x,y)) |
| 124 |
llX = sum(log(dnorm(x,mean=meanX,sd=sdX))) |
| 125 |
llY = sum(log(dnorm(y,mean=meanY,sd=sdY))) |
| 126 |
llXY = sum(log(dnorm(c(x,y),mean=meanXY,sd=sdXY))) |
| 127 |
ratio = -(llX + llY - llXY) |
| 128 |
return(ratio) |
| 129 |
### Returns the likelihood ratio. |
| 130 |
}, ex=function(){
|
| 131 |
# LOD score for 2 set of values |
| 132 |
mean1=5; sd1=2; card2 = 250 |
| 133 |
mean2=6; sd2=3; card1 = 200 |
| 134 |
x1 = rnorm(card1, mean1, sd1) |
| 135 |
x2 = rnorm(card2, mean2, sd2) |
| 136 |
min = floor(min(c(x1,x2))) |
| 137 |
max = ceiling(max(c(x1,x2))) |
| 138 |
hist(c(x1,x2), xlim=c(min, max), breaks=min:max) |
| 139 |
lines(min:max,dnorm(min:max,mean1,sd1)*card1,col=2) |
| 140 |
lines(min:max,dnorm(min:max,mean2,sd2)*card2,col=3) |
| 141 |
lines(min:max,dnorm(min:max,mean(c(x1,x2)),sd(c(x1,x2)))*card2,col=4) |
| 142 |
lod_score_vecs(x1,x2) |
| 143 |
}) |
| 144 |
|
| 145 |
dfadd = structure(function# Adding list to a dataframe. |
| 146 |
### Add a list \emph{l} to a dataframe \emph{df}. Create it if \emph{df} is \emph{NULL}. Return the dataframe \emph{df}.
|
| 147 |
(df, ##<< A dataframe |
| 148 |
l ##<< A list |
| 149 |
) {
|
| 150 |
if (is.null(df)) {
|
| 151 |
df = data.frame(l,stringsAsFactors=FALSE) |
| 152 |
} else {
|
| 153 |
df = rbind(df, data.frame(l,stringsAsFactors=FALSE)) |
| 154 |
} |
| 155 |
return(df) |
| 156 |
### Return the dataframe \emph{df}.
|
| 157 |
}, ex=function(){
|
| 158 |
## Here dataframe is NULL |
| 159 |
print(df) |
| 160 |
df = NULL |
| 161 |
|
| 162 |
# Initialize df |
| 163 |
df = dfadd(df, list(key1 = "value1", key2 = "value2")) |
| 164 |
print(df) |
| 165 |
|
| 166 |
# Adding elements to df |
| 167 |
df = dfadd(df, list(key1 = "value1'", key2 = "value2'")) |
| 168 |
print(df) |
| 169 |
}) |
| 170 |
|
| 171 |
|
| 172 |
sign_from_strand = function( |
| 173 |
### Get the sign of strand |
| 174 |
strands) {
|
| 175 |
apply(t(strands), 2, function(strand) { if (strand == "F") return(1) else return(-1)})
|
| 176 |
### If strand in forward then returns 1 else returns -1 |
| 177 |
} |
| 178 |
|
| 179 |
flat_reads = function( |
| 180 |
### Extract reads coordinates from TempleteFilter input sequence |
| 181 |
reads, ##<< TemplateFilter input reads |
| 182 |
nuc_width ##<< Width used to shift F and R reads. |
| 183 |
) {
|
| 184 |
F_flatted_reads = unlist(apply(t(reads[reads$V3=="F",]),2,function(r){rep(as.integer(r[2]), r[4])}))
|
| 185 |
R_flatted_reads = unlist(apply(t(reads[reads$V3=="R",]),2,function(r){rep(as.integer(r[2]), r[4])}))
|
| 186 |
flatted_reads = c(F_flatted_reads + rep(nuc_width/2, length(F_flatted_reads)), R_flatted_reads - rep(nuc_width/2, length(R_flatted_reads)) ) |
| 187 |
return(list(F_flatted_reads, R_flatted_reads, flatted_reads)) |
| 188 |
### Returns a list of F reads, R reads and joint/shifted F and R reads. |
| 189 |
} |
| 190 |
|
| 191 |
|
| 192 |
|
| 193 |
filter_tf_outputs = function(# Filter TemplateFilter outputs |
| 194 |
### This function filters TemplateFilter outputs according, not only genome area observerved properties, but also correlation and overlap threshold. |
| 195 |
tf_outputs, ##<< TemplateFilter outputs. |
| 196 |
chr, ##<< Chromosome observed, here chr is an integer. |
| 197 |
x_min, ##<< Coordinate of the first bp observed. |
| 198 |
x_max, ##<< Coordinate of the last bp observed. |
| 199 |
nuc_width = 160, ##<< Nucleosome width. |
| 200 |
ol_bp = 59, ##<< Overlap Threshold. |
| 201 |
corr_thres = 0.5 ##<< Correlation threshold. |
| 202 |
) {
|
| 203 |
if (x_min < 0) {
|
| 204 |
tf_outputs = tf_outputs[tf_outputs$chr == paste("chr", chr, sep="") & tf_outputs$center > (-x_max - nuc_width/2) & tf_outputs$center < (-x_min + nuc_width/2),]
|
| 205 |
} else {
|
| 206 |
tf_outputs = tf_outputs[tf_outputs$chr == paste("chr", chr, sep="") & tf_outputs$center > (x_min - nuc_width/2) & tf_outputs$center < (x_max + nuc_width/2),]
|
| 207 |
} |
| 208 |
tf_outputs$lower_bound = tf_outputs$center - tf_outputs$width/2 |
| 209 |
tf_outputs$upper_bound = tf_outputs$center + tf_outputs$width/2 |
| 210 |
tf_outputs = tf_outputs[tf_outputs$correlation >= corr_thres,] |
| 211 |
tf_outputs = tf_outputs[order(tf_outputs$correlation,decreasing=TRUE),] |
| 212 |
i = 1 |
| 213 |
while (i <= length(tf_outputs[,1])) {
|
| 214 |
lb = tf_outputs[i,]$low |
| 215 |
ub = tf_outputs[i,]$up |
| 216 |
tf_outputs = tf_outputs[!(tf_outputs$low <= (ub-ol_bp) & tf_outputs$up > ub) & !(tf_outputs$up >= (lb+ol_bp) & tf_outputs$low < lb),] |
| 217 |
i = i+1 |
| 218 |
} |
| 219 |
return(tf_outputs) |
| 220 |
### Returns filtered TemplateFilter Outputs |
| 221 |
} |
| 222 |
|
| 223 |
|
| 224 |
|
| 225 |
filter_tf_inputs = function(# Filter TemplateFilter inputs |
| 226 |
### This function filters TemplateFilter inputs according genome area observed properties. It takes into account reads that are at the frontier of this area and the strand of these reads. |
| 227 |
inputs, ##<< TF inputs to be filtered. |
| 228 |
chr, ##<< Chromosome observed, here chr is an integer. |
| 229 |
x_min, ##<< Coordinate of the first bp observed. |
| 230 |
x_max, ##<< Coordinate of the last bp observed. |
| 231 |
nuc_width = 160, ##<< Nucleosome width. |
| 232 |
only_f = FALSE, ##<< Filter only F reads. |
| 233 |
only_r = FALSE ##<< Filter only R reads. |
| 234 |
) {
|
| 235 |
if (only_f) {
|
| 236 |
inputs = inputs[inputs[,1]==chr & inputs[,2] >= x_min - nuc_width & inputs[,3] == "F" & inputs[,2] <= x_max + nuc_width,] |
| 237 |
} else if (only_r) {
|
| 238 |
inputs = inputs[inputs[,1]==chr & inputs[,2] >= x_min - nuc_width & inputs[,3] == "R" & inputs[,2] <= x_max + nuc_width,] |
| 239 |
} else {
|
| 240 |
inputs = inputs[inputs[,1]==chr & inputs[,2] >= x_min - nuc_width & inputs[,2] <= x_max + nuc_width,] |
| 241 |
} |
| 242 |
corrected_inputs_coords = inputs[,2] + nuc_width/2 * sign_from_strand(inputs[,3]) |
| 243 |
inputs = inputs[inputs[,1]==chr & corrected_inputs_coords >= x_min & corrected_inputs_coords <= x_max,] |
| 244 |
return(inputs) |
| 245 |
### Returns filtred inputs. |
| 246 |
} |
| 247 |
|
| 248 |
|
| 249 |
|
| 250 |
get_comp_strand = function( |
| 251 |
### Compute the complementatry strand. |
| 252 |
strand ##<< The original strand. |
| 253 |
) {
|
| 254 |
apply(t(strand),2, function(n){
|
| 255 |
if (n=="a") {return("t")}
|
| 256 |
if (n=="t") {return("a")}
|
| 257 |
if (n=="c") {return("g")}
|
| 258 |
if (n=="g") {return("c")}
|
| 259 |
}) |
| 260 |
### Returns the complementatry strand. |
| 261 |
} |
| 262 |
|
| 263 |
|
| 264 |
aggregate_intra_strain_nucs = structure(function(# Aggregate replicated sample's nucleosomes. |
| 265 |
### This function aggregates nucleosome for replicated samples. It uses TemplateFilter ouput of each sample as replicate. Each sample owns a set of nucleosomes computed using TemplateFilter and ordered by the position of their center. Adajacent nucleosomes are compared two by two. Comparison is based on a log likelihood ratio score. The issue of comparison is adjacents nucleosomes merge or separation. Finally the function returns a list of clusters and all computed \emph{lod_scores}. Each cluster ows an attribute \emph{wp} for "well positionned". This attribute is set as \emph{TRUE} if the cluster is composed of exactly one nucleosomes of each sample.
|
| 266 |
samples, ##<< A list of samples. Each sample is a list like \emph{sample = list(id=..., marker=..., strain=..., roi=..., inputs=..., outputs=...)} with \emph{roi = list(name=..., begin=..., end=..., chr=..., genome=...)}.
|
| 267 |
lod_thres=-20, ##<< Log likelihood ration threshold. |
| 268 |
coord_max=20000000 ##<< A too big value to be a coord for a nucleosome lower bound. |
| 269 |
){
|
| 270 |
end_of_tracks = function(tracks) {
|
| 271 |
if (length(tracks) == 0) {
|
| 272 |
return(TRUE) |
| 273 |
} |
| 274 |
for (lower_bound in tracks) {
|
| 275 |
if(!is.na(lower_bound)) {
|
| 276 |
if (lower_bound < coord_max) {
|
| 277 |
return(FALSE) |
| 278 |
} |
| 279 |
} |
| 280 |
} |
| 281 |
return(TRUE) |
| 282 |
} |
| 283 |
store_cluster = function(clusters, new_cluster, nb_nucs_in_cluster,nuc_from_track,nb_tracks, min_nuc_center, max_nuc_center) {
|
| 284 |
if ( nb_nucs_in_cluster==nb_tracks & sum(nuc_from_track)==nb_tracks) {
|
| 285 |
new_cluster$wp = TRUE |
| 286 |
center = (new_cluster$lower_bound + new_cluster$upper_bound) / 2 |
| 287 |
if (is.null(min_nuc_center) | ((min_nuc_center <= center) & (center < max_nuc_center))) {
|
| 288 |
clusters[[length(clusters) + 1]] = new_cluster |
| 289 |
# print(new_cluster) |
| 290 |
} |
| 291 |
} else {
|
| 292 |
new_cluster$wp = FALSE |
| 293 |
center = (new_cluster$lower_bound + new_cluster$upper_bound) / 2 |
| 294 |
if (is.null(min_nuc_center) | ((min_nuc_center <= center) & (center < max_nuc_center))) {
|
| 295 |
clusters[[length(clusters) + 1]] = new_cluster |
| 296 |
} |
| 297 |
} |
| 298 |
return(clusters) |
| 299 |
} |
| 300 |
strain = samples[[1]]$strain |
| 301 |
lod_scores = c() |
| 302 |
min_nuc_center = min(samples[[1]]$roi$begin, samples[[1]]$roi$end) |
| 303 |
max_nuc_center = max(samples[[1]]$roi$begin, samples[[1]]$roi$end) |
| 304 |
# compute clusters |
| 305 |
clusters = list() |
| 306 |
cluster_contents = list() |
| 307 |
# Init reader |
| 308 |
indexes = c() |
| 309 |
track_readers = c() |
| 310 |
current_nuc = NULL |
| 311 |
lod_score = lod_thres - 1 |
| 312 |
# Read nucs from TF outputs |
| 313 |
tf_outs = list() |
| 314 |
i = 1 |
| 315 |
for (sample in samples) {
|
| 316 |
# print(sample$roi$chr) |
| 317 |
# print(min_nuc_center) |
| 318 |
# print(max_nuc_center) |
| 319 |
# print(sample$outputs) |
| 320 |
# tf_outs[[i]] = filter_tf_outputs(sample$outputs, sample$roi$chr, min_nuc_center, max_nuc_center) |
| 321 |
# print(tf_outs[[i]]) |
| 322 |
tf_outs[[i]] = sample$outputs |
| 323 |
tf_outs[[i]] = tf_outs[[i]][order(tf_outs[[i]]$center),] |
| 324 |
indexes[i] = 1 |
| 325 |
if (is.na(tf_outs[[i]][indexes[i],]$center)) {
|
| 326 |
track_readers[i] = coord_max |
| 327 |
} else {
|
| 328 |
track_readers[i] = tf_outs[[i]][indexes[i],]$center |
| 329 |
} |
| 330 |
i = i+1 |
| 331 |
} |
| 332 |
# print(track_readers) |
| 333 |
new_cluster = NULL |
| 334 |
nb_nucs_in_cluster = 0 |
| 335 |
nuc_from_track = c() |
| 336 |
for (i in 1:length(tf_outs)){
|
| 337 |
nuc_from_track[i] = FALSE |
| 338 |
} |
| 339 |
# Start clustering |
| 340 |
while (!end_of_tracks(track_readers)) {
|
| 341 |
new_center = min(track_readers) |
| 342 |
current_track = which(track_readers == new_center)[1] |
| 343 |
new_nuc = as.list(tf_outs[[current_track]][indexes[current_track],]) |
| 344 |
new_nuc$chr = substr(new_nuc$chr,4,1000000L) |
| 345 |
new_nuc$inputs = samples[[current_track]]$inputs |
| 346 |
new_nuc$chr = samples[[current_track]]$roi$chr |
| 347 |
new_nuc$track = current_track |
| 348 |
|
| 349 |
new_nuc$inputs = filter_tf_inputs(samples[[current_track]]$inputs, new_nuc$chr, new_nuc$lower_bound, new_nuc$upper_bound, new_nuc$width) |
| 350 |
flatted_reads = flat_reads(new_nuc$inputs, new_nuc$width) |
| 351 |
new_nuc$original_reads = flatted_reads[[3]] |
| 352 |
|
| 353 |
new_upper_bound = new_nuc$upper_bound |
| 354 |
|
| 355 |
if (!is.null(current_nuc)) {
|
| 356 |
lod_score = lod_score_vecs(current_nuc$original_reads,new_nuc$original_reads) |
| 357 |
lod_scores = c(lod_scores,lod_score) |
| 358 |
} |
| 359 |
# print(paste(lod_score, length(current_nuc$original_reads), length(new_nuc$original_reads), sep=" ")) |
| 360 |
if (is.na(lod_score)) {
|
| 361 |
lod_score = lod_thres - 1 |
| 362 |
} |
| 363 |
if (lod_score > lod_thres) {
|
| 364 |
# Store lod_score |
| 365 |
new_nuc$lod_score = lod_score |
| 366 |
# aggregate to current cluster |
| 367 |
# update bound |
| 368 |
if (new_nuc$upper_bound > new_cluster$upper_bound) {
|
| 369 |
new_cluster$upper_bound = new_nuc$upper_bound |
| 370 |
} |
| 371 |
if (new_nuc$lower_bound < new_cluster$lower_bound) {
|
| 372 |
new_cluster$lower_bound = new_nuc$lower_bound |
| 373 |
} |
| 374 |
# add nucleosome to current cluster |
| 375 |
nuc_from_track[current_track] = TRUE |
| 376 |
nb_nucs_in_cluster = nb_nucs_in_cluster + 1 |
| 377 |
new_cluster$nucs[[length(new_cluster$nucs)+1]] = new_nuc |
| 378 |
} else {
|
| 379 |
if (!is.null(new_cluster)) {
|
| 380 |
# store old cluster |
| 381 |
clusters = store_cluster(clusters, new_cluster, nb_nucs_in_cluster,nuc_from_track,length(tf_outs),min_nuc_center, max_nuc_center) |
| 382 |
} |
| 383 |
# Reinit current cluster composition stuff |
| 384 |
nb_nucs_in_cluster = 0 |
| 385 |
nuc_from_track = c() |
| 386 |
for (i in 1:length(tf_outs)){
|
| 387 |
nuc_from_track[i] = FALSE |
| 388 |
} |
| 389 |
# create new cluster |
| 390 |
new_cluster = list(lower_bound=new_nuc$low, upper_bound=new_nuc$up, chr=new_nuc$chr, strain_ref=strain , nucs=list()) |
| 391 |
# update upper bound |
| 392 |
current_upper_bound = new_upper_bound |
| 393 |
# add nucleosome to current cluster |
| 394 |
nb_nucs_in_cluster = nb_nucs_in_cluster + 1 |
| 395 |
nuc_from_track[current_track] = TRUE |
| 396 |
new_cluster$nucs[[length(new_cluster$nucs)+1]] = new_nuc |
| 397 |
|
| 398 |
} |
| 399 |
|
| 400 |
current_nuc = new_nuc |
| 401 |
|
| 402 |
# update indexes |
| 403 |
if (indexes[current_track] < length(tf_outs[[current_track]]$center)) {
|
| 404 |
indexes[current_track] = indexes[current_track] + 1 |
| 405 |
# update track |
| 406 |
track_readers[current_track] = tf_outs[[current_track]][indexes[current_track],]$center |
| 407 |
} else {
|
| 408 |
# update track |
| 409 |
track_readers[current_track] = coord_max |
| 410 |
} |
| 411 |
} |
| 412 |
# store last cluster |
| 413 |
if (!is.null(new_cluster)) {
|
| 414 |
# store old cluster |
| 415 |
clusters = store_cluster(clusters, new_cluster, nb_nucs_in_cluster,nuc_from_track,length(tf_outs),min_nuc_center, max_nuc_center) |
| 416 |
} |
| 417 |
return(list(clusters, lod_scores)) |
| 418 |
### Returns a list of clusterized nucleosomes, and all computed lod scores. |
| 419 |
}, ex=function(){
|
| 420 |
# Dealing with a region of interest |
| 421 |
roi =list(name="example", begin=1000, end=1300, chr="1", genome=rep("A",301))
|
| 422 |
samples = list() |
| 423 |
for (i in 1:3) {
|
| 424 |
# Create TF output |
| 425 |
tf_nuc = list("chr"=paste("chr", roi$chr, sep=""), "center"=(roi$end + roi$begin)/2, "width"= 150, "correlation.score"= 0.9)
|
| 426 |
outputs = dfadd(NULL,tf_nuc) |
| 427 |
outputs = filter_tf_outputs(outputs, roi$chr, roi$begin, roi$end) |
| 428 |
# Generate corresponding reads |
| 429 |
nb_reads = round(runif(1,170,230)) |
| 430 |
reads = round(rnorm(nb_reads, tf_nuc$center,20)) |
| 431 |
u_reads = sort(unique(reads)) |
| 432 |
strands = sample(c(rep("R",ceiling(length(u_reads)/2)),rep("F",floor(length(u_reads)/2))))
|
| 433 |
counts = apply(t(u_reads), 2, function(r) { sum(reads == r)})
|
| 434 |
shifts = apply(t(strands), 2, function(s) { if (s == "F") return(-tf_nuc$width/2) else return(tf_nuc$width/2)})
|
| 435 |
u_reads = u_reads + shifts |
| 436 |
inputs = data.frame(list("V1" = rep(roi$chr, length(u_reads)),
|
| 437 |
"V2" = u_reads, |
| 438 |
"V3" = strands, |
| 439 |
"V4" = counts), stringsAsFactors=FALSE) |
| 440 |
samples[[length(samples) + 1]] = list(id=1, marker="Mnase_Seq", strain="strain_ex", total_reads = 10000000, roi=roi, inputs=inputs, outputs=outputs) |
| 441 |
} |
| 442 |
print(aggregate_intra_strain_nucs(samples)) |
| 443 |
}) |
| 444 |
|
| 445 |
|
| 446 |
|
| 447 |
|
| 448 |
|
| 449 |
|
| 450 |
align_inter_strain_nucs = structure(function(# Aligns nucleosomes between 2 strains. |
| 451 |
### This function aligns nucs between two strains for a given genome region. |
| 452 |
replicates, ##<< Set of replicates, ideally 3 per strain. |
| 453 |
wp_nucs_strain_ref1=NULL, ##<< List of aggregates nucleosome for strain 1. If it's null this list will be computed. |
| 454 |
wp_nucs_strain_ref2=NULL, ##<< List of aggregates nucleosome for strain 2. If it's null this list will be computed. |
| 455 |
corr_thres=0.5, ##<< Correlation threshold. |
| 456 |
lod_thres=-100, ##<< LOD cut off. |
| 457 |
config=NULL, ##<< GLOBAL config variable |
| 458 |
... ##<< A list of parameters that will be passed to \emph{aggregate_intra_strain_nucs} if needed.
