Mercurial > repos > devteam > dwt_ivc_all
changeset 1:506ae7b0d85d draft default tip
"planemo upload for repository https://github.com/galaxyproject/tools-devteam/tree/master/tools/dwt_ivc_all commit f929353ffb0623f2218d7dec459c7da62f3b0d24"
| author | devteam |
|---|---|
| date | Mon, 06 Jul 2020 20:31:56 -0400 |
| parents | 0b89b03ad760 |
| children | |
| files | execute_dwt_IvC_all.R execute_dwt_IvC_all.pl execute_dwt_IvC_all.xml test-data/in1.tsv test-data/in2.tsv test-data/out2.pdf |
| diffstat | 6 files changed, 223 insertions(+), 221 deletions(-) [+] |
line wrap: on
line diff
--- /dev/null Thu Jan 01 00:00:00 1970 +0000 +++ b/execute_dwt_IvC_all.R Mon Jul 06 20:31:56 2020 -0400 @@ -0,0 +1,163 @@ +########################################################################################### +## code to do wavelet Indel vs. Control +## signal is the difference I-C; function is second moment i.e. variance from zero not mean +## to perform wavelet transf. of signal, scale-by-scale analysis of the function +## create null bands by permuting the original data series +## generate plots and table matrix of correlation coefficients including p-values +############################################################################################ +library("wavethresh"); +library("waveslim"); + +options(echo = FALSE) + +## normalize data +norm <- function(data) { + v <- (data - mean(data)) / sd(data); + if (sum(is.na(v)) >= 1) { + v <- data; + } + return(v); +} + +dwt_cor <- function(data_short, names_short, data_long, names_long, test, pdf, table, filter = 4, bc = "symmetric", wf = "haar", boundary = "reflection") { + print(test); + print(pdf); + print(table); + + pdf(file = pdf); + final_pvalue <- NULL; + title <- NULL; + + short_levels <- wavethresh::wd(data_short[, 1], filter.number = filter, bc = bc)$nlevels; + title <- c("motif"); + for (i in 1:short_levels) { + title <- c(title, paste(i, "moment2", sep = "_"), paste(i, "pval", sep = "_"), paste(i, "test", sep = "_")); + } + print(title); + + ## loop to compare a vs a + for (i in seq_len(length(names_short))) { + wave1_dwt <- NULL; + m2_dwt <- NULL; + diff <- NULL; + var_dwt <- NULL; + out <- NULL; + out <- vector(length = length(title)); + + print(names_short[i]); + print(names_long[i]); + + ## need exit if not comparing motif(a) vs motif(a) + if (names_short[i] != names_long[i]) { + stop(paste("motif", names_short[i], "is not the same as", names_long[i], sep = " ")); + } + else { + ## signal is the difference I-C data sets + diff <- data_short[, i] - data_long[, i]; + + ## normalize the signal + diff <- norm(diff); + + ## function is 2nd moment + ## 2nd moment m_j = 1/N[sum_N(W_j + V_J)^2] = 1/N sum_N(W_j)^2 + (X_bar)^2 + wave1_dwt <- waveslim::dwt(diff, wf = wf, short_levels, boundary = boundary); + var_dwt <- waveslim::wave.variance(wave1_dwt); + m2_dwt <- vector(length = short_levels) + for (level in 1:short_levels) { + m2_dwt[level] <- var_dwt[level, 1] + (mean(diff)^2); + } + + ## CI bands by permutation of time series + feature1 <- NULL; + feature2 <- NULL; + feature1 <- data_short[, i]; + feature2 <- data_long[, i]; + null <- NULL; + results <- NULL; + med <- NULL; + m2_25 <- NULL; + m2_975 <- NULL; + + for (k in 1:1000) { + nk_1 <- NULL; + nk_2 <- NULL; + m2_null <- NULL; + var_null <- NULL; + null_levels <- NULL; + null_wave1 <- NULL; + null_diff <- NULL; + nk_1 <- sample(feature1, length(feature1), replace = FALSE); + nk_2 <- sample(feature2, length(feature2), replace = FALSE); + null_levels <- wavethresh::wd(nk_1, filter.number = filter, bc = bc)$nlevels; + null_diff <- nk_1 - nk_2; + null_diff <- norm(null_diff); + null_wave1 <- waveslim::dwt(null_diff, wf = wf, short_levels, boundary = boundary); + var_null <- waveslim::wave.variance(null_wave1); + m2_null <- vector(length = null_levels); + for (level in 1:null_levels) { + m2_null[level] <- var_null[level, 1] + (mean(null_diff)^2); + } + null <- rbind(null, m2_null); + } + + null <- apply(null, 2, sort, na.last = TRUE); + m2_25 <- null[25, ]; + m2_975 <- null[975, ]; + med <- apply(null, 2, median, na.rm = TRUE); + + ## plot + results <- cbind(m2_dwt, m2_25, m2_975); + matplot(results, type = "b", pch = "*", lty = 1, col = c(1, 2, 2), xlab = "Wavelet Scale", ylab = c("Wavelet 2nd Moment", test), main = (names_short[i]), cex.main = 0.75); + abline(h = 1); + + ## get pvalues by comparison to null distribution + out <- c(names_short[i]); + for (m in seq_len(length(m2_dwt))) { + print(paste("scale", m, sep = " ")); + print(paste("m2", m2_dwt[m], sep = " ")); + print(paste("median", med[m], sep = " ")); + out <- c(out, format(m2_dwt[m], digits = 4)); + pv <- NULL; + if (is.na(m2_dwt[m])) { + pv <- "NA"; + } + else { + if (m2_dwt[m] >= med[m]) { + ## R tail test + tail <- "R"; + pv <- (length(which(null[, m] >= m2_dwt[m]))) / (length(na.exclude(null[, m]))); + } + else{ + if (m2_dwt[m] < med[m]) { + ## L tail test + tail <- "L"; + pv <- (length(which(null[, m] <= m2_dwt[m]))) / (length(na.exclude(null[, m]))); + } + } + } + out <- c(out, pv); + print(pv); + out <- c(out, tail); + } + final_pvalue <- rbind(final_pvalue, out); + print(out); + } + } + + colnames(final_pvalue) <- title; + write.table(final_pvalue, file = table, sep = "\t", quote = FALSE, row.names = FALSE); + dev.off(); +} +## execute +## read in data +args <- commandArgs(trailingOnly = TRUE) + +input_data <- read.delim(args[1]); +input_data_names <- colnames(input_data); + +control_data <- read.delim(args[2]); +control_data_names <- colnames(control_data); + +## call the test function to implement IvC test +dwt_cor(input_data, input_data_names, control_data, control_data_names, test = "IvC", pdf = args[3], table = args[4]); +print("done with the correlation test");
--- a/execute_dwt_IvC_all.pl Mon Jan 27 09:25:56 2014 -0500 +++ /dev/null Thu Jan 01 00:00:00 1970 +0000 @@ -1,210 +0,0 @@ -#!/usr/bin/perl -w -use warnings; -use IO::Handle; - -$usage = "execute_dwt_IvC_all.pl [TABULAR.in] [TABULAR.in] [TABULAR.out] [PDF.out] \n"; -die $usage unless @ARGV == 4; - -#get the input arguments -my $firstInputFile = $ARGV[0]; -my $secondInputFile = $ARGV[1]; -my $firstOutputFile = $ARGV[2]; -my $secondOutputFile = $ARGV[3]; - -open (INPUT1, "<", $firstInputFile) || die("Could not open file $firstInputFile \n"); -open (INPUT2, "<", $secondInputFile) || die("Could not open file $secondInputFile \n"); -open (OUTPUT1, ">", $firstOutputFile) || die("Could not open file $firstOutputFile \n"); -open (OUTPUT2, ">", $secondOutputFile) || die("Could not open file $secondOutputFile \n"); -open (ERROR, ">", "error.txt") or die ("Could not open file error.txt \n"); - -#save all error messages into the error file $errorFile using the error file handle ERROR -STDERR -> fdopen( \*ERROR, "w" ) or die ("Could not direct errors to the error file error.txt \n"); - - -print "There are two input data files: \n"; -print "The input data file is: $firstInputFile \n"; -print "The control data file is: $secondInputFile \n"; - -# IvC test -$test = "IvC"; - -# construct an R script to implement the IvC test -print "\n"; - -$r_script = "get_dwt_IvC_test.r"; -print "$r_script \n"; - -# R script -open(Rcmd, ">", "$r_script") or die "Cannot open $r_script \n\n"; -print Rcmd " - ########################################################################################### - # code to do wavelet Indel vs. Control - # signal is the difference I-C; function is second moment i.e. variance from zero not mean - # to perform wavelet transf. of signal, scale-by-scale analysis of the function - # create null bands by permuting the original data series - # generate plots and table matrix of correlation coefficients including p-values - ############################################################################################ - library(\"Rwave\"); - library(\"wavethresh\"); - library(\"waveslim\"); - - options(echo = FALSE) - - # normalize data - norm <- function(data){ - v <- (data - mean(data))/sd(data); - if(sum(is.na(v)) >= 