|
| 459 |
) {
|
| 460 |
if (length(replicates) < 2) {
|
| 461 |
stop("ERROR, align_inter_strain_nucs needs 2 replicate sets.")
|
| 462 |
} else if (length(replicates) > 2) {
|
| 463 |
print("WARNING, align_inter_strain_nucs will use 2 first sets of replicates as inputs.")
|
| 464 |
} |
| 465 |
common_nuc = NULL |
| 466 |
lod_scores = c() |
| 467 |
chr = replicates[[1]][[1]]$roi$chr |
| 468 |
min_nuc_center = min(replicates[[1]][[1]]$roi$begin, replicates[[1]][[1]]$roi$end) |
| 469 |
max_nuc_center = max(replicates[[1]][[1]]$roi$begin, replicates[[1]][[1]]$roi$end) |
| 470 |
strain_ref1 = replicates[[1]][[1]]$strain |
| 471 |
strain_ref2 = replicates[[2]][[1]]$strain |
| 472 |
big_roi = replicates[[1]][[1]]$roi |
| 473 |
if(big_roi$end - big_roi$begin < 0) {
|
| 474 |
tmp_begin = big_roi$begin |
| 475 |
big_roi$begin = big_roi$end |
| 476 |
big_roi$end = tmp_begin |
| 477 |
} |
| 478 |
# print(big_roi) |
| 479 |
# GO! |
| 480 |
if (is.null(wp_nucs_strain_ref1)) {
|
| 481 |
wp_nucs_strain_ref1 = aggregate_intra_strain_nucs(replicates[[1]], ...)[[1]] |
| 482 |
} |
| 483 |
if (is.null(wp_nucs_strain_ref2)) {
|
| 484 |
wp_nucs_strain_ref2 = aggregate_intra_strain_nucs(replicates[[2]], ...)[[1]] |
| 485 |
} |
| 486 |
# foo <<- wp_nucs_strain_ref1 |
| 487 |
# print(apply(t(wp_nucs_strain_ref1), 2, function(l){c(l[[1]]$lower_bound, l[[1]]$upper_bound, l[[1]]$wp)}))
|
| 488 |
# print(apply(t(wp_nucs_strain_ref2), 2, function(l){c(l[[1]]$lower_bound, l[[1]]$upper_bound, l[[1]]$wp)}))
|
| 489 |
# dealing with matching_nas |
| 490 |
lws = c() |
| 491 |
ups = c() |
| 492 |
for (na in wp_nucs_strain_ref2) {
|
| 493 |
lws = c(lws, na$lower_bound) |
| 494 |
ups = c(ups, na$upper_bound) |
| 495 |
} |
| 496 |
|
| 497 |
print(paste("Exploring chr" , chr , ", " , length(wp_nucs_strain_ref1) , ", [" , min_nuc_center , ", " , max_nuc_center , "] nucs...", sep=""))
|
| 498 |
roi_strain_ref1 = NULL |
| 499 |
roi_strain_ref2 = NULL |
| 500 |
if (length(wp_nucs_strain_ref1) > 0) {
|
| 501 |
for(index_nuc_strain_ref1 in 1:length(wp_nucs_strain_ref1)){
|
| 502 |
# print(paste("" , index_nuc_strain_ref1 , "/" , length(wp_nucs_strain_ref1), sep=""))
|
| 503 |
nuc_strain_ref1 = wp_nucs_strain_ref1[[index_nuc_strain_ref1]] |
| 504 |
# Filtering on Well Positionned |
| 505 |
if (nuc_strain_ref1$wp) {
|
| 506 |
roi_strain_ref1 = list(name=paste("strain_chr_id_" , strain_ref1 , "_" , chr , "_" , "i" , "_", sep=""), begin=nuc_strain_ref1$lower_bound, end=nuc_strain_ref1$upper_bound, chr=chr, strain_ref = strain_ref1)
|
| 507 |
roi_strain_ref2 = translate_roi(roi_strain_ref1, strain_ref2, big_roi, config) |
| 508 |
if (!is.null(roi_strain_ref2)){
|
| 509 |
# LOADING INTRA_STRAIN_NUCS_FILENAME_STRAIN_REF2 FILE(S) TO COMPUTE MATCHING_NAS (FILTER) |
| 510 |
lower_bound_roi_strain_ref2 = min(roi_strain_ref2$end,roi_strain_ref2$begin) |
| 511 |
upper_bound_roi_strain_ref2 = max(roi_strain_ref2$end,roi_strain_ref2$begin) |
| 512 |
matching_nas = which( lower_bound_roi_strain_ref2 <= ups & lws <= upper_bound_roi_strain_ref2) |
| 513 |
for (index_nuc_strain_ref2 in matching_nas) {
|
| 514 |
nuc_strain_ref2 = wp_nucs_strain_ref2[[index_nuc_strain_ref2]] |
| 515 |
# Filtering on Well Positionned |
| 516 |
if (nuc_strain_ref2$wp) {
|
| 517 |
# Filtering on correlation Score and collecting reads |
| 518 |
SKIP = FALSE |
| 519 |
# TODO: This for loop could be done before working on strain_ref2. Isn't it? |
| 520 |
reads_strain_ref1 = c() |
| 521 |
for (nuc in nuc_strain_ref1$nucs){
|
| 522 |
reads_strain_ref1 = c(reads_strain_ref1, nuc$original_reads) |
| 523 |
if (nuc$corr < corr_thres) {
|
| 524 |
SKIP = TRUE |
| 525 |
} |
| 526 |
} |
| 527 |
reads_strain_ref2 = c() |
| 528 |
for (nuc in nuc_strain_ref2$nucs){
|
| 529 |
reads_strain_ref2 = c(reads_strain_ref2, nuc$original_reads) |
| 530 |
if (nuc$corr < corr_thres) {
|
| 531 |
SKIP = TRUE |
| 532 |
} |
| 533 |
} |
| 534 |
# Filtering on correlation Score |
| 535 |
if (!SKIP) {
|
| 536 |
# tranlation of reads into strain 2 coords |
| 537 |
diff = ((roi_strain_ref1$begin + roi_strain_ref1$end) - (roi_strain_ref2$begin + roi_strain_ref2$end)) / 2 |
| 538 |
reads_strain_ref1 = reads_strain_ref1 - rep(diff, length(reads_strain_ref1)) |
| 539 |
lod_score = lod_score_vecs(reads_strain_ref1, reads_strain_ref2) |
| 540 |
lod_scores = c(lod_scores, lod_score) |
| 541 |
# Filtering on LOD Score |
| 542 |
if (lod_score > lod_thres) {
|
| 543 |
tmp_nuc = list() |
| 544 |
# strain_ref1 |
| 545 |
tmp_nuc[[paste("chr_", strain_ref1, sep="")]] = chr
|
| 546 |
tmp_nuc[[paste("lower_bound_", strain_ref1, sep="")]] = nuc_strain_ref1$lower_bound
|
| 547 |
tmp_nuc[[paste("upper_bound_", strain_ref1, sep="")]] = nuc_strain_ref1$upper_bound
|
| 548 |
tmp_nuc[[paste("mean_", strain_ref1, sep="")]] = signif(mean(reads_strain_ref1),5)
|
| 549 |
tmp_nuc[[paste("sd_", strain_ref1, sep="")]] = signif(sd(reads_strain_ref1),5)
|
| 550 |
tmp_nuc[[paste("nb_reads_", strain_ref1, sep="")]] = length(reads_strain_ref1)
|
| 551 |
tmp_nuc[[paste("index_nuc_", strain_ref1, sep="")]] = index_nuc_strain_ref1
|
| 552 |
# tmp_nuc[[paste("corr1_", strain_ref1, sep="")]] = signif(nuc_strain_ref1$nucs[[1]]$corr,5)
|
| 553 |
# tmp_nuc[[paste("corr2_", strain_ref1, sep="")]] = signif(nuc_strain_ref1$nucs[[2]]$corr,5)
|
| 554 |
# tmp_nuc[[paste("corr3_", strain_ref1, sep="")]] = signif(nuc_strain_ref1$nucs[[3]]$corr,5)
|
| 555 |
# strain_ref2 |
| 556 |
tmp_nuc[[paste("chr_", strain_ref2, sep="")]] = roi_strain_ref2$chr
|
| 557 |
tmp_nuc[[paste("lower_bound_", strain_ref2, sep="")]] = nuc_strain_ref2$lower_bound
|
| 558 |
tmp_nuc[[paste("upper_bound_", strain_ref2, sep="")]] = nuc_strain_ref2$upper_bound
|
| 559 |
tmp_nuc[[paste("means_", strain_ref2, sep="")]] = signif(mean(reads_strain_ref2),5)
|
| 560 |
tmp_nuc[[paste("sd_", strain_ref2, sep="")]] = signif(sd(reads_strain_ref2),5)
|
| 561 |
tmp_nuc[[paste("nb_reads_", strain_ref2, sep="")]] = length(reads_strain_ref2)
|
| 562 |
tmp_nuc[[paste("index_nuc_", strain_ref2, sep="")]] = index_nuc_strain_ref2
|
| 563 |
# tmp_nuc[[paste("corr1_", strain_ref2, sep="")]] = signif(nuc_strain_ref2$nucs[[1]]$corr,5)
|
| 564 |
# tmp_nuc[[paste("corr2_", strain_ref2, sep="")]] = signif(nuc_strain_ref2$nucs[[2]]$corr,5)
|
| 565 |
# tmp_nuc[[paste("corr3_", strain_ref2, sep="")]] = signif(nuc_strain_ref2$nucs[[3]]$corr,5)
|
| 566 |
# common |
| 567 |
tmp_nuc[["lod_score"]] = signif(lod_score,5) |
| 568 |
# print(tmp_nuc) |
| 569 |
common_nuc = dfadd(common_nuc, tmp_nuc) |
| 570 |
} |
| 571 |
} |
| 572 |
} |
| 573 |
} |
| 574 |
} else {
|
| 575 |
print("WARNING! No roi for strain ref 2.")
|
| 576 |
} |
| 577 |
} |
| 578 |
} |
| 579 |
|
| 580 |
if(length(unique(common_nuc[,1:3])[,1]) != length((common_nuc[,1:3])[,1])) {
|
| 581 |
index_redundant = which(apply(common_nuc[,1:3][-length(common_nuc[,1]),] == common_nuc[,1:3][-1,] ,1,sum) == 3) |
| 582 |
to_remove_list = c() |
| 583 |
for (i in 1:length(index_redundant)) {
|
| 584 |
if (common_nuc[index_redundant[i],15] < common_nuc[index_redundant[i]+1,15]) {
|
| 585 |
to_remove = index_redundant[i] |
| 586 |
} else {
|
| 587 |
to_remove = index_redundant[i] + 1 |
| 588 |
} |
| 589 |
to_remove_list = c(to_remove_list, to_remove) |
| 590 |
} |
| 591 |
common_nuc = common_nuc[-to_remove_list,] |
| 592 |
} |
| 593 |
|
| 594 |
if(length(unique(common_nuc[,8:10])[,1]) != length((common_nuc[,8:10])[,1])) {
|
| 595 |
index_redundant = which(apply(common_nuc[,8:10][-length(common_nuc[,1]),] == common_nuc[,8:10][-1,] ,1,sum) == 3) |
| 596 |
to_remove_list = c() |
| 597 |
for (i in 1:length(index_redundant)) {
|
| 598 |
if (common_nuc[index_redundant[i],15] < common_nuc[index_redundant[i]+1,15]) {
|
| 599 |
to_remove = index_redundant[i] |
| 600 |
} else {
|
| 601 |
to_remove = index_redundant[i] + 1 |
| 602 |
} |
| 603 |
to_remove_list = c(to_remove_list, to_remove) |
| 604 |
} |
| 605 |
common_nuc = common_nuc[-to_remove_list,] |
| 606 |
} |
| 607 |
|
| 608 |
return(list(common_nuc, lod_scores)) |
| 609 |
} else {
|
| 610 |
print("WARNING, no nucs for strain_ref1.")
|
| 611 |
return(NULL) |
| 612 |
} |
| 613 |
### Returns a list of clusterized nucleosomes, and all computed lod scores. |
| 614 |
}, ex=function(){
|
| 615 |
|
| 616 |
# Define new translate_roi function... |
| 617 |
translate_roi = function(roi, strain2, big_roi=NULL, config=NULL) {
|
| 618 |
return(roi) |
| 619 |
} |
| 620 |
# Binding it by uncomment follwing lines. |
| 621 |
unlockBinding("translate_roi", as.environment("package:nucleominer"))
|
| 622 |
unlockBinding("translate_roi", getNamespace("nucleominer"))
|
| 623 |
assign("translate_roi", translate_roi, "package:nucleominer")
|
| 624 |
assign("translate_roi", translate_roi, getNamespace("nucleominer"))
|
| 625 |
lockBinding("translate_roi", getNamespace("nucleominer"))
|
| 626 |
lockBinding("translate_roi", as.environment("package:nucleominer"))
|
| 627 |
|
| 628 |
# Dealing with a region of interest |
| 629 |
roi =list(name="example", begin=1000, end=1300, chr="1", genome=rep("A",301), strain_ref1 = "STRAINREF1")
|
| 630 |
roi2 = translate_roi(roi, roi$strain_ref1) |
| 631 |
replicates = list() |
| 632 |
for (j in 1:2) {
|
| 633 |
samples = list() |
| 634 |
for (i in 1:3) {
|
| 635 |
# Create TF output |
| 636 |
tf_nuc = list("chr"=paste("chr", roi$chr, sep=""), "center"=(roi$end + roi$begin)/2, "width"= 150, "correlation.score"= 0.9)
|
| 637 |
outputs = dfadd(NULL,tf_nuc) |
| 638 |
outputs = filter_tf_outputs(outputs, roi$chr, roi$begin, roi$end) |
| 639 |
# Generate corresponding reads |
| 640 |
nb_reads = round(runif(1,170,230)) |
| 641 |
reads = round(rnorm(nb_reads, tf_nuc$center,20)) |
| 642 |
u_reads = sort(unique(reads)) |
| 643 |
strands = sample(c(rep("R",ceiling(length(u_reads)/2)),rep("F",floor(length(u_reads)/2))))
|
| 644 |
counts = apply(t(u_reads), 2, function(r) { sum(reads == r)})
|
| 645 |
shifts = apply(t(strands), 2, function(s) { if (s == "F") return(-tf_nuc$width/2) else return(tf_nuc$width/2)})
|
| 646 |
u_reads = u_reads + shifts |
| 647 |
inputs = data.frame(list("V1" = rep(roi$chr, length(u_reads)),
|
| 648 |
"V2" = u_reads, |
| 649 |
"V3" = strands, |
| 650 |
"V4" = counts), stringsAsFactors=FALSE) |
| 651 |
samples[[length(samples) + 1]] = list(id=1, marker="Mnase_Seq", strain=paste("strain_ex",j,sep=""), total_reads = 10000000, roi=roi, inputs=inputs, outputs=outputs)
|
| 652 |
} |
| 653 |
replicates[[length(replicates) + 1]] = samples |
| 654 |
} |
| 655 |
print(align_inter_strain_nucs(replicates)) |
| 656 |
}) |
| 657 |
|
| 658 |
|
| 659 |
|
| 660 |
|
| 661 |
|
| 662 |
|
| 663 |
|
| 664 |
|
| 665 |
|
| 666 |
|
| 667 |
|
| 668 |
|
| 669 |
|
| 670 |
|
| 671 |
|
| 672 |
|
| 673 |
|
| 674 |
|
| 675 |
|
| 676 |
|
| 677 |
|
| 678 |
|
| 679 |
|
| 680 |
|
| 681 |
|
| 682 |
|
| 683 |
fetch_mnase_replicates = function(# Prefetch data |
| 684 |
### Fetch and filter inputs and outpouts per region of interest. Organize it per replicates. |
| 685 |
strain, ##<< The strain we want mnase replicatesList of replicates. Each replicates is a vector of sample ids. |
| 686 |
roi, ##<< Region of interest. |
| 687 |
all_samples, ##<< Global list of samples. |
| 688 |
config=NULL, ##<< GLOBAL config variable |
| 689 |
only_fetch=FALSE, ##<< If TRUE, only fetch and not filtering. It is used tio load sample files into memory before forking. |
| 690 |
get_genome=FALSE, ##<< If TRUE, load corresponding genome sequence. |
| 691 |
get_ouputs=TRUE##<< If TRUE, get also ouput corresponding TF output files. |
| 692 |
) {
|
| 693 |
samples=list() |
| 694 |
samples_ids = unique(all_samples[all_samples$marker == "Mnase_Seq" & all_samples$strain == strain,]$id) |
| 695 |
for (i in samples_ids) {
|
| 696 |
sample = as.list(all_samples[all_samples$id==i,]) |
| 697 |
sample$roi = translate_roi(roi, sample$strain, config) |
| 698 |
if (get_genome) {
|
| 699 |
# Get Genome |
| 700 |
sample$roi$genome = get_content(config$FASTA_REFERENCE_GENOME_FILES[[sample$strain]], "fasta")[[switch_pairlist(config$FASTA_INDEXES[[sample$strain]])[[sample$roi$chr]]]][sample$roi$begin:sample$roi$end] |
| 701 |
} |
| 702 |
# Get inputs |
| 703 |
sample$inputs = get_content(paste(config$ALIGN_DIR, "/TF/sample_", i, "_TF.txt", sep=""), "table", stringsAsFactors=FALSE) |
| 704 |
sample$total_reads = sum(sample$inputs[,4]) |
| 705 |
if (!only_fetch) {
|
| 706 |
sample$inputs = filter_tf_inputs(sample$inputs, sample$roi$chr, min(sample$roi$begin, sample$roi$end), max(sample$roi$begin, sample$roi$end), 300) |
| 707 |
} |
| 708 |
# Get TF outputs for Mnase_Seq samples |
| 709 |
if (sample$marker == "Mnase_Seq" & get_ouputs) {
|
| 710 |
sample$outputs = get_content(paste(config$ALIGN_DIR, "/TF/sample_", i, "_all_nucs.tab", sep=""), "table", header=TRUE, sep="\t") |
| 711 |
if (!only_fetch) {
|
| 712 |
sample$outputs = filter_tf_outputs(sample$outputs, sample$roi$chr, min(sample$roi$begin, sample$roi$end), max(sample$roi$begin, sample$roi$end), 300) |
| 713 |
} |
| 714 |
} |
| 715 |
samples[[length(samples) + 1]] = sample |
| 716 |
} |
| 717 |
return(samples) |
| 718 |
} |
| 719 |
|
| 720 |
substract_region = function(# Substract to a list of regions an other list of regions that intersect it. |
| 721 |
### This fucntion embed a recursive part. It occurs when a substracted region split an original region on two. |
| 722 |
region1, ##<< Original regions. |
| 723 |
region2 ##<< Regions to substract. |
| 724 |
) {
|
| 725 |
rec_substract_region = function(region1, region2) {
|
| 726 |
non_inter_fuzzy = apply(t(1:length(region1[,1])), 2, function(i) {
|
| 727 |
cur_fuzzy = region1[i,] |
| 728 |
inter_wp = region2[region2$lower_bound <= cur_fuzzy$upper_bound & region2$upper_bound >= cur_fuzzy$lower_bound,] |
| 729 |
if (length(inter_wp[,1]) > 0) {
|
| 730 |
ret = c() |
| 731 |
for (j in 1:length(inter_wp[,1])) {
|
| 732 |
cur_wp = inter_wp[j,] |
| 733 |
if (cur_wp$lower_bound <= cur_fuzzy$lower_bound & cur_fuzzy$upper_bound <= cur_wp$upper_bound) {
|
| 734 |
# remove cur_fuzzy |
| 735 |
ret = c() |
| 736 |
break |
| 737 |
} else if (cur_wp$lower_bound <= cur_fuzzy$lower_bound & cur_wp$upper_bound < cur_fuzzy$upper_bound) {
|
| 738 |
# crop fuzzy |
| 739 |
cur_fuzzy$lower_bound = cur_wp$upper_bound + 1 |
| 740 |
ret = cur_fuzzy |
| 741 |
} else if (cur_fuzzy$lower_bound < cur_wp$lower_bound & cur_fuzzy$upper_bound <= cur_wp$upper_bound) {
|
| 742 |
# crop fuzzy |
| 743 |
cur_fuzzy$upper_bound = cur_wp$lower_bound - 1 |
| 744 |
ret = cur_fuzzy |
| 745 |
} else if (cur_wp$lower_bound > cur_fuzzy$lower_bound & cur_wp$upper_bound < cur_fuzzy$upper_bound) {
|
| 746 |
# split fuzzy |
| 747 |
tmp_ret_fuzzy_1 = cur_fuzzy |
| 748 |
tmp_ret_fuzzy_1$upper_bound = cur_wp$lower_bound - 1 |
| 749 |
tmp_ret_fuzzy_2 = cur_fuzzy |
| 750 |
tmp_ret_fuzzy_2$lower_bound = cur_wp$upper_bound + 1 |
| 751 |
ret = rec_substract_region(rbind(tmp_ret_fuzzy_1, tmp_ret_fuzzy_2), inter_wp) |
| 752 |
# print(ret) |
| 753 |
# ret = cur_fuzzy |
| 754 |
break |
| 755 |
} else {
|
| 756 |
stop("WARNING NO ADAPTED CASE!")
|
| 757 |
} |
| 758 |
} |
| 759 |
return(ret) |
| 760 |
} else {
|
| 761 |
return(cur_fuzzy) |
| 762 |
} |
| 763 |
}) |
| 764 |
} |
| 765 |
non_inter_fuzzy = rec_substract_region(region1, region2) |
| 766 |
if (is.null(non_inter_fuzzy)) {return(non_inter_fuzzy)}
|
| 767 |
tmp_ulist = unlist(non_inter_fuzzy) |
| 768 |
tmp_names = names(tmp_ulist)[1:4] |
| 769 |
non_inter_fuzzy = data.frame(matrix(tmp_ulist,ncol=4, byrow=TRUE), stringsAsFactors=FALSE) |
| 770 |
names(non_inter_fuzzy) = tmp_names |
| 771 |
non_inter_fuzzy$lower_bound = as.numeric(non_inter_fuzzy$lower_bound) |
| 772 |
non_inter_fuzzy$upper_bound = as.numeric(non_inter_fuzzy$upper_bound) |
| 773 |
return(non_inter_fuzzy) |
| 774 |
} |
| 775 |
|
| 776 |
union_regions = function(# Aggregate regions that intersect themnselves. |
| 777 |
### This function is based on sort of lower bounds to detect regions that intersect. We compare lower bound and upper bound of the porevious item. This function embed a while loop and break break regions list become stable. |
| 778 |
regions ##<< The Regions to be aggregated |
| 779 |
) {
|
| 780 |
old_length = length(regions[,1]) |
| 781 |
new_length = 0 |
| 782 |
|
| 783 |
while (old_length != new_length) {
|
| 784 |
regions = regions[order(regions$lower_bound), ] |
| 785 |
regions$stop = !c(regions$lower_bound[-1] - regions$upper_bound[-length(regions$lower_bound)] <= 0, TRUE) |
| 786 |
vec_end_1 = which(regions$stop) |
| 787 |
if (length(vec_end_1) == 0) {
|
| 788 |
vec_end_1 = c(length(regions$stop)) |
| 789 |
} |
| 790 |
if (vec_end_1[length(vec_end_1)] != length(regions$stop)) {
|
| 791 |
vec_end_1 = c(vec_end_1, length(regions$stop)) |
| 792 |
} |
| 793 |
vec_beg_1 = c(1, vec_end_1[-length(vec_end_1)] + 1) |
| 794 |
union = apply(t(1:length(vec_beg_1)), 2, function(i) {
|
| 795 |
chr = regions$chr[vec_beg_1[i]] |
| 796 |
lower_bound = min(regions$lower_bound[vec_beg_1[i]:vec_end_1[i]]) |
| 797 |
upper_bound = max(regions$upper_bound[vec_beg_1[i]:vec_end_1[i]]) |
| 798 |
roi_index = regions$roi_index[vec_beg_1[i]] |
| 799 |
data.frame(list(chr=chr, lower_bound=lower_bound, upper_bound=upper_bound, roi_index=roi_index)) |
| 800 |
}) |
| 801 |
union = do.call("rbind", union)
|
| 802 |
union$lower_bound = as.numeric(union$lower_bound) |
| 803 |
union$upper_bound = as.numeric(union$upper_bound) |
| 804 |
old_length = length(regions[,1]) |
| 805 |
new_length = length(union[,1]) |
| 806 |
regions = union |
| 807 |
} |
| 808 |
return(union) |
| 809 |
} |
| 810 |
|
| 811 |
remove_aligned_wp = function(# Remove wp nucs from common nucs list. |
| 812 |
### It is based on common wp nucs index on nucs and region. |
| 813 |
strain_maps, ##<< Nuc maps. |
| 814 |
roi_index, ##<< The region of interest index. |
| 815 |
tmp_common_nucs, ##<< the list of wp nucs. |
| 816 |
strain##<< The strain to consider. |
| 817 |
){
|
| 818 |
fuzzy_nucs = strain_maps[[strain]] |
| 819 |
fuzzy_nucs = fuzzy_nucs[fuzzy_nucs$roi_index == roi_index,] |
| 820 |
fuzzy_nucs = fuzzy_nucs[order(fuzzy_nucs$index_nuc),] |
| 821 |
if (length(fuzzy_nucs[,1]) == 0) {return(fuzzy_nucs)}
|
| 822 |
if (sum(fuzzy_nucs$index_nuc == min(fuzzy_nucs$index_nuc):max(fuzzy_nucs$index_nuc)) != max(fuzzy_nucs$index_nuc)) {"Warning in index!"}
|
| 823 |
anti_index_1 = tmp_common_nucs[[paste("index_nuc", strain, sep="_")]]
|
| 824 |
fuzzy_nucs = fuzzy_nucs[-anti_index_1,] |
| 825 |
return(fuzzy_nucs) |
| 826 |
} |
| 827 |
|
| 828 |
translate_regions = function(# Translate a list of regions from a strain ref to another. |
| 829 |
### This function is an eloborated call to translate_roi. |
| 830 |
regions, ##<< Regions to be translated. |
| 831 |
combi, ##<< Combination of strains. |
| 832 |
roi_index, ##<< The region of interest index. |
| 833 |
config=NULL, ##<< GLOBAL config variable |
| 834 |
roi ##<< The region of interest. |
| 835 |
) {
|
| 836 |
tr_regions = apply(t(1:length(regions[,1])), 2, function(i) {
|
| 837 |
tmp_regions_ref2 = list(name="foo", begin=regions[i,]$lower_bound, end=regions[i,]$upper_bound, chr=as.character(regions[i,]$chr), strain_ref = combi[2]) |
| 838 |
big_roi = translate_roi(roi, tmp_regions_ref2$strain_ref, config) |
| 839 |
tmp_min = min(big_roi$begin, big_roi$end) |
| 840 |
tmp_max = max(big_roi$begin, big_roi$end) |
| 841 |
big_roi$begin = tmp_min |
| 842 |
big_roi$end = tmp_max |
| 843 |
trs_tmp_regions_ref2 = translate_roi(tmp_regions_ref2, combi[1], config, big_roi) |
| 844 |
data.frame(list(chr=trs_tmp_regions_ref2$chr, lower_bound=min(trs_tmp_regions_ref2$begin, trs_tmp_regions_ref2$end), upper_bound=max(trs_tmp_regions_ref2$begin, trs_tmp_regions_ref2$end), roi_index=roi_index)) |
| 845 |
}) |
| 846 |
tr_regions = do.call("rbind", tr_regions)
|
| 847 |
tr_regions$lower_bound = as.numeric(tr_regions$lower_bound) |
| 848 |
tr_regions$upper_bound = as.numeric(tr_regions$upper_bound) |
| 849 |
tr_regions = tr_regions[order(tr_regions$lower_bound),] |
| 850 |
return(tr_regions) |
| 851 |
} |
| 852 |
|
| 853 |
extract_wp = function(# Extract wp nucs from nuc map. |
| 854 |
### Function based on common wp nuc index and roi_index. |
| 855 |
strain_maps, ##<< Nuc maps. |
| 856 |
roi_index, ##<< The region of interest index. |
| 857 |
strain, ##<< The strain to consider. |
| 858 |
tmp_common_nucs ##<< the list of wp nucs. |
| 859 |
) {
|
| 860 |
wp_nucs = apply(t(tmp_common_nucs[[paste("index_nuc", strain, sep="_")]]), 2, function(i) {
|
| 861 |
tmp_wp_nucs = strain_maps[[strain]] |
| 862 |
tmp_wp_nucs = tmp_wp_nucs[tmp_wp_nucs$roi_index == roi_index & tmp_wp_nucs$index_nuc == i,] |
| 863 |
return(tmp_wp_nucs) |
| 864 |
}) |
| 865 |
wp_nucs = do.call("rbind", wp_nucs)
|
| 866 |
wp_nucs$lower_bound = as.numeric(wp_nucs$lower_bound) |
| 867 |
wp_nucs$upper_bound = as.numeric(wp_nucs$upper_bound) |
| 868 |
wp_nucs = wp_nucs[order(wp_nucs$lower_bound),] |
| 869 |
return(wp_nucs) |
| 870 |
} |
| 871 |
|
| 872 |
|
| 873 |
|
| 874 |
|
| 875 |
|
| 876 |
crop_fuzzy = function(# Crop bound of regions according to region of interest bound |
| 877 |
### The fucntion is no more necessary since we remove "big_roi" bug in translate_roi function. |
| 878 |
tmp_fuzzy_nucs, ##<< the regiuons to be croped. |
| 879 |
roi, ##<< The region of interest. |
| 880 |
strain, ##<< The strain to consider. |
| 881 |
config=NULL ##<< GLOBAL config variable |
| 882 |
) {
|
| 883 |
tr_roi = translate_roi(roi, strain, config) |
| 884 |
tr_roi_begin = min(tr_roi$begin, tr_roi$end) |
| 885 |
tr_roi_end = max(tr_roi$begin, tr_roi$end) |
| 886 |
if (length(tmp_fuzzy_nucs[tmp_fuzzy_nucs$lower_bound < tr_roi_begin,1]) > 0) {
|
| 887 |
tmp_fuzzy_nucs[tmp_fuzzy_nucs$lower_bound < tr_roi_begin,]$lower_bound = tr_roi_begin |
| 888 |
} |
| 889 |
if (length(tmp_fuzzy_nucs[tmp_fuzzy_nucs$upper_bound < tr_roi_begin,1]) > 0) {
|
| 890 |
tmp_fuzzy_nucs[tmp_fuzzy_nucs$upper_bound < tr_roi_begin,]$upper_bound = tr_roi_begin |
| 891 |
} |
| 892 |
if (length(tmp_fuzzy_nucs[tmp_fuzzy_nucs$lower_bound > tr_roi_end,1]) > 0) {
|
| 893 |
tmp_fuzzy_nucs[tmp_fuzzy_nucs$lower_bound > tr_roi_end,]$lower_bound = tr_roi_end |
| 894 |
} |
| 895 |
if (length(tmp_fuzzy_nucs[tmp_fuzzy_nucs$upper_bound > tr_roi_end,1]) > 0) {
|
| 896 |
tmp_fuzzy_nucs[tmp_fuzzy_nucs$upper_bound > tr_roi_end,]$upper_bound = tr_roi_end |
| 897 |
} |
| 898 |
tmp_fuzzy_nucs = tmp_fuzzy_nucs[tmp_fuzzy_nucs$upper_bound != tmp_fuzzy_nucs$lower_bound,] |
| 899 |
return(tmp_fuzzy_nucs) |
| 900 |
} |
| 901 |
|
| 902 |
|
| 903 |
get_fuzzy = function(# Compute the fuzzy nucs. |
| 904 |
### This function aggregate non common wp nucs for each strain and substract common wp nucs. It does not take care about the size of the resulting fuzzy regions. It will be take into account in the count read part og the pipeline. |
| 905 |
combi, ##<< The strain combination to consider. |
| 906 |
roi, ##<< The region of interest. |
| 907 |
roi_index, ##<< The region of interest index. |
| 908 |
strain_maps, ##<< Nuc maps. |
| 909 |
common_nuc_results, ##<< Common wp nuc maps |
| 910 |
config=NULL ##<< GLOBAL config variable |
| 911 |
) {
|
| 912 |
print(roi_index) |
| 913 |
PLOT = FALSE |
| 914 |
tmp_common_nucs = common_nuc_results[[paste(combi[1], combi[2], sep="_")]] |
| 915 |
tmp_common_nucs = tmp_common_nucs[tmp_common_nucs$roi_index == roi_index, ] |
| 916 |
|
| 917 |
print(paste("Dealing with fuzzy from", combi[1]))
|
| 918 |
tmp_fuzzy_nucs_1 = remove_aligned_wp(strain_maps, roi_index, tmp_common_nucs, combi[1]) |
| 919 |
tmp_fuzzy_nucs_1 = crop_fuzzy(tmp_fuzzy_nucs_1, roi, combi[1], config) |
| 920 |
if (length(tmp_fuzzy_nucs_1[,1]) == 0) {return(NULL)}
|
| 921 |
agg_fuzzy_1 = union_regions(tmp_fuzzy_nucs_1) |
| 922 |
if (PLOT) for (i in 1:length(agg_fuzzy_1[,1])) {
|
| 923 |
lines(c(agg_fuzzy_1[i,]$lower_bound, agg_fuzzy_1[i,]$upper_bound), c(+3.1,+3.1), col=2) |
| 924 |
} |
| 925 |
|
| 926 |
print(paste("Dealing with fuzzy from ", combi[2]))
|
| 927 |
tmp_fuzzy_nucs_2 = remove_aligned_wp(strain_maps, roi_index, tmp_common_nucs, combi[2]) |
| 928 |
if (length(tmp_fuzzy_nucs_2[,1]) == 0) {return(NULL)}
|
| 929 |
agg_fuzzy_2 = union_regions(tmp_fuzzy_nucs_2) |
| 930 |
agg_fuzzy_2 = crop_fuzzy(agg_fuzzy_2, roi, combi[2], config) |
| 931 |
tr_agg_fuzzy_2 = translate_regions(agg_fuzzy_2, combi, roi_index, roi) |
| 932 |
tr_agg_fuzzy_2 = crop_fuzzy(tr_agg_fuzzy_2, roi, combi[2], config) |
| 933 |
# tr_agg_fuzzy_2 = union_regions(tr_agg_fuzzy_2) |
| 934 |
if (PLOT) for (i in 1:length(tr_agg_fuzzy_2[,1])) {
|
| 935 |
lines(c(tr_agg_fuzzy_2[i,]$lower_bound, tr_agg_fuzzy_2[i,]$upper_bound), c(+3.3,+3.3), col=2) |
| 936 |
} |
| 937 |
|
| 938 |
print("Dealing with fuzzy from both...")