1){ - v <- data; - } - return(v); - } - - dwt_cor <- function(data.short, names.short, data.long, names.long, test, pdf, table, filter = 4, bc = \"symmetric\", wf = \"haar\", boundary = \"reflection\") { - print(test); - print(pdf); - print(table); - - pdf(file = pdf); - final_pvalue = NULL; - title = NULL; - - short.levels <- wd(data.short[, 1], filter.number = filter, bc = bc)\$nlevels; - title <- c(\"motif\"); - for (i in 1:short.levels){ - title <- c(title, paste(i, \"moment2\", sep = \"_\"), paste(i, \"pval\", sep = \"_\"), paste(i, \"test\", sep = \"_\")); - } - print(title); - - # loop to compare a vs a - for(i in 1:length(names.short)){ - wave1.dwt = NULL; - m2.dwt = diff = var.dwt = NULL; - out = NULL; - out <- vector(length = length(title)); - - print(names.short[i]); - print(names.long[i]); - - # need exit if not comparing motif(a) vs motif(a) - if (names.short[i] != names.long[i]){ - stop(paste(\"motif\", names.short[i], \"is not the same as\", names.long[i], sep = \" \")); - } - else { - # signal is the difference I-C data sets - diff<-data.short[,i]-data.long[,i]; - - # normalize the signal - diff<-norm(diff); - - # function is 2nd moment - # 2nd moment m_j = 1/N[sum_N(W_j + V_J)^2] = 1/N sum_N(W_j)^2 + (X_bar)^2 - wave1.dwt <- dwt(diff, wf = wf, short.levels, boundary = boundary); - var.dwt <- wave.variance(wave1.dwt); - m2.dwt <- vector(length = short.levels) - for(level in 1:short.levels){ - m2.dwt[level] <- var.dwt[level, 1] + (mean(diff)^2); - } - - # CI bands by permutation of time series - feature1 = feature2 = NULL; - feature1 = data.short[, i]; - feature2 = data.long[, i]; - null = results = med = NULL; - m2_25 = m2_975 = NULL; - - for (k in 1:1000) { - nk_1 = nk_2 = NULL; - m2_null = var_null = NULL; - null.levels = null_wave1 = null_diff = NULL; - nk_1 <- sample(feature1, length(feature1), replace = FALSE); - nk_2 <- sample(feature2, length(feature2), replace = FALSE); - null.levels <- wd(nk_1, filter.number = filter, bc = bc)\$nlevels; - null_diff <- nk_1-nk_2; - null_diff <- norm(null_diff); - null_wave1 <- dwt(null_diff, wf = wf, short.levels, boundary = boundary); - var_null <- wave.variance(null_wave1); - m2_null <- vector(length = null.levels); - for(level in 1:null.levels){ - m2_null[level] <- var_null[level, 1] + (mean(null_diff)^2); - } - null= rbind(null, m2_null); - } - - null <- apply(null, 2, sort, na.last = TRUE); - m2_25 <- null[25,]; - m2_975 <- null[975,]; - med <- apply(null, 2, median, na.rm = TRUE); - - # plot - results <- cbind(m2.dwt, m2_25, m2_975); - matplot(results, type = \"b\", pch = \"*\", lty = 1, col = c(1, 2, 2), xlab = \"Wavelet Scale\", ylab = c(\"Wavelet 2nd Moment\", test), main = (names.short[i]), cex.main = 0.75); - abline(h = 1); - - # get pvalues by comparison to null distribution - out <- c(names.short[i]); - for (m in 1:length(m2.dwt)){ - print(paste(\"scale\", m, sep = \" \")); - print(paste(\"m2\", m2.dwt[m], sep = \" \")); - print(paste(\"median\", med[m], sep = \" \")); - out <- c(out, format(m2.dwt[m], digits = 4)); - pv = NULL; - if(is.na(m2.dwt[m])){ - pv <- \"NA\"; - } - else { - if (m2.dwt[m] >= med[m]){ - # R tail test - tail <- \"R\"; - pv <- (length(which(null[, m] >= m2.dwt[m])))/(length(na.exclude(null[, m]))); - } - else{ - if (m2.dwt[m] < med[m]){ - # L tail test - tail <- \"L\"; - pv <- (length(which(null[, m] <= m2.dwt[m])))/(length(na.exclude(null[, m]))); - } - } - } - out <- c(out, pv); - print(pv); - out <- c(out, tail); - } - final_pvalue <-rbind(final_pvalue, out); - print(out); - } - } - - colnames(final_pvalue) <- title; - write.table(final_pvalue, file = table, sep = \"\\t\", quote = FALSE, row.names = FALSE); - dev.off(); - }\n"; - -print Rcmd " - # execute - # read in data - - inputData <- read.delim(\"$firstInputFile\"); - inputDataNames <- colnames(inputData); - - controlData <- read.delim(\"$secondInputFile\"); - controlDataNames <- colnames(controlData); - - # call the test function to implement IvC test - dwt_cor(inputData, inputDataNames, controlData, controlDataNames, test = \"$test\", pdf = \"$secondOutputFile\", table = \"$firstOutputFile\"); - print (\"done with the correlation test\"); -\n"; - -print Rcmd "#eof\n"; - -close Rcmd; - -system("echo \"wavelet IvC test started on \`hostname\` at \`date\`\"\n"); -system("R --no-restore --no-save --no-readline < $r_script > $r_script.out\n"); -system("echo \"wavelet IvC test ended on \`hostname\` at \`date\`\"\n"); - -#close the input and output and error files -close(ERROR); -close(OUTPUT2); -close(OUTPUT1); -close(INPUT2); -close(INPUT1); \ No newline at end of file
--- a/execute_dwt_IvC_all.xml Mon Jan 27 09:25:56 2014 -0500 +++ b/execute_dwt_IvC_all.xml Mon Jul 06 20:31:56 2020 -0400 @@ -1,20 +1,35 @@ -<tool id="compute_p-values_second_moments_feature_occurrences_between_two_datasets_using_discrete_wavelet_transfom" name="Compute P-values and Second Moments for Feature Occurrences" version="1.0.0"> +<tool id="compute_p-values_second_moments_feature_occurrences_between_two_datasets_using_discrete_wavelet_transfom" name="Compute P-values and Second Moments for Feature Occurrences" version="1.0.1"> <description>between two datasets using Discrete Wavelet Transfoms</description> - - <command interpreter="perl"> - execute_dwt_IvC_all.pl $inputFile1 $inputFile2 $outputFile1 $outputFile2 + <requirements> + <requirement type="package" version="1.7.5">r-waveslim</requirement> + <requirement type="package" version="4.6.8">r-wavethresh</requirement> + </requirements> + <command detect_errors="exit_code"> + Rscript --vanilla '$__tool_directory__/execute_dwt_IvC_all.R' + '$inputFile1' + '$inputFile2' + '$outputFile2' + '$outputFile1' </command> - <inputs> - <param format="tabular" name="inputFile1" type="data" label="Select the first input file"/> - <param format="tabular" name="inputFile2" type="data" label="Select the second input file"/> + <param format="tabular" name="inputFile1" type="data" label="Select the first input file"/> + <param format="tabular" name="inputFile2" type="data" label="Select the second input file"/> </inputs> - <outputs> - <data format="tabular" name="outputFile1"/> - <data format="pdf" name="outputFile2"/> + <data format="tabular" name="outputFile1" label="${tool.name} on ${on_string}: statistics"/> + <data format="pdf" name="outputFile2" label="${tool.name} on ${on_string}: pdf"/> </outputs> - + <tests> + <test> + <param ftype="tabular" name="inputFile1" value="in1.tsv"/> + <param ftype="tabular" name="inputFile2" value="in2.tsv"/> + <output name="outputFile1" ftype="tabular"> + <assert_contents><has_line_matching expression="^motif\t1_moment2.*"/></assert_contents> + <assert_contents><has_line_matching expression="^translinTarget.*" /></assert_contents> + </output> + <output name="outputFile2" ftype="pdf" file="out2.pdf" compare="sim_size"/> + </test> + </tests> <help> .. class:: infomark
--- /dev/null Thu Jan 01 00:00:00 1970 +0000 +++ b/test-data/in1.tsv Mon Jul 06 20:31:56 2020 -0400 @@ -0,0 +1,17 @@ +deletionHoptspot insertionHoptspot dnaPolPauseFrameshift topoisomeraseCleavageSite translinTarget +269 366 330 238 1129 +239 328 327 283 1188 +254 351 358 297 1151 +262 371 355 256 1107 +254 361 352 234 1192 +265 354 367 240 1182 +255 359 333 235 1217 +271 389 387 272 1241 +240 305 341 249 1159 +272 351 337 257 1169 +275 351 337 233 1158 +305 331 361 253 1172 +277 341 343 253 1113 +266 362 355 267 1162 +235 326 329 241 1230 +254 335 360 251 1172
--- /dev/null Thu Jan 01 00:00:00 1970 +0000 +++ b/test-data/in2.tsv Mon Jul 06 20:31:56 2020 -0400 @@ -0,0 +1,17 @@ +deletionHoptspot insertionHoptspot dnaPolPauseFrameshift topoisomeraseCleavageSite translinTarget +104 146 142 113 478 +89 146 151 94 495 +100 176 151 88 435 +96 163 128 114 468 +99 138 144 91 513 +112 126 162 106 468 +86 127 145 83 491 +104 145 171 110 496 +91 121 147 104 469 +103 141 145 98 458 +92 134 142 117 468 +97 146 145 107 471 +115 121 136 109 470 +113 135 138 101 491 +111 150 138 102 451 +94 128 151 138 481