|
| 939 |
all_fuzzy = union_regions(rbind(agg_fuzzy_1, tr_agg_fuzzy_2)) |
| 940 |
if (PLOT) for (i in 1:length(all_fuzzy[,1])) {
|
| 941 |
lines(c(all_fuzzy[i,]$lower_bound, all_fuzzy[i,]$upper_bound), c(+3.2, +3.2), col=1) |
| 942 |
} |
| 943 |
|
| 944 |
print(paste("Dealing with wp from", combi[1]))
|
| 945 |
wp_nucs_1 = extract_wp(strain_maps, roi_index, combi[1], tmp_common_nucs) |
| 946 |
if (PLOT) for (i in 1:length(wp_nucs_1[,1])) {
|
| 947 |
lines(c(wp_nucs_1[i,]$lower_bound, wp_nucs_1[i,]$upper_bound), c(+3.5,+3.5), col=3) |
| 948 |
} |
| 949 |
|
| 950 |
print(paste("Dealing with wp from", combi[2]))
|
| 951 |
wp_nucs_2 = extract_wp(strain_maps, roi_index, combi[2], tmp_common_nucs) |
| 952 |
tr_wp_nucs_2 = translate_regions(wp_nucs_2, combi, roi_index, roi) |
| 953 |
if (PLOT) for (i in 1:length(tr_wp_nucs_2[,1])) {
|
| 954 |
lines(c(tr_wp_nucs_2[i,]$lower_bound, tr_wp_nucs_2[i,]$upper_bound), c(+3.7,+3.7), col=3) |
| 955 |
} |
| 956 |
|
| 957 |
print("Dealing with wp from both...")
|
| 958 |
all_wp = union_regions(rbind(wp_nucs_1[,1:4], tr_wp_nucs_2)) |
| 959 |
if (PLOT) for (i in 1:length(all_wp[,1])) {
|
| 960 |
lines(c(all_wp[i,]$lower_bound, all_wp[i,]$upper_bound), c(+3.6, +3.6), col=1) |
| 961 |
} |
| 962 |
|
| 963 |
print("Dealing with fuzzy and wp...")
|
| 964 |
non_inter_fuzzy = substract_region(all_fuzzy, all_wp) |
| 965 |
if (is.null(non_inter_fuzzy)) { return(NULL) }
|
| 966 |
|
| 967 |
non_inter_fuzzy$len = non_inter_fuzzy$upper_bound - non_inter_fuzzy$lower_bound |
| 968 |
# non_inter_fuzzy = non_inter_fuzzy[non_inter_fuzzy$len >= min_fuzz_width,] |
| 969 |
if (PLOT) for (i in 1:length(non_inter_fuzzy[,1])) {
|
| 970 |
lines(c(non_inter_fuzzy[i,]$lower_bound, non_inter_fuzzy[i,]$upper_bound), c(+3.9, +3.9), col=1) |
| 971 |
} |
| 972 |
|
| 973 |
non_inter_fuzzy$index_nuc = 1:length(non_inter_fuzzy[,1]) |
| 974 |
return (non_inter_fuzzy) |
| 975 |
} |
| 976 |
|
| 977 |
|
| 978 |
|
| 979 |
get_all_reads = function(# Retrieve Reads |
| 980 |
### Retrieve reads for a given marker, combi, form. |
| 981 |
marker, ##<< The marker to considere. |
| 982 |
combi, ##<< The starin combination to considere. |
| 983 |
form="wp" ##<< The nuc form to considere. |
| 984 |
) {
|
| 985 |
all_reads = NULL |
| 986 |
for (manip in c("Mnase_Seq", marker)) {
|
| 987 |
if (form == "fuzzy") {
|
| 988 |
out_filename = paste("results/2013-08/",combi[1],"_",combi[2],"_",manip,"_fuzzy_and_nbreads.tab",sep="")
|
| 989 |
tmp_res = read.table(file=out_filename, header=TRUE) |
| 990 |
tmp_res = tmp_res[tmp_res[,3] - tmp_res[,2] > 75,] |
| 991 |
tmp_res$form = form |
| 992 |
} else if (form == "wp") {
|
| 993 |
out_filename = paste("results/2013-08/",combi[1],"_",combi[2],"_",manip,"_wp_and_nbreads.tab",sep="")
|
| 994 |
tmp_res = read.table(file=out_filename, header=TRUE) |
| 995 |
tmp_res$form = form |
| 996 |
} else if (form == "wpfuzzy") {
|
| 997 |
out_filename = paste("results/2013-08/",combi[1],"_",combi[2],"_",manip,"_wp_and_nbreads.tab",sep="")
|
| 998 |
tmp_res = read.table(file=out_filename, header=TRUE) |
| 999 |
tmp_res$form = "wp" |
| 1000 |
out_filename = paste("results/2013-08/",combi[1],"_",combi[2],"_",manip,"_fuzzy_and_nbreads.tab",sep="")
|
| 1001 |
tmp_res2 = read.table(file=out_filename, header=TRUE) |
| 1002 |
tmp_res2 = tmp_res2[tmp_res2[,3] - tmp_res2[,2] > 75,] |
| 1003 |
tmp_res2$form = "fuzzy" |
| 1004 |
tmp_res = rbind(tmp_res, tmp_res2) |
| 1005 |
} |
| 1006 |
if (is.null(all_reads)) {
|
| 1007 |
all_reads = tmp_res[,c(1:9,length(tmp_res))] |
| 1008 |
} |
| 1009 |
tmp_res = tmp_res[,-c(1:9,length(tmp_res))] |
| 1010 |
all_reads = cbind(all_reads, tmp_res) |
| 1011 |
} |
| 1012 |
return(all_reads) |
| 1013 |
} |
| 1014 |
|
| 1015 |
get_design = function(# Build the design for deseq |
| 1016 |
### This function build the design according sample properties. |
| 1017 |
marker, ##<< The marker to considere. |
| 1018 |
combi, ##<< The starin combination to considere. |
| 1019 |
all_samples ##<< Global list of samples. |
| 1020 |
) {
|
| 1021 |
off1 = 0 |
| 1022 |
off2 = 0 |
| 1023 |
manips = c("Mnase_Seq", marker)
|
| 1024 |
design_rownames = c() |
| 1025 |
design_manip = c() |
| 1026 |
design_strain = c() |
| 1027 |
off2index = function(off) {
|
| 1028 |
switch(toString(off), |
| 1029 |
"1"=c(0,1,1), |
| 1030 |
"2"=c(1,0,1), |
| 1031 |
"3"=c(1,1,0), |
| 1032 |
c(1,1,1) |
| 1033 |
) |
| 1034 |
} |
| 1035 |
for (manip in manips) {
|
| 1036 |
tmp_samples = all_samples[ ((all_samples$strain == combi[1] | all_samples$strain == combi[2]) & all_samples$marker == manip), ] |
| 1037 |
tmp_samples = tmp_samples[order(tmp_samples$strain), ] |
| 1038 |
if (manip == "H3K4me1" & (off1 != 0 & off2 ==0 )) {
|
| 1039 |
tmp_samples = tmp_samples[c(off2index(off1), c(1,1)) == 1,] |
| 1040 |
} else {
|
| 1041 |
if (manip != "Mnase_Seq" & (off1 != 0 | off2 !=0)) {
|
| 1042 |
tmp_samples = tmp_samples[c(off2index(off1), off2index(off2)) == 1,] |
| 1043 |
} |
| 1044 |
} |
| 1045 |
design_manip = c(design_manip, rep(manip, length(tmp_samples$id))) |
| 1046 |
for (strain in combi) {
|
| 1047 |
cols = apply(t(tmp_samples[ (tmp_samples$strain == strain & tmp_samples$marker == manip), ]$id), 2, function(i){paste(strain, manip, i, sep="_")})
|
| 1048 |
design_strain = c(design_strain, rep(strain, length(cols))) |
| 1049 |
design_rownames = c(design_rownames, cols) |
| 1050 |
} |
| 1051 |
} |
| 1052 |
snep_design = data.frame( row.names=design_rownames, manip=design_manip, strain=design_strain) |
| 1053 |
return(snep_design) |
| 1054 |
} |
| 1055 |
|
| 1056 |
plot_dist_samples = function(# Plot the distribution of reads. |
| 1057 |
### This fuxntion use the deseq nomalization feature to compare qualitatively the distribution. |
| 1058 |
strain, ##<< The strain to considere. |
| 1059 |
marker, ##<< The marker to considere. |
| 1060 |
res, ##<< Data |
| 1061 |
all_samples, ##<< Global list of samples. |
| 1062 |
NEWPLOT = TRUE ##<< If FALSE the curve will be add to the current plot. |
| 1063 |
) {
|
| 1064 |
cols = apply(t(all_samples[ (all_samples$strain == strain & all_samples$marker == marker), ]$id), 2, function(i){paste(strain, marker, i, sep="_")})
|
| 1065 |
snepCountTable = res[,cols] |
| 1066 |
snepDesign = data.frame( |
| 1067 |
row.names = cols, |
| 1068 |
manip = rep(marker, length(cols)), |
| 1069 |
strain = rep(strain, length(cols)) |
| 1070 |
) |
| 1071 |
cdsFull = newCountDataSet(snepCountTable, snepDesign) |
| 1072 |
sizeFactors = estimateSizeFactors(cdsFull) |
| 1073 |
# print(sizeFactors[[1]]) |
| 1074 |
sample_ids = all_samples[ (all_samples$strain == strain & all_samples$marker == marker), ]$id |
| 1075 |
if (NEWPLOT) {
|
| 1076 |
plot(density(res[,paste(strain, marker, sample_ids[1], sep="_")] / sizeFactors[[1]][1]), col=0, main=paste(strain, marker)) |
| 1077 |
NEWPLOT = FALSE |
| 1078 |
} |
| 1079 |
for (it in 1:length(sample_ids)) {
|
| 1080 |
sample_id = sample_ids[it] |
| 1081 |
lines(density(res[,paste(strain, marker, sample_id, sep="_")] / sizeFactors[[1]][it]), col = it + 1, lty = it) |
| 1082 |
} |
| 1083 |
legend("topright", col=(1:length(sample_ids))+1, lty=1:length(sample_ids), legend=cols)
|
| 1084 |
} |
| 1085 |
|
| 1086 |
analyse_design = function(# Launch deseq methods. |
| 1087 |
### This function is based on deseq example. It mormalizes data, fit data to GLM model with and without interaction term and compare the two l;=models. |
| 1088 |
snep_design, ##<< The design to considere. |
| 1089 |
reads ##<< The data to considere. |
| 1090 |
) {
|
| 1091 |
snep_count_table = reads[, rownames(snep_design)] |
| 1092 |
cdsFull = newCountDataSet(snep_count_table, snep_design) |
| 1093 |
cdsFull1 = estimateDispersions(estimateSizeFactors(cdsFull), fitType="local", method="pooled", sharingMode="maximum") |
| 1094 |
fit1 = fitNbinomGLMs(cdsFull1, count ~ manip * strain) |
| 1095 |
|
| 1096 |
cdsFull0 = estimateDispersions(estimateSizeFactors(cdsFull), fitType="local", method="pooled", sharingMode="maximum") |
| 1097 |
fit0 = fitNbinomGLMs(cdsFull0, count ~ manip + strain) |
| 1098 |
|
| 1099 |
pvalsGLM = nbinomGLMTest( fit1, fit0 ) |
| 1100 |
return(list(fit1, fit0, snep_design, pvalsGLM)) |
| 1101 |
} |
| 1102 |
|
| 1103 |
|
| 1104 |
|
| 1105 |
|
| 1106 |
|
| 1107 |
|
| 1108 |
|
| 1109 |
|
| 1110 |
get_sneps = structure(function(# Compute the list of SNEPs for a given set of marker, strain combination and nuc form. |
| 1111 |
### This function uses |
| 1112 |
marker, ##<< The marker involved. |
| 1113 |
combi, ##<< The strain combination involved. |
| 1114 |
form, ##<< the nuc form involved. |
| 1115 |
all_samples ##<< Global list of samples. |
| 1116 |
) {
|
| 1117 |
# PRETREAT |
| 1118 |
d = get_design(marker, combi, all_samples) |
| 1119 |
reads = get_all_reads(marker, combi, form) |
| 1120 |
# RUN ANALYSE |
| 1121 |
tmp_analyse = analyse_design(d, reads) |
| 1122 |
# RESULTS |
| 1123 |
fit1 = tmp_analyse[[1]] |
| 1124 |
fit0 = tmp_analyse[[2]] |
| 1125 |
k = names(fit1) |
| 1126 |
reads[[k[2]]] = signif(fit1[[k[2]]], 5) |
| 1127 |
reads[[k[3]]] = signif(fit1[[k[3]]], 5) |
| 1128 |
reads[[k[4]]] = signif(fit1[[k[4]]], 5) |
| 1129 |
reads$pvalsGLM = signif(tmp_analyse[[4]], 5) |
| 1130 |
snep_design = tmp_analyse[[3]] |
| 1131 |
print(snep_design) |
| 1132 |
fdr = 0.0001 |
| 1133 |
thres = FDR(reads$pvalsGLM, fdr) |
| 1134 |
reads$snep_index = reads$pvalsGLM < thres |
| 1135 |
print(paste(sum(reads$snep_index), " SNEPs found for ", length(reads[,1])," nucs and ", fdr*100,"% of FDR.", sep = "")) |
| 1136 |
return(reads) |
| 1137 |
}, ex=function(){
|
| 1138 |
marker = "H3K4me1" |
| 1139 |
combi = c("BY", "YJM")
|
| 1140 |
form = "wpfuzzy" # "wp" | "fuzzy" | "wpfuzzy" |
| 1141 |
# foo = get_sneps(marker, combi, form) |
| 1142 |
# foo = get_sneps("H4K12ac", c("BY", "RM"), "wp")
|
| 1143 |
}) |
| 1144 |
|
| 1145 |
|
| 1146 |
|
| 1147 |
|
| 1148 |
|
| 1149 |
|
| 1150 |
|
| 1151 |
|
| 1152 |
|
| 1153 |
|
| 1154 |
|
| 1155 |
|
| 1156 |
|
| 1157 |
|
| 1158 |
|
| 1159 |
|
| 1160 |
|
| 1161 |
|
| 1162 |
|
| 1163 |
|
| 1164 |
|
| 1165 |
|
| 1166 |
|
| 1167 |
|
| 1168 |
ROM2ARAB = function(# Roman to Arabic pair list. |
| 1169 |
### Util to convert Roman to Arabic |
| 1170 |
){list(
|
| 1171 |
"I" = 1, |
| 1172 |
"II" = 2, |
| 1173 |
"III" = 3, |
| 1174 |
"IV" = 4, |
| 1175 |
"V" = 5, |
| 1176 |
"VI" = 6, |
| 1177 |
"VII" = 7, |
| 1178 |
"VIII" = 8, |
| 1179 |
"IX" = 9, |
| 1180 |
"X" = 10, |
| 1181 |
"XI" = 11, |
| 1182 |
"XII" = 12, |
| 1183 |
"XIII" = 13, |
| 1184 |
"XIV" = 14, |
| 1185 |
"XV" = 15, |
| 1186 |
"XVI" = 16, |
| 1187 |
"XVII" = 17, |
| 1188 |
"XVIII" = 18, |
| 1189 |
"XIX" = 19, |
| 1190 |
"XX" = 20 |
| 1191 |
)} |
| 1192 |
|
| 1193 |
switch_pairlist = structure(function(# Switch a pairlist |
| 1194 |
### Take a pairlist key:value and return the switched pairlist value:key. |
| 1195 |
l ##<< The pairlist to switch. |
| 1196 |
) {
|
| 1197 |
ret = list() |
| 1198 |
for (name in names(l)) {
|
| 1199 |
ret[[as.character(l[[name]])]] = name |
| 1200 |
} |
| 1201 |
ret |
| 1202 |
### The switched pairlist. |
| 1203 |
}, ex=function(){
|
| 1204 |
l = list(key1 = "value1", key2 = "value2") |
| 1205 |
print(switch_pairlist(l)) |
| 1206 |
}) |
| 1207 |
|
| 1208 |
ARAB2ROM = function(# Arabic to Roman pair list. |
| 1209 |
### Util to convert Arabicto Roman |
| 1210 |
){switch_pairlist(ROM2ARAB())}
|
| 1211 |
|
| 1212 |
|
| 1213 |
# translate_roi = structure(function(# Translate coords of a genome region. |
| 1214 |
# ### This function is used in the examples, usualy you have to define your own translation function and overwrite this one using \emph{unlockBinding} features. Please, refer to the example.
|
| 1215 |
# roi, ##<< Original genome region of interest. |
| 1216 |
# strain2, ##<< The strain in wich you want the genome region of interest. |
| 1217 |
# big_roi=NULL ##<< A largest region than roi use to filter c2c if it is needed. |
| 1218 |
# ) {
|
| 1219 |
# ### This function translate a genome region of interest from a strain coord system to an other. This is a minimal fucntion that will be called by \emph{align_inter_strain_nucs} and its exemnple, you need to overwrite it by your own fucntion.
|
| 1220 |
# strain1 = roi$strain_ref |
| 1221 |
# if (strain1 == strain2 | strain2 == "strain_ex2") {
|
| 1222 |
# return(roi) |
| 1223 |
# } else {
|
| 1224 |
# stop("ERROR, you need to overwrite your own function to convert convert strain coords. To binf the new function, have a
|
| 1225 |
# look in the translate_roi example.") |
| 1226 |
# } |
| 1227 |
# ### The translated genome region of interest. |
| 1228 |
# }, ex=function(){
|
| 1229 |
# # Define new translate_roi function... |
| 1230 |
# translate_roi = function(roi, strain2) {
|
| 1231 |
# strain1 = roi$strain_ref |
| 1232 |
# if (strain1 == strain2) {
|
| 1233 |
# return(roi) |
| 1234 |
# } else {
|
| 1235 |
# stop("Here is my new translate_roi function...")
|
| 1236 |
# } |
| 1237 |
# } |
| 1238 |
# # Binding it by uncomment follwing lines. |
| 1239 |
# # unlockBinding("translate_roi", as.environment("package:nm"))
|
| 1240 |
# # unlockBinding("translate_roi", getNamespace("nm"))
|
| 1241 |
# # assign("translate_roi", translate_roi, "package:nm")
|
| 1242 |
# # assign("translate_roi", translate_roi, getNamespace("nm"))
|
| 1243 |
# # lockBinding("translate_roi", getNamespace("nm"))
|
| 1244 |
# # lockBinding("translate_roi", as.environment("package:nm"))
|
| 1245 |
# }) |
| 1246 |
# |
| 1247 |
# |
| 1248 |
|
| 1249 |
|
| 1250 |
translate_roi = structure(function(# Translate coords of a genome region. |
| 1251 |
### This function is used in the examples, usualy you have to define your own translation function and overwrite this one using \emph{unlockBinding} features. Please, refer to the example.
|
| 1252 |
roi, ##<< Original genome region of interest. |
| 1253 |
strain2, ##<< The strain in wich you want the genome region of interest. |
| 1254 |
config=NULL, ##<< GLOBAL config variable |
| 1255 |
big_roi=NULL ##<< A largest region than roi use to filter c2c if it is needed. |
| 1256 |
) {
|
| 1257 |
strain1 = roi$strain_ref |
| 1258 |
reverse = (strain1=="RM" & strain2=="BY") | strain1=="YJM" |
| 1259 |
if (strain1 == strain2) {
|
| 1260 |
roi$length = roi$end - roi$begin + sign(roi$end - roi$begin) * 1 |
| 1261 |
return(roi) |
| 1262 |
} |
| 1263 |
# Launch c2c file |
| 1264 |
if (reverse) {
|
| 1265 |
c2c_file = list(filename=config$C2C_FILES[[paste(strain2, "-", strain1, sep="")]]) |
| 1266 |
} else {
|
| 1267 |
c2c_file = list(filename=config$C2C_FILES[[paste(strain1, "-", strain2, sep="")]]) |
| 1268 |
} |
| 1269 |
c2c = get_content(c2c_file$filename, "table", stringsAsFactors=FALSE) |
| 1270 |
# filtering it |
| 1271 |
c2c = c2c[c2c$V6=="-",] |
| 1272 |
# Reverse |
| 1273 |
if (reverse) {
|
| 1274 |
tmp_col = c2c$V1 |
| 1275 |
c2c$V1 = c2c$V7 |
| 1276 |
c2c$V7 = tmp_col |
| 1277 |
tmp_col = c2c$V2 |
| 1278 |
c2c$V2 = c2c$V9 |
| 1279 |
c2c$V9 = tmp_col |
| 1280 |
tmp_col = c2c$V3 |
| 1281 |
c2c$V3 = c2c$V10 |
| 1282 |
c2c$V10 = tmp_col |
| 1283 |
} |
| 1284 |
# Restrict c2c to big_roi |
| 1285 |
# if (FALSE) {
|
| 1286 |
if (!is.null(big_roi)) {
|
| 1287 |
if (roi$strain_ref == big_roi$strain_ref) {
|
| 1288 |
if (strain1 == "BY") {
|
| 1289 |
big_chro_1 = paste("chr", ARAB2ROM()[[big_roi$chr]], sep="")
|
| 1290 |
} else if (strain1 == "RM") {
|
| 1291 |
big_chro_1 = paste("supercontig_1.",big_roi$chr,sep="")
|
| 1292 |
} else if (strain1 == "YJM") {
|
| 1293 |
big_chro_1 = switch_pairlist(config$FASTA_INDEXES$YJM)[[big_roi$chr]] |
| 1294 |
} |
| 1295 |
big_begin_1 = big_roi$begin |
| 1296 |
big_end_1 = big_roi$end |
| 1297 |
c2c = c2c[c2c$V1==big_chro_1,] |
| 1298 |
if (length(c2c[c2c$V3<big_begin_1, 1] > 0)) {c2c[c2c$V3 < big_begin_1,c("V2", "V3") ] = big_begin_1}
|
| 1299 |
if (length(c2c[c2c$V2>big_end_1, 1] > 0)) {c2c[c2c$V2 > big_end_1, c("V2", "V3")] = big_end_1}
|
| 1300 |
c2c = c2c[c2c$V2 - c2c$V3 != 0,] |
| 1301 |
} else {
|
| 1302 |
stop("ERROR, big_roi and roi not in the same strain_ref")
|
| 1303 |
} |
| 1304 |
} |
| 1305 |
# Convert initial roi$chr into c2c format |
| 1306 |
if (strain1 == "BY") {
|
| 1307 |
chro_1 = paste("chr", ARAB2ROM()[[roi$chr]], sep="")
|
| 1308 |
} else if (strain1 == "RM") {
|
| 1309 |
chro_1 = paste("supercontig_1.",roi$chr,sep="")
|
| 1310 |
} else if (strain1 == "YJM") {
|
| 1311 |
chro_1 = switch_pairlist(config$FASTA_INDEXES$YJM)[[roi$chr]] |
| 1312 |
} |
| 1313 |
begin_1 = roi$begin |
| 1314 |
end_1 = roi$end |
| 1315 |
# Computing equivalent strain_2 alignment coordinates |
| 1316 |
if (reverse) {
|
| 1317 |
tmptransfostart = c2c[c2c$V1==chro_1 & ((c2c$V3>=begin_1 & c2c$V2<=begin_1 & c2c$V8==1) | (c2c$V2>=begin_1 & c2c$V3<=begin_1 & c2c$V8==-1)),] |
| 1318 |
tmptransfostop = c2c[c2c$V1==chro_1 & ((c2c$V3>=end_1 & c2c$V2<=end_1 & c2c$V8==1) | (c2c$V2>=end_1 & c2c$V3<=end_1 & c2c$V8==-1)),] |
| 1319 |
} else {
|
| 1320 |
tmptransfostart = c2c[c2c$V1==chro_1 & c2c$V3>=begin_1 & c2c$V2<=begin_1,] |
| 1321 |
tmptransfostop = c2c[c2c$V1==chro_1 & c2c$V3>=end_1 & c2c$V2<=end_1,] |
| 1322 |
} |
| 1323 |
# Never happend conditions ... |
| 1324 |
{
|
| 1325 |
if (length(tmptransfostart$V8) == 0) {
|
| 1326 |
# begin_1 is between to lines: shift begin_1 to the start of 2nd line. |
| 1327 |
tmp_c2c = c2c[c2c$V1==chro_1 & c2c$V2>=begin_1,] |
| 1328 |
begin_1 = sort(tmp_c2c$V2)[1] |
| 1329 |
if (reverse) {
|
| 1330 |
tmptransfostart = c2c[c2c$V1==chro_1 & ((c2c$V3>=begin_1 & c2c$V2<=begin_1 & c2c$V8==1) | (c2c$V2>=begin_1 & c2c$V3<=begin_1 & c2c$V8==-1)),] |
| 1331 |
} else {
|
| 1332 |
tmptransfostart = c2c[c2c$V1==chro_1 & c2c$V3>=begin_1 & c2c$V2<=begin_1,] |
| 1333 |
} |
| 1334 |
if (length(tmptransfostart$V8) == 0) {
|
| 1335 |
if (!is.null(big_roi)) {
|
| 1336 |
return(NULL) |
| 1337 |
tmptransfostart = c2c[c2c$V1==chro_1 & c2c$V3>=big_roi$begin & c2c$V2<=big_roi$begin,] |
| 1338 |
} else {
|
| 1339 |
# return(NULL) |
| 1340 |
# print(c2c[c2c$V1==chro_1 & c2c$V2<=end_1 & c2c$V3>=begin_1,]) |
| 1341 |
# print(c2c[c2c$V1==chro_1,]) |
| 1342 |
print(tmptransfostart) |
| 1343 |
print(tmptransfostop) |
| 1344 |
stop("Never happend condition 1.")
|
| 1345 |
} |
| 1346 |
} |
| 1347 |
} |
| 1348 |
if (length(tmptransfostop$V8) == 0) {
|
| 1349 |
# end_1 is between to lines: shift end_1 to the end of 2nd line. |
| 1350 |
tmp_c2c = c2c[c2c$V1==chro_1 & c2c$V3<=end_1,] |
| 1351 |
end_1 = sort(tmp_c2c$V3)[length(tmp_c2c$V2)] |
| 1352 |
if (reverse) {
|
| 1353 |
tmptransfostop = c2c[c2c$V1==chro_1 & ((c2c$V3>=end_1 & c2c$V2<=end_1 & c2c$V8==1) | (c2c$V2>=end_1 & c2c$V3<=end_1 & c2c$V8==-1)),] |
| 1354 |
} else {
|
| 1355 |
tmptransfostop = c2c[c2c$V1==chro_1 & c2c$V3>=end_1 & c2c$V2<=end_1,] |
| 1356 |
} |
| 1357 |
if (length(tmptransfostop$V8) == 0) {
|
| 1358 |
if (!is.null(big_roi)) {
|
| 1359 |
return(NULL) |
| 1360 |
tmptransfostop = c2c[c2c$V1==chro_1 & c2c$V3>=big_roi$end & c2c$V2<=big_roi$end,] |
| 1361 |
} else {
|
| 1362 |
# return(NULL) |
| 1363 |
print(c2c[c2c$V1==chro_1,]) |
| 1364 |
print(tmptransfostart) |
| 1365 |
print(tmptransfostop) |
| 1366 |
stop("Never happend condition 2.")
|
| 1367 |
} |
| 1368 |
} |
| 1369 |
} |
| 1370 |
if (length(tmptransfostart$V8) != 1) {
|
| 1371 |
# tmptransfostart = tmptransfostart[1,] |
| 1372 |
# print("many start")
|
| 1373 |
# print(c2c[c2c$V1==chro_1,]) |
| 1374 |
tmptransfostart = tmptransfostart[tmptransfostart$V1==chro_1 & tmptransfostart$V3>=begin_1 & tmptransfostart$V2==begin_1,] |
| 1375 |
if (length(tmptransfostart$V8) != 1) {
|
| 1376 |
# return(NULL) |
| 1377 |
print(tmptransfostart) |
| 1378 |
print(tmptransfostop) |
| 1379 |
stop("Never happend condition 3.")
|
| 1380 |
} |
| 1381 |
} |
| 1382 |
if (length(tmptransfostop$V8) != 1) {
|
| 1383 |
# tmptransfostop = tmptransfostop[length(tmptransfostop$V8),] |
| 1384 |
# print("many stop")
|
| 1385 |
# print(tmptransfostop) |
| 1386 |
# print(roi) |
| 1387 |
tmptransfostop = tmptransfostop[tmptransfostop$V1==chro_1 & tmptransfostop$V3==end_1 & tmptransfostop$V2<=end_1,] |
| 1388 |
if (length(tmptransfostop$V8) != 1) {
|
| 1389 |
# return(NULL) |
| 1390 |
print(tmptransfostart) |
| 1391 |
print(tmptransfostop) |
| 1392 |
stop("Never happend condition 4.")
|
| 1393 |
} |
| 1394 |
} |
| 1395 |
if (tmptransfostart$V7 != tmptransfostop$V7) {
|
| 1396 |
print(tmptransfostart) |
| 1397 |
print(tmptransfostop) |
| 1398 |
stop("Problem with genome region of interest of strain 1. \nIt is translated over many contigs into strain 2 ref. \nSorry, but you have to redefine your region of interest.")
|
| 1399 |
} |
| 1400 |
} |
| 1401 |
# Deal with strand |
| 1402 |
if (tmptransfostart$V8 == 1) {
|
| 1403 |
begin_2 = tmptransfostart$V9 + (begin_1 - tmptransfostart$V2) |
| 1404 |
end_2 = tmptransfostop$V9 + (end_1 - tmptransfostop$V2) |
| 1405 |
} else {
|
| 1406 |
begin_2 = tmptransfostart$V9 - (begin_1 - tmptransfostart$V2) |
| 1407 |
end_2 = tmptransfostop$V9 - (end_1 - tmptransfostop$V2) |
| 1408 |
} |
| 1409 |
# Build returned roi |
| 1410 |
roi$strain_ref = strain2 |
| 1411 |
if (roi$strain_ref == "BY") {
|
| 1412 |
roi$chr = ROM2ARAB()[[substr(tmptransfostart$V7, 4, 12)]] |
| 1413 |
} else {
|
| 1414 |
roi$chr = config$FASTA_INDEXES[[strain2]][[tmptransfostart$V7]] |
| 1415 |
} |
| 1416 |
roi$begin = begin_2 |
| 1417 |
roi$end = end_2 |
| 1418 |
if (sign(roi$end - roi$begin) == 0) {
|
| 1419 |
roi$length = 1 |
| 1420 |
} else {
|
| 1421 |
roi$length = roi$end - roi$begin + sign(roi$end - roi$begin) * 1 |
| 1422 |
} |
| 1423 |
return(roi) |
| 1424 |
}, ex=function(){
|
| 1425 |
# Define new translate_roi function... |
| 1426 |
translate_roi = function(roi, strain2, config) {
|
| 1427 |
strain1 = roi$strain_ref |
| 1428 |
if (strain1 == strain2) {
|
| 1429 |
return(roi) |
| 1430 |
} else {
|
| 1431 |
stop("Here is my new translate_roi function...")
|
| 1432 |
} |
| 1433 |
} |
| 1434 |
# Binding it by uncomment follwing lines. |
| 1435 |
# unlockBinding("translate_roi", as.environment("package:nm"))
|
| 1436 |
# unlockBinding("translate_roi", getNamespace("nm"))
|
| 1437 |
# assign("translate_roi", translate_roi, "package:nm")
|
| 1438 |
# assign("translate_roi", translate_roi, getNamespace("nm"))
|
| 1439 |
# lockBinding("translate_roi", getNamespace("nm"))
|
| 1440 |
# lockBinding("translate_roi", as.environment("package:nm"))
|
| 1441 |
}) |
| 1442 |
|
| 1443 |
|
| 1444 |
|
| 1445 |
|
| 1446 |
|
| 1447 |
|
| 1448 |
|
| 1449 |
compute_inter_all_strain_curs = function (# Compute Common Uninterrupted Regions (CUR) |
| 1450 |
### CURs are regions that can be aligned between the genomes |
| 1451 |
diff_allowed = 10, ##<< the maximum indel width allowe din a CUR |
| 1452 |
min_cur_width = 200, ##<< The minimum width of a CUR |
| 1453 |
config=NULL, ##<< GLOBAL config variable |
| 1454 |
plot = FALSE ##<< Plot CURs or not |
| 1455 |
) {
|
| 1456 |
get_inter_strain_rois = function(strain1, strain2, diff_allowed = 10, min_cur_width = 200, plot=FALSE) {
|
| 1457 |
c2c_file = list(filename=config$C2C_FILES[[paste(strain1, "-", strain2, sep="")]]) |
| 1458 |
c2c = get_content(c2c_file$filename, "table", stringsAsFactors=FALSE) |
| 1459 |
# Filtering unagapped |
| 1460 |
c2c = c2c[c2c$V6=="-",] |
| 1461 |
# filtering some things (chr...) |
| 1462 |
# c2c = c2c[c2c$V1 == "chrIV",] |
| 1463 |
diff = c2c$V2[-1] - c2c$V3[-length(c2c$V2)] |
| 1464 |
diff2 = c2c$V9[-1] - c2c$V10[-length(c2c$V2)] |
| 1465 |
# Plot diffs to define a threshold (diff_allowed) |
| 1466 |
# hist(abs(c(diff2, diff)),breaks=c(0:2000, 200000000000), xlim=c(0,100)) |
| 1467 |
# Filtering |
| 1468 |
indexes_stop = which(abs(diff) > diff_allowed | abs(diff2) > diff_allowed) |
| 1469 |
indexes_start = c(1, indexes_stop[-length(indexes_stop)] + rep(1, length(indexes_stop) -1)) |
| 1470 |
|
| 1471 |
rois = NULL |
| 1472 |
for(i in 1:length(indexes_start)) {
|
| 1473 |
start = indexes_start[i] |
| 1474 |
stop = indexes_stop[i] |
| 1475 |
sub_c2c = c2c[start:stop,] |
| 1476 |
if (strain1 == "BY") {
|
| 1477 |
chr = ROM2ARAB()[[substr(sub_c2c[1,]$V1,4,10)]] |
| 1478 |
} else {
|
| 1479 |
chr = config$FASTA_INDEXES[[strain1]][[sub_c2c[1,]$V1]] |
| 1480 |
} |
| 1481 |
roi = list(chr=chr, begin=sub_c2c[1,]$V2, end=sub_c2c[length(sub_c2c$V1),]$V3, strain_ref=strain1) |
| 1482 |
roi[["length"]] = roi$end - roi$begin |
| 1483 |
if (roi$length >= min_cur_width) {
|
| 1484 |
rois = dfadd(rois,roi) |
| 1485 |
} |
| 1486 |
if (length(unique(sub_c2c[,c(1,7,8)])[,2]) != 1) {
|
| 1487 |
print("*************** ERROR, non homogenous region! ********************")
|
| 1488 |
} |
| 1489 |
# print(i) |
| 1490 |
# print(roi) |
| 1491 |
# print(sub_c2c) |
| 1492 |
# print("________________________________________________________________")
|
| 1493 |
} |
| 1494 |
if (plot) {
|
| 1495 |
print(paste(length(indexes_stop), "area of interest.")) |
| 1496 |
# Plot rois |
| 1497 |
fasta_ref = list(filename=config$FASTA_REFERENCE_GENOME_FILES[[strain1]]) |
| 1498 |
genome = get_content(fasta_ref$filename, "fasta") |
| 1499 |
plot(0,0, ylim=(c(1,length(genome))), xlim = c(0, max(apply(t(genome), 2, function(chr){length(unlist(chr))}))))
|
| 1500 |
for (name in names(genome)) {
|
| 1501 |
if (strain1 == "BY") {
|
| 1502 |
chr_ref = paste("chr", ARAB2ROM()[[config$FASTA_INDEXES[[strain1]][[name]]]], sep="")
|
| 1503 |
} else {
|
| 1504 |
chr_ref = name |
| 1505 |
} |
| 1506 |
y_lev = as.integer(config$FASTA_INDEXES[[strain1]][[name]]) |
| 1507 |
lines(c(0,length(unlist(genome[[name]]))), c(y_lev,y_lev)) |
| 1508 |
text( length(unlist(genome[[name]]))/2, y_lev, labels = chr_ref) |
| 1509 |
} |
| 1510 |
col=1 |
| 1511 |
for (roi_index in 1:length(rois$chr)) {
|
| 1512 |
roi = rois[roi_index,] |
| 1513 |
y_lev = as.integer(roi$chr) + 0.3 |
| 1514 |
lines(c(roi$begin,roi$end), c(y_lev,y_lev), col=col) |
| 1515 |
text( mean(c(roi$begin,roi$end)), y_lev, labels = roi_index) |
| 1516 |
col = col + 1 |
| 1517 |
} |
| 1518 |
} |
| 1519 |
return(rois) |
| 1520 |
} |
| 1521 |
rois = NULL |
| 1522 |
rois_BY_RM = get_inter_strain_rois("BY", "RM", min_cur_width = min_cur_width, diff_allowed = diff_allowed)
|
| 1523 |
rois_BY_YJM = get_inter_strain_rois("BY", "YJM", min_cur_width = min_cur_width, diff_allowed = diff_allowed)
|
| 1524 |
for (roi_1_index in 1:length(rois_BY_RM[,1])) {
|
| 1525 |
roi_1 = rois_BY_RM[roi_1_index,] |
| 1526 |
roi_2_candidates = rois_BY_YJM[rois_BY_YJM$chr== roi_1$chr & rois_BY_YJM$begin <= roi_1$end & rois_BY_YJM$end >= roi_1$begin , ] ; |
| 1527 |
# print(length(roi_2_candidates[,1])) |
| 1528 |
if (length(roi_2_candidates[,1]) > 0) {
|
| 1529 |
for(roi_2_index in 1:length(roi_2_candidates[,1])) {
|
| 1530 |
roi_2 = roi_2_candidates[roi_2_index,] |
| 1531 |
roi = list(chr=roi_1$chr, begin=max(roi_1$begin, roi_2$begin), end=min(roi_1$end, roi_2$end), strain_ref="BY") |
| 1532 |
roi[["length"]] = roi$end - roi$begin + 1 |
| 1533 |
if (roi$length >= min_cur_width) {
|
| 1534 |
# if (length(rois[,1]) == 153) {
|
| 1535 |
# print(paste(length(rois[,1]), roi_1_index, roi_2_index )) |
| 1536 |
# print(roi_1) |
| 1537 |
# print(roi_2) |
| 1538 |
# print(roi) |
| 1539 |
# } |
| 1540 |
rois = dfadd(rois,roi) |
| 1541 |
} |
| 1542 |
} |
| 1543 |
} |
| 1544 |
} |
| 1545 |
print(length(rois[,1])) |
| 1546 |
print(sum(rois$length)) |
| 1547 |
rois_1st_round = rois |
| 1548 |
rois_2nd_round = NULL |
| 1549 |
rois_RM_YJM = get_inter_strain_rois("RM", "YJM", min_cur_width = min_cur_width, diff_allowed = diff_allowed)
|
| 1550 |
for (roi_1_index in 1:length(rois_1st_round[,1])) {
|
| 1551 |
roi_1 = rois_1st_round[roi_1_index,] |
| 1552 |
translated_roi_1 = translate_roi(roi_1, "RM", config) |
| 1553 |
t_b = min(translated_roi_1$begin, translated_roi_1$end) |
| 1554 |
t_e = max(translated_roi_1$begin, translated_roi_1$end) |
| 1555 |
roi_2_candidates = rois_RM_YJM[rois_RM_YJM$chr== translated_roi_1$chr & rois_RM_YJM$begin <= t_e & rois_RM_YJM$end >= t_b , ] ; |
| 1556 |
if (length(roi_2_candidates[,1]) > 0) {
|
| 1557 |
for(roi_2_index in 1:length(roi_2_candidates[,1])) {
|
| 1558 |
roi_2 = roi_2_candidates[roi_2_index,] |
| 1559 |
roi = list(chr=translated_roi_1$chr, begin=max(t_b, roi_2$begin), end=min(t_e, roi_2$end), strain_ref="RM") |
| 1560 |
roi[["length"]] = roi$end - roi$begin + 1 |
| 1561 |
if (roi$length >= min_cur_width) {
|
| 1562 |
rois_2nd_round = dfadd(rois_2nd_round,roi) |
| 1563 |
} |
| 1564 |
} |
| 1565 |
} |
| 1566 |
} |
| 1567 |
print(length(rois_2nd_round[,1])) |
| 1568 |
print(sum(rois_2nd_round$length)) |
| 1569 |
rois = rois_2nd_round |
| 1570 |
|
| 1571 |
rois_translator_round = list() |
| 1572 |
for (roi_index in 1:length(rois[,1])) {
|
| 1573 |
roi = rois[roi_index,] |
| 1574 |
BY_roi = translate_roi(roi, "BY", config) |
| 1575 |
tmp_BY_roi = BY_roi |
| 1576 |
BY_roi$begin = min(tmp_BY_roi$begin, tmp_BY_roi$end) |
| 1577 |
BY_roi$end = max(tmp_BY_roi$begin, tmp_BY_roi$end) |
| 1578 |
BY_roi$length = abs(BY_roi$length) |
| 1579 |
rois_translator_round = dfadd(rois_translator_round, BY_roi) |
| 1580 |
} |
| 1581 |
rois = rois_translator_round |
| 1582 |
|
| 1583 |
rois_3rd_round = NULL |
| 1584 |
for (roi_index in 1:length(rois[,1])) {
|
| 1585 |
# for (roi_index in 1:2) {
|
| 1586 |
current_roi = rois[roi_index,] |
| 1587 |
# print(roi_index) |
| 1588 |
|
| 1589 |
to_be_check_rois = dfadd(NULL, current_roi) |
| 1590 |
NEED_RERUN = TRUE |
| 1591 |
while (NEED_RERUN) {
|
| 1592 |
# print("RERUN"),
|
| 1593 |
NEED_RERUN = FALSE |
| 1594 |
to_be_check_again = NULL |
| 1595 |
for (to_be_check_roi_index in 1:length(to_be_check_rois[,1])) {
|
| 1596 |
# print(to_be_check_rois) |
| 1597 |
to_be_check_roi = to_be_check_rois[to_be_check_roi_index,] |
| 1598 |
combis = list(c("BY", "RM"), c("BY", "YJM"), c("RM", "YJM"), c("RM", "BY"), c("YJM", "BY"), c("YJM", "RM"))
|
| 1599 |
for (combi in combis) {
|
| 1600 |
# print(combi) |
| 1601 |
strain1 = combi[1] |
| 1602 |
strain2 = combi[2] |
| 1603 |
trans_roi = translate_roi(to_be_check_roi, strain1, config) |
| 1604 |
lower_bound=min(trans_roi$begin, trans_roi$end) |
| 1605 |
upper_bound=max(trans_roi$begin, trans_roi$end) |
| 1606 |
|
| 1607 |
check_overlaping = structure(function(strain1 = "BY", strain2 = "RM", chr = NULL, lower_bound=NULL, upper_bound=NULL) {
|
| 1608 |
reverse = (strain1=="RM" & strain2=="BY") | strain1=="YJM" |
| 1609 |
if (strain1 == strain2) {
|
| 1610 |
roi$length = roi$end - roi$begin + sign(roi$end - roi$begin) * 1 |
| 1611 |
return(roi) |
| 1612 |
} |
| 1613 |
# Launch c2c file |
| 1614 |
if (reverse) {
|
| 1615 |
c2c_file = list(filename=config$C2C_FILES[[paste(strain2, "-", strain1, sep="")]]) |
| 1616 |
} else {
|
| 1617 |
c2c_file = list(filename=config$C2C_FILES[[paste(strain1, "-", strain2, sep="")]]) |
| 1618 |
} |
| 1619 |
c2c = get_content(c2c_file$filename, "table", stringsAsFactors=FALSE) |
| 1620 |
# filtering it |
| 1621 |
c2c = c2c[c2c$V6=="-",] |
| 1622 |
# Reverse |
| 1623 |
if (reverse) {
|
| 1624 |
tmp_col = c2c$V1 |
| 1625 |
c2c$V1 = c2c$V7 |
| 1626 |
c2c$V7 = tmp_col |
| 1627 |
tmp_col = c2c$V2 |
| 1628 |
c2c$V2 = c2c$V9 |
| 1629 |
c2c$V9 = tmp_col |
| 1630 |
tmp_col = c2c$V3 |
| 1631 |
c2c$V3 = c2c$V10 |
| 1632 |
c2c$V10 = tmp_col |
| 1633 |
} |
| 1634 |
|
| 1635 |
if (strain1 == "BY") {
|
| 1636 |
chro_1 = paste("chr", ARAB2ROM()[[chr]], sep="")
|
| 1637 |
} else if (strain1 == "RM") {
|
| 1638 |
chro_1 = paste("supercontig_1.",chr,sep="")
|
| 1639 |
} else if (strain1 == "YJM") {
|
| 1640 |
chro_1 = switch_pairlist(config$FASTA_INDEXES$YJM)[[chr]] |
| 1641 |
} |
| 1642 |
# print(chro_1) |
| 1643 |
if (!is.null(lower_bound) & !is.null(upper_bound)) {
|
| 1644 |
if (reverse) {
|
| 1645 |
tmp_c2c = c2c[c2c$V1==chro_1 & ((c2c$V3>=lower_bound & c2c$V2<=upper_bound & c2c$V8==1) | (c2c$V2>=lower_bound & c2c$V3<=upper_bound & c2c$V8==-1)),] |
| 1646 |
} else {
|
| 1647 |
tmp_c2c = c2c[c2c$V1==chro_1 & c2c$V3>=lower_bound & c2c$V2<=upper_bound,] |
| 1648 |
} |
| 1649 |
} else {
|
| 1650 |
tmp_c2c = c2c[c2c$V1 == chr,] |
| 1651 |
} |
| 1652 |
if (length(tmp_c2c[,1]) > 1) {
|
| 1653 |
pbs = apply(t(1:(length(tmp_c2c[,1]) - 1)), 2, function(i){
|
| 1654 |
# print(paste(i, "/", length(tmp_c2c[,1]))) |
| 1655 |
apply(t((i+1):length(tmp_c2c[,1])), 2, function(j){
|
| 1656 |
l1 = tmp_c2c[i,] |
| 1657 |
b1 = min(l1$V2, l1$V3) |
| 1658 |
e1 = max(l1$V2, l1$V3) |
| 1659 |
l2 = tmp_c2c[j,] |
| 1660 |
b2 = min(l2$V2, l2$V3) |
| 1661 |
e2 = max(l2$V2, l2$V3) |
| 1662 |
if ((e1>=b2 & b1<=e2) | (e2>=b1 & b2<=e1)) {
|
| 1663 |
print(paste("WARNING! Overlaping", " (", strain1, ",", strain2, ") chr: ",chr, " [", b1, ",", e1, "] [", b2, ",", e2, "]", sep=""))
|
| 1664 |
pb = list(strain1, strain2, chr, b1, e1, b2, e2) |
| 1665 |
pb |
| 1666 |
} else {
|
| 1667 |
NULL |
| 1668 |
} |
| 1669 |
}) |
| 1670 |
}) |
| 1671 |
return(pbs) |
| 1672 |
} |
| 1673 |
|
| 1674 |
}, ex=function(){
|
| 1675 |
source("src/nucleo_miner/yeast_strain_conversion.R");
|
| 1676 |
pbs1 = check_overlaping(strain1 = "BY", strain2 = "RM", dest=TRUE) |
| 1677 |
pbs3 = check_overlaping(strain1 = "BY", strain2 = "YJM", dest=TRUE) |
| 1678 |
pbs5 = check_overlaping(strain1 = "RM", strain2 = "YJM", dest=TRUE) |
| 1679 |
pbs2 = check_overlaping(strain1 = "BY", strain2 = "RM", dest=FALSE) |
| 1680 |
pbs4 = check_overlaping(strain1 = "BY", strain2 = "YJM", dest=FALSE) |
| 1681 |
pbs6 = check_overlaping(strain1 = "RM", strain2 = "YJM", dest=FALSE) |
| 1682 |
}) |
| 1683 |
|
| 1684 |
|
| 1685 |
res = check_overlaping(strain1 = strain1, strain2 = strain2, chr = trans_roi$chr, lower_bound=lower_bound, upper_bound=upper_bound) |
| 1686 |
if (!is.null(res)) {
|
| 1687 |
df_res = data.frame(matrix(unlist(res), ncol = 7, byrow=TRUE), stringsAsFactors=FALSE) |
| 1688 |
interval = df_res[1,] |
| 1689 |
inter_min = as.numeric(max( min(interval$X4, interval$X5), min(interval$X6, interval$X7))) |
| 1690 |
inter_max = as.numeric(min( max(interval$X4, interval$X5), max(interval$X6, interval$X7))) |
| 1691 |
# print(paste("SPLIT ROI", roi_index, "for", combi[1], combi[2]))
|
| 1692 |
new_roi1 = trans_roi |
| 1693 |
new_roi2 = trans_roi |
| 1694 |
new_roi1$begin = lower_bound |
| 1695 |
new_roi1$end = inter_min - 1 |
| 1696 |
new_roi1$length = new_roi1$end - new_roi1$begin + 1 |
| 1697 |
new_roi2$begin = inter_max + 1 |
| 1698 |
new_roi2$end = upper_bound |
| 1699 |
new_roi2$length = new_roi2$end - new_roi2$begin + 1 |
| 1700 |
if (new_roi1$length > min_cur_width) {
|
| 1701 |
BY_roi = translate_roi(new_roi1, "BY", config) |
| 1702 |
tmp_BY_roi = BY_roi |
| 1703 |
BY_roi$begin = min(tmp_BY_roi$begin, tmp_BY_roi$end) |
| 1704 |
BY_roi$end = max(tmp_BY_roi$begin, tmp_BY_roi$end) |
| 1705 |
BY_roi$length = abs(BY_roi$length) |
| 1706 |
to_be_check_again = dfadd(to_be_check_again, BY_roi) |
| 1707 |
} |
| 1708 |
if (new_roi2$length > min_cur_width) {
|
| 1709 |
BY_roi = translate_roi(new_roi2, "BY", config) |
| 1710 |
tmp_BY_roi = BY_roi |
| 1711 |
BY_roi$begin = min(tmp_BY_roi$begin, tmp_BY_roi$end) |
| 1712 |
BY_roi$end = max(tmp_BY_roi$begin, tmp_BY_roi$end) |
| 1713 |
BY_roi$length = abs(BY_roi$length) |
| 1714 |
to_be_check_again = dfadd(to_be_check_again, BY_roi) |
| 1715 |
} |
| 1716 |
if (to_be_check_roi_index < length(to_be_check_rois[,1])) {
|
| 1717 |
for (i in (to_be_check_roi_index + 1):length(to_be_check_rois[,1])) {
|
| 1718 |
to_be_check_again = dfadd(to_be_check_again, to_be_check_rois[i,]) |
| 1719 |
} |
| 1720 |
} |
| 1721 |
NEED_RERUN = TRUE |
| 1722 |
break |
| 1723 |
} |
| 1724 |
} |
| 1725 |
if (NEED_RERUN) {
|
| 1726 |
to_be_check_rois = to_be_check_again |
| 1727 |
break |
| 1728 |
} |
| 1729 |
} |
| 1730 |
} |
| 1731 |
|
| 1732 |
checked_rois = to_be_check_rois |
| 1733 |
for (checked_roi_index in 1:length(checked_rois[,1])) {
|
| 1734 |
rois_3rd_round = dfadd(rois_3rd_round, checked_rois[checked_roi_index,]) |
| 1735 |
} |
| 1736 |
} |
| 1737 |
|
| 1738 |
print(length(rois_3rd_round[,1])) |
| 1739 |
print(sum(rois_3rd_round$length)) |
| 1740 |
rois = rois_3rd_round |
| 1741 |
|
| 1742 |
|
| 1743 |
if (plot) {
|
| 1744 |
print(paste(length(rois$chr), "area of interest.")) |
| 1745 |
# Plot rois |
| 1746 |
fasta_ref = list(filename=config$FASTA_REFERENCE_GENOME_FILES[["BY"]]) |
| 1747 |
genome = get_content(fasta_ref$filename, "fasta") |
| 1748 |
plot(0,0, ylim=(c(1,length(genome))), xlim = c(0, max(apply(t(genome), 2, function(chr){length(unlist(chr))}))))
|
| 1749 |
for (name in names(genome)) {
|
| 1750 |
if (TRUE) {
|
| 1751 |
chr_ref = paste("chr", ARAB2ROM()[[config$FASTA_INDEXES[["BY"]][[name]]]], sep="")
|
| 1752 |
} else {
|
| 1753 |
chr_ref = name |
| 1754 |
} |
| 1755 |
y_lev = as.integer(config$FASTA_INDEXES[["BY"]][[name]]) |
| 1756 |
lines(c(0,length(unlist(genome[[name]]))), c(y_lev,y_lev)) |
| 1757 |
text( length(unlist(genome[[name]]))/2, y_lev, labels = chr_ref) |
| 1758 |
} |
| 1759 |
col=1 |
| 1760 |
for (roi_index in 1:length(rois$chr)) {
|
| 1761 |
roi = rois[roi_index,] |
| 1762 |
y_lev = as.integer(roi$chr) + 0.3 |
| 1763 |
lines(c(roi$begin,roi$end), c(y_lev,y_lev), col=col) |
| 1764 |
text( mean(c(roi$begin,roi$end)), y_lev, labels = roi_index) |
| 1765 |
col = col + 1 |
| 1766 |
} |
| 1767 |
} |
| 1768 |
return (rois) |
| 1769 |
} |
| 1770 |
|
| 1771 |
|
| 1772 |
|
| 1773 |
|
| 1774 |
|
| 1775 |
|
| 1776 |
|
| 1777 |
|
| 1778 |
|
| 1779 |
|
| 1780 |
|
| 1781 |
|
| 1782 |
|
| 1783 |
|
| 1784 |
|
| 1785 |
|
| 1786 |
|
| 1787 |
|
| 1788 |
|
| 1789 |
|
| 1790 |
|
| 1791 |
|
| 1792 |
|
| 1793 |
|
| 1794 |
|
| 1795 |
|
| 1796 |
|
| 1797 |
|
| 1798 |
|
| 1799 |
|
| 1800 |
|
| 1801 |
|
| 1802 |
|
| 1803 |
|
| 1804 |
|
| 1805 |
|
| 1806 |
|
| 1807 |
|
| 1808 |
|
| 1809 |
|
| 1810 |
|
| 1811 |
|
| 1812 |
|
| 1813 |
|
| 1814 |
|
| 1815 |
|
| 1816 |
|
| 1817 |
|
| 1818 |
|
| 1819 |
|
| 1820 |
|
| 1821 |
|
| 1822 |
|
| 1823 |
|
| 1824 |
|
| 1825 |
|
| 1826 |
|
| 1827 |
|
| 1828 |
|
| 1829 |
|
| 1830 |
|
| 1831 |
|
| 1832 |
|
| 1833 |
|
| 1834 |
|
| 1835 |
|
| 1836 |
|
| 1837 |
|
| 1838 |
|
| 1839 |
|
| 1840 |
|
| 1841 |
|
| 1842 |
perform_anovas = function(# Performaing ANOVAs |
| 1843 |
### Counts reads and Performs ANOVAS for each common nucleosomes involved. |
| 1844 |
replicates, ##<< Set of replicates, each replicate is a list of samples (ideally 3). Each sample is a list like \emph{sample = list(id=..., marker=..., strain=..., roi=..., inputs=..., outputs=...)} with \emph{roi = list(name=..., begin=..., end=..., chr=..., genome=...)}. In the \emph{perform_anovas} contexte, we need 4 replicates (4 * (3 samples)): 2 strains * (1 marker + 1 input (Mnase_Seq)).
|
| 1845 |
aligned_inter_strain_nucs, ##<< List of common nucleosomes. |
| 1846 |
inputs_name="Mnase_Seq", ##<< Name of the input. |
| 1847 |
plot_anova_boxes=FALSE ##<< Plot (or not) boxplot for each nuc. |
| 1848 |
) {
|
| 1849 |
anova_results = NULL |
| 1850 |
for (inter_strain_nuc_index in 1:length(aligned_inter_strain_nucs[,1])) {
|
| 1851 |
inter_strain_nuc = aligned_inter_strain_nucs[inter_strain_nuc_index,] |
| 1852 |
# counting reads |
| 1853 |
my_data = NULL |
| 1854 |
for (replicate_rank in 1:length(replicates)) {
|
| 1855 |
samples = replicates[[replicate_rank]] |
| 1856 |
strain = samples[[1]]$strain |
| 1857 |
marker = samples[[1]]$marker |
| 1858 |
for (sample_rank in 1:length(samples)) {
|
| 1859 |
sample = samples[[sample_rank]] |
| 1860 |
nuc_width = as.integer(inter_strain_nuc[[paste("upper_bound_",strain,sep="")]]) - inter_strain_nuc[[paste("lower_bound_",strain,sep="")]]
|
| 1861 |
nuc_reads = filter_tf_inputs(sample$inputs, inter_strain_nuc[[paste("chr_",strain,sep="")]], inter_strain_nuc[[paste("lower_bound_",strain,sep="")]], inter_strain_nuc[[paste("upper_bound_",strain,sep="")]], nuc_width - 1)
|
| 1862 |
indic = sum(nuc_reads[,4]) * 1000000/sample$total_reads / nuc_width |
| 1863 |
my_data = dfadd(my_data, list(strain = strain, marker = marker, indic = indic)) |
| 1864 |
} |
| 1865 |
} |
| 1866 |
my_data$strain = as.factor(my_data$strain) |
| 1867 |
my_data$marker = as.factor(my_data$marker) |
| 1868 |
|
| 1869 |
strain_ref1 = replicates[[1]][[1]]$strain |
| 1870 |
strain_ref2 = replicates[[2]][[1]]$strain |
| 1871 |
marker = replicates[[length(replicates)]][[1]]$marker |
| 1872 |
|
| 1873 |
# Collecting anova results |
| 1874 |
anova_result = list() |
| 1875 |
# nucs info |
| 1876 |
for (name in names(inter_strain_nuc)) {
|
| 1877 |
anova_result[[name]] = inter_strain_nuc[[name]] |
| 1878 |
} |
| 1879 |
# |
| 1880 |
for (strain in c(strain_ref1, strain_ref2)) {
|
| 1881 |
for (manip in c(inputs_name, marker)) {
|
| 1882 |
replicat = 1 |
| 1883 |
for(indic in my_data[my_data$strain == strain & my_data$marker == manip,]$indic) {
|
| 1884 |
anova_result[[paste(strain, manip, replicat, sep="_")]] = indic |
| 1885 |
replicat = replicat + 1 |
| 1886 |
} |
| 1887 |
} |
| 1888 |
} |
| 1889 |
|
| 1890 |
# Compute ANOVAs |
| 1891 |
mnase_data = my_data[my_data$marker == inputs_name,] |
| 1892 |
mnase_aov = aov(indic~strain,mnase_data) |
| 1893 |
mnase_effects = model.tables(mnase_aov)$tables |
| 1894 |
mnase_pvalues = summary(mnase_aov)[[1]]$Pr |
| 1895 |
# boxplot(indic~ma*st,mnase_data) |
| 1896 |
anova_result[["mnase_st"]] = mnase_effects$strain[1] |
| 1897 |
anova_result[["mnase_st_pvalue"]] = mnase_pvalues[1] |
| 1898 |
# |
| 1899 |
marker_data = my_data[my_data$marker == marker,] |
| 1900 |
marker_aov = aov(indic~strain,marker_data) |
| 1901 |
marker_effects = model.tables(marker_aov)$tables |
| 1902 |
marker_pvalues = summary(marker_aov)[[1]]$Pr |
| 1903 |
# boxplot(indic~ma*st,marker_data) |
| 1904 |
anova_result[["marker_st"]] = marker_effects$strain[1] |
| 1905 |
anova_result[["marker_st_pvalue"]] = marker_pvalues[1] |
| 1906 |
# |
| 1907 |
st1_data = my_data[my_data$strain == strain_ref1,] |
| 1908 |
st1_aov = aov(indic~marker,st1_data) |
| 1909 |
st1_effects = model.tables(st1_aov)$tables |
| 1910 |
st1_pvalues = summary(st1_aov)[[1]]$Pr |
| 1911 |
# boxplot(indic~ma*st,st1_data) |
| 1912 |
anova_result[["st1_ma"]]=st1_effects$ma[inputs_name] |
| 1913 |
anova_result[["st1_ma_pvalue"]]=st1_pvalues[1] |
| 1914 |
# |
| 1915 |
st2_data = my_data[my_data$strain == strain_ref2,] |
| 1916 |
st2_aov = aov(indic~marker,st2_data) |
| 1917 |
st2_effects = model.tables(st2_aov)$tables |
| 1918 |
st2_pvalues = summary(st2_aov)[[1]]$Pr |
| 1919 |
# boxplot(indic~ma*st,st2_data) |
| 1920 |
anova_result[["st2_ma"]]=st2_effects$ma[inputs_name] |
| 1921 |
anova_result[["st2_ma_pvalue"]]=st2_pvalues[1] |
| 1922 |
# |
| 1923 |
correl_data = my_data |
| 1924 |
correl_aov = aov(indic~strain*marker,correl_data) |
| 1925 |
correl_effects = model.tables(correl_aov)$tables |
| 1926 |
correl_pvalues = summary(correl_aov)[[1]]$Pr |
| 1927 |
if (plot_anova_boxes) {
|
| 1928 |
x11() |
| 1929 |
boxplot(indic~marker*strain,correl_data) |
| 1930 |
} |
| 1931 |
anova_result[["correl_st"]]=correl_effects$strain[1] |
| 1932 |
anova_result[["correl_ma"]]=correl_effects$ma[inputs_name] |
| 1933 |
anova_result[["correl_st_ma"]]=correl_effects$"strain:marker"[1,1] |
| 1934 |
anova_result[["correl_st_pvalue"]]=correl_pvalues[1] |
| 1935 |
anova_result[["correl_ma_pvalue"]]=correl_pvalues[2] |
| 1936 |
anova_result[["correl_st_ma_pvalue"]]=correl_pvalues[3] |
| 1937 |
|
| 1938 |
anova_results = dfadd(anova_results, anova_result) |
| 1939 |
} |
| 1940 |
return(anova_results) |
| 1941 |
### Returns ANOVA results and comunted reads. |
| 1942 |
} |
| 1943 |
|
| 1944 |
|
| 1945 |
watch_samples = function(# Watching analysis of samples |
| 1946 |
### This function allows to view analysis for a particuler region of the genome. |
| 1947 |
replicates, ##<< replicates under the form... |
| 1948 |
read_length, ##<< length of the reads |
| 1949 |
plot_ref_genome = TRUE, ##<< Plot (or not) reference genome. |
| 1950 |
plot_arrow_raw_reads = TRUE, ##<< Plot (or not) arrows for raw reads. |
| 1951 |
plot_arrow_nuc_reads = TRUE, ##<< Plot (or not) arrows for reads aasiocied to a nucleosome. |
| 1952 |
plot_squared_reads = TRUE, ##<< Plot (or not) reads in the square fashion. |
| 1953 |
plot_coverage = FALSE, ##<< Plot (or not) reads in the covergae fashion. fashion. |
| 1954 |
plot_gaussian_reads = TRUE, ##<< Plot (or not) gaussian model of a F anf R reads. |
| 1955 |
plot_gaussian_unified_reads = TRUE, ##<< Plot (or not) gaussian model of a nuc. |
| 1956 |
plot_ellipse_nucs = TRUE, ##<< Plot (or not) ellipse for a nuc. |
| 1957 |
plot_wp_nucs = TRUE, ##<< Plot (or not) cluster of nucs |
| 1958 |
plot_wp_nuc_model = TRUE, ##<< Plot (or not) gaussian model for a cluster of nucs |
| 1959 |
plot_common_nucs = TRUE, ##<< Plot (or not) aligned reads. |
| 1960 |
plot_anovas = FALSE, ##<< Plot (or not) scatter for each nuc. |
| 1961 |
plot_anova_boxes = FALSE, ##<< Plot (or not) boxplot for each nuc. |
| 1962 |
plot_wp_nucs_4_nonmnase = FALSE, ##<< Plot (or not) clusters for non inputs samples. |
| 1963 |
aggregated_intra_strain_nucs = NULL, ##<< list of aggregated intra strain nucs. If NULL, it will be computed. |
| 1964 |
aligned_inter_strain_nucs = NULL, ##<< list of aligned inter strain nucs. If NULL, it will be computed. |
| 1965 |
height = 10, ##<< Number of reads in per million read for each sample, graphical parametre for the y axis. |
| 1966 |
config=NULL ##<< GLOBAL config variable |
| 1967 |
){
|
| 1968 |
returned_list = list() |
| 1969 |
# Computing global display parameters |
| 1970 |
if (replicates[[1]][[1]]$roi[["begin"]] < replicates[[1]][[1]]$roi[["end"]]) {
|
| 1971 |
x_min_glo = replicates[[1]][[1]]$roi[["begin"]] |
| 1972 |
x_max_glo = replicates[[1]][[1]]$roi[["end"]] |
| 1973 |
} else {
|
| 1974 |
x_min_glo = - replicates[[1]][[1]]$roi[["begin"]] |
| 1975 |
x_max_glo = - replicates[[1]][[1]]$roi[["end"]] |
| 1976 |
} |
| 1977 |
base_glo = 0 |
| 1978 |
nb_rank_glo = 0 |
| 1979 |
for (samples in replicates) {
|
| 1980 |
nb_rank_glo = nb_rank_glo + length(samples) |
| 1981 |
} |
| 1982 |
ylim_glo = c(base_glo, base_glo + height * nb_rank_glo) |
| 1983 |
y_min_glo = min(ylim_glo) |
| 1984 |
y_max_glo = max(ylim_glo) |
| 1985 |
delta_y_glo = y_max_glo - y_min_glo |
| 1986 |
# Plot main frame |
| 1987 |
plot(c(x_min_glo,x_max_glo), c(0,0), ylim=ylim_glo, col=0, yaxt="n", ylab="#reads (per million reads)", xlab=paste("Ref strain:", replicates[[1]][[1]]$strain, "chr: ", replicates[[1]][[1]]$roi$chr), main="NucleoMiner2" )
|
| 1988 |
axis(2, at=0:(nb_rank_glo*2) * delta_y_glo / (nb_rank_glo*2), labels=c(rep(c(height/2,0),nb_rank_glo),height/2)) |
| 1989 |
# Go |
| 1990 |
replicates_wp_nucs = list() |
| 1991 |
for (replicate_rank in 1:length(replicates)) {
|
| 1992 |
# Computing replicate parameters |
| 1993 |
nb_rank = length(samples) |
| 1994 |
base = (replicate_rank-1) * height * nb_rank |
| 1995 |
ylim = c(base, base + height * nb_rank) |
| 1996 |
y_min = min(ylim) |
| 1997 |
y_max = max(ylim) |
| 1998 |
delta_y = y_max - y_min |
| 1999 |
samples = replicates[[replicate_rank]] |
| 2000 |
for (sample_rank in 1:length(samples)) {
|
| 2001 |
# computing sample parameters |
| 2002 |
sample = samples[[sample_rank]] |
| 2003 |
y_lev = y_min + (sample_rank - 0.5) * delta_y/nb_rank |
| 2004 |
text(x_min_glo, y_lev + height/2 - delta_y_glo/100, labels=paste("(",sample$id,") ",sample$strain, " ", sample$marker, sep=""))
|
| 2005 |
|
| 2006 |
if (sample$roi[["begin"]] < sample$roi[["end"]]) {
|
| 2007 |
x_min = sample$roi[["begin"]] |
| 2008 |
x_max = sample$roi[["end"]] |
| 2009 |
} else {
|
| 2010 |
x_min = - sample$roi[["begin"]] |
| 2011 |
x_max = - sample$roi[["end"]] |
| 2012 |
} |
| 2013 |
shift = x_min_glo - x_min |
| 2014 |
# Plot Genome seq |
| 2015 |
if (plot_ref_genome) {
|
| 2016 |
text(1:length(sample$roi$genome) + x_min - 1 + shift, rep(y_lev - height/2, length(sample$roi$genome)), labels=toupper(sample$roi$genome), cex=dev.size()[1]*9/(x_max-x_min), family="Courier") |
| 2017 |
} |
| 2018 |
# Plot reads |
| 2019 |
reads = sample$inputs |
| 2020 |
signs = sign_from_strand(reads[,3]) |
| 2021 |
if (plot_arrow_raw_reads) {
|
| 2022 |
arrows(sign(x_min) * reads[,2] + shift, sign(x_min) * signs * reads[,4] * 1000000/sample$total_reads + y_lev, sign(x_min) * (reads[,2] + signs * read_length) + shift, sign(x_min) * signs * reads[,4] * 1000000/sample$total_reads + y_lev, |
| 2023 |
col=1, length=0.15/nb_rank) |
| 2024 |
} |
| 2025 |
if (plot_squared_reads) {
|
| 2026 |
# require(plotrix) |
| 2027 |
rect(sign(x_min) * reads[,2] + shift, rep(y_lev,length(reads[,1])), sign(x_min) * (reads[,2] + signs * read_length) + shift, y_lev + sign(x_min) * signs * reads[,4] * 1000000/sample$total_reads, col=adjustcolor(1, alpha.f = 0.1),border=0) |
| 2028 |
} |
| 2029 |
if (plot_coverage) {
|
| 2030 |
if (length(reads[,1]) != 0) {
|
| 2031 |
reads[1,1] = 0 |
| 2032 |
reads[1,2] = 0 |
| 2033 |
reads[1,3] = 0 |
| 2034 |
reads[1,4] = 0 |
| 2035 |
signs = c(1) |
| 2036 |
step_h = sign(x_min) * signs * reads[,4] |
| 2037 |
step_b = sign(x_min) * reads[,2] + shift |
| 2038 |
step_e = sign(x_min) * (reads[,2] + signs * 150) + shift |
| 2039 |
steps_x = min(step_b, step_e):max(step_b, step_e) |
| 2040 |
steps_y = rep(0, length(steps_x)) |
| 2041 |
for (i in 1:length(step_h)) {
|
| 2042 |
steps_y[which(steps_x==min(step_b[i], step_e[i]))] = steps_y[which(steps_x==min(step_b[i], step_e[i]))] + abs(step_h[i]) |
| 2043 |
steps_y[which(steps_x==max(step_b[i], step_e[i]))] = steps_y[which(steps_x==max(step_b[i], step_e[i]))] - abs(step_h[i]) |
| 2044 |
} |
| 2045 |
tmp_index = which(steps_y != 0) |
| 2046 |
steps_x = steps_x[tmp_index] |
| 2047 |
steps_y = steps_y[tmp_index] |
| 2048 |
tmp_current_level = 0 |
| 2049 |
for (i in 1:length(steps_y)) {
|
| 2050 |
steps_y[i] = tmp_current_level + steps_y[i] |
| 2051 |
tmp_current_level = steps_y[i] |
| 2052 |
} |
| 2053 |
steps_y = c(0, steps_y) |
| 2054 |
steps_y = steps_y * 1000000/sample$total_reads |
| 2055 |
} else {
|
| 2056 |
steps_y = c(0, 0, 0) |
| 2057 |
steps_x = c(sample$roi$begin, sample$roi$end) |
| 2058 |
} |
| 2059 |
lines(stepfun(steps_x, steps_y + y_lev), pch="") |
| 2060 |
abline(y_lev,0) |
| 2061 |
returned_list[[paste("cov", sample$id, sep="_")]] = stepfun(steps_x, steps_y)
|
| 2062 |
} |
| 2063 |
# Plot nucs |
| 2064 |
if (sample$marker == "Mnase_Seq" & (plot_squared_reads | plot_gaussian_reads | plot_gaussian_unified_reads | plot_arrow_nuc_reads)) {
|
| 2065 |
nucs = sample$outputs |
| 2066 |
if (length(nucs$center) > 0) {
|
| 2067 |
col = 1 |
| 2068 |
for (i in 1:length(nucs$center)) {
|
| 2069 |
col = col + 1 |
| 2070 |
nuc = nucs[i,] |
| 2071 |
involved_reads = filter_tf_inputs(reads, sample$roi$chr, nuc$lower_bound, nuc$upper_bound, nuc_width = nuc$width) |
| 2072 |
involved_signs = apply(t(involved_reads[,3]), 2, function(strand) { if (strand == "F") return(1) else return(-1)})
|
| 2073 |
total_involved_reads = sum(involved_reads[,4]) |
| 2074 |
if (plot_arrow_nuc_reads ) {
|
| 2075 |
arrows(sign(x_min) * involved_reads[,2] + shift, sign(x_min) * involved_signs * involved_reads[,4] * 1000000/sample$total_reads + y_lev, sign(x_min) * (involved_reads[,2] + involved_signs * read_length) + shift, sign(x_min) * involved_signs * involved_reads[,4] * 1000000/sample$total_reads + y_lev, |
| 2076 |
col=col, length=0.15/nb_rank) |
| 2077 |
} |
| 2078 |
if (plot_gaussian_reads | plot_gaussian_unified_reads) {
|
| 2079 |
flatted_reads = flat_reads(involved_reads, nuc$width) |
| 2080 |
delta_x = (nuc$center - nuc$width):(nuc$center + nuc$width) |
| 2081 |
} |
| 2082 |
if (plot_gaussian_reads ) {
|
| 2083 |
flatted_reads = flat_reads(involved_reads, nuc$width) |
| 2084 |
delta_x = (nuc$center - nuc$width):(nuc$center + nuc$width) |
| 2085 |
lines(sign(x_min) * delta_x + shift, dnorm(delta_x, mean(flatted_reads[[1]]), sd(flatted_reads[[1]])) * length(flatted_reads[[1]]) * sign(x_min) + y_lev, col=col) |
| 2086 |
lines(sign(x_min) * delta_x + shift, dnorm(delta_x, mean(flatted_reads[[2]]), sd(flatted_reads[[2]])) * length(flatted_reads[[2]]) * -1 * sign(x_min) + y_lev, col=col) |
| 2087 |
} |
| 2088 |
if (plot_gaussian_unified_reads ) {
|
| 2089 |
lines(sign(x_min) * delta_x + shift, dnorm(delta_x, mean(flatted_reads[[3]]), sd(flatted_reads[[3]])) * length(flatted_reads[[3]]) + y_lev, col=col, lty=2) |
| 2090 |
} |
| 2091 |
if (plot_ellipse_nucs) {
|
| 2092 |
# require(plotrix) |
| 2093 |
draw.ellipse(sign(x_min) * nuc$center + shift, y_lev, nuc$width/2, total_involved_reads/nuc$width, border=col) |
| 2094 |
} |
| 2095 |
} |
| 2096 |
} else {
|
| 2097 |
print("WARNING! No nucs to print.")
|
| 2098 |
} |
| 2099 |
} |
| 2100 |
} |
| 2101 |
# Plot wp nucs |
| 2102 |
if ((plot_wp_nucs_4_nonmnase | sample$marker == "Mnase_Seq") & plot_wp_nucs) {
|
| 2103 |
if (samples[[1]]$marker == "Mnase_Seq") {
|
| 2104 |
if (is.null(aggregated_intra_strain_nucs)) {
|
| 2105 |
wp_nucs = aggregate_intra_strain_nucs(samples)[[1]] |
| 2106 |
} else {
|
| 2107 |
wp_nucs = aggregated_intra_strain_nucs[[replicate_rank]] |
| 2108 |
} |
| 2109 |
} else {
|
| 2110 |
wp_nucs = replicates_wp_nucs[[replicate_rank-2]] |
| 2111 |
} |
| 2112 |
replicates_wp_nucs[[replicate_rank]] = wp_nucs |
| 2113 |
for (wp_nuc in wp_nucs) {
|
| 2114 |
if (wp_nuc$wp){
|
| 2115 |
rect(sign(x_min) * wp_nuc$lower_bound + shift, y_min, sign(x_min) * wp_nuc$upper_bound + shift, y_max, col=adjustcolor(2, alpha.f = 0.1), border=1) |
| 2116 |
all_original_reads = c() |
| 2117 |
for(initial_nuc in wp_nuc$nucs) {
|
| 2118 |
all_original_reads = c(all_original_reads, initial_nuc$original_reads) |
| 2119 |
} |
| 2120 |
delta_x = wp_nuc$lower_bound:wp_nuc$upper_bound |
| 2121 |
if (FALSE) {
|
| 2122 |
rect(sign(x_min) * wp_nuc$lower_bound + shift, y_min, sign(x_min) * wp_nuc$upper_bound + shift, y_max, col="#EEEEEE", border=1) |
| 2123 |
} |
| 2124 |
if (plot_wp_nuc_model) {
|
| 2125 |
lines(sign(x_min) * delta_x + shift, dnorm(delta_x, mean(all_original_reads), sd(all_original_reads)) * length(all_original_reads) + y_min, col=1) |
| 2126 |
} |
| 2127 |
} |
| 2128 |
} |
| 2129 |
} |
| 2130 |
} |
| 2131 |
|
| 2132 |
if (plot_common_nucs | plot_anovas | plot_anova_boxes) {
|
| 2133 |
if (is.null(aligned_inter_strain_nucs)) {
|
| 2134 |
aligned_inter_strain_nucs = align_inter_strain_nucs(replicates, replicates_wp_nucs[[1]], replicates_wp_nucs[[2]], config=config)[[1]] |
| 2135 |
} |
| 2136 |
if (plot_common_nucs) {
|
| 2137 |
#Plot common wp nucs |
| 2138 |
mid_y = shift = x_min = x_max = nb_rank = base = ylim = ymin = y_max = delta_y = list() |
| 2139 |
for (replicate_rank in 1:length(replicates)) {
|
| 2140 |
nb_rank[[replicate_rank]] = length(samples) |
| 2141 |
base[[replicate_rank]] = (replicate_rank-1) * height * nb_rank[[replicate_rank]] |
| 2142 |
ylim[[replicate_rank]] = c(base[[replicate_rank]], base[[replicate_rank]] + height * nb_rank[[replicate_rank]]) |
| 2143 |
y_min[[replicate_rank]] = min(ylim[[replicate_rank]]) |
| 2144 |
y_max[[replicate_rank]] = max(ylim[[replicate_rank]]) |
| 2145 |
delta_y[[replicate_rank]] = y_max[[replicate_rank]] - y_min[[replicate_rank]] |
| 2146 |
mid_y[[replicate_rank]] = (y_max[[replicate_rank]] + y_min[[replicate_rank]]) / 2 |
| 2147 |
|
| 2148 |
samples = replicates[[replicate_rank]] |
| 2149 |
for (sample_rank in 1:length(samples)) {
|
| 2150 |
sample = samples[[sample_rank]] |
| 2151 |
y_lev = y_min[[replicate_rank]] + (sample_rank - 0.5) * delta_y[[replicate_rank]]/nb_rank[[replicate_rank]] |
| 2152 |
if (sample$roi[["begin"]] < sample$roi[["end"]]) {
|
| 2153 |
x_min[[replicate_rank]] = sample$roi[["begin"]] |
| 2154 |
x_max[[replicate_rank]] = sample$roi[["end"]] |
| 2155 |
} else {
|
| 2156 |
x_min[[replicate_rank]] = - sample$roi[["begin"]] |
| 2157 |
x_max[[replicate_rank]] = - sample$roi[["end"]] |
| 2158 |
} |
| 2159 |
shift[[replicate_rank]] = x_min[[1]] - x_min[[replicate_rank]] |
| 2160 |
} |
| 2161 |
} |
| 2162 |
for (inter_strain_nuc_index in 1:length(aligned_inter_strain_nucs[,1])) {
|
| 2163 |
inter_strain_nuc = aligned_inter_strain_nucs[inter_strain_nuc_index,] |
| 2164 |
tmp_xs = tmp_ys = c() |
| 2165 |
for (replicate_rank in 1:length(replicates)) {
|
| 2166 |
samples = replicates[[replicate_rank]] |
| 2167 |
strain = samples[[1]]$strain |
| 2168 |
tmp_xs = c(tmp_xs, sign(x_min[[replicate_rank]]) * (inter_strain_nuc[[paste("lower_bound_",strain,sep="")]] + inter_strain_nuc[[paste("upper_bound_",strain,sep="")]])/2 + shift[[replicate_rank]])
|
| 2169 |
tmp_ys = c(tmp_ys, mid_y[[replicate_rank]]) |
| 2170 |
} |
| 2171 |
tmp_ys <<- tmp_ys |
| 2172 |
tmp_xs <<- tmp_xs |
| 2173 |
lines(tmp_xs, tmp_ys, col=2, type="b", lwd=dev.size()[1]*100/(x_max[[1]]-x_min[[1]]), cex=dev.size()[1]*200/(x_max[[1]]-x_min[[1]]), pch=19) |
| 2174 |
} |
| 2175 |
} |
| 2176 |
if (plot_anovas | plot_anova_boxes) {
|
| 2177 |
anova_results = perform_anovas(replicates, aligned_inter_strain_nucs, plot_anova_boxes=plot_anova_boxes) |
| 2178 |
thres = FDR(anova_results[,"correl_st_ma_pvalue"],0.05) |
| 2179 |
if (is.na(thres)) {
|
| 2180 |
# Boneferroni multiple test |
| 2181 |
thres = 0.05/length(anova_results[,1]) |
| 2182 |
} |
| 2183 |
filtred_anova_results = anova_results[anova_results[,"correl_st_ma_pvalue"]<thres,] |
| 2184 |
x11() |
| 2185 |
if (plot_anovas) {
|
| 2186 |
plot(anova_results[,"mnase_st"],anova_results[,"correl_st_ma"], pch=1, xlim=c(-0.07,0.07), ylim=c(-0.07,0.07),main="SNEPs" ,xlab = "strain effect", ylab = "snep effect") |
| 2187 |
points(x=filtred_anova_results[,"mnase_st"],y=filtred_anova_results[,"correl_st_ma"],col=2, pch="+") |
| 2188 |
} |
| 2189 |
} |
| 2190 |
} |
| 2191 |
return(returned_list) |
| 2192 |
} |
| 2193 |
|
| 2194 |
|
| 2195 |
|
| 2196 |
|