diff --git a/NAMESPACE b/NAMESPACE
index 23a2879f70b2c867c124ba1507cdb8e566287606..1ae3984ecd4a3fd299c824ae0f780fb02e6bcef3 100644
--- a/NAMESPACE
+++ b/NAMESPACE
@@ -1,22 +1,10 @@
# Generated by roxygen2: do not edit by hand
export(calculate_lr)
-export(calculate_maf)
-export(count_alleles)
-export(count_confidences)
-export(extract_ranges)
-export(find_cpts)
-export(merge_segments)
-export(normalise_lr)
export(read_axiom)
-export(regroup_data)
-export(summarise_segments)
-export(transform_lr)
-import(GenomicRanges)
-import(changepoint)
+export(standardise_lr)
+export(standardize_lr)
import(dplyr)
-import(plyranges)
-import(purrr)
import(readr)
import(stringr)
import(tidyr)
diff --git a/R/calculate_lr.R b/R/calculate_lr.R
index 076e8ef1e45b00bac26e8a7129ad22c23b3aadb9..43d1da1f8fff6bad35723c15614c2bffce3ace90 100644
--- a/R/calculate_lr.R
+++ b/R/calculate_lr.R
@@ -1,7 +1,6 @@
#' Calculate lr values from summary file
#'
#' @param summary_file imported summary file (usually named 'summary')
-#' @param format "wide" or "long" table (defaults to "long")
#'
#' @import dplyr
#' @import tidyr
@@ -11,10 +10,10 @@
#'
#' @examples
#' \dontrun{
-#' calculate_lr(summary_file, format = "long")
+#' calculate_lr(summary_file)
#' }
-calculate_lr <- function(summary_file, format = "long") {
+calculate_lr <- function(summary_file) {
# Set NULL variables
@@ -47,15 +46,6 @@ calculate_lr <- function(summary_file, format = "long") {
dplyr::mutate(lr = log2(sqrt(signal_a^2 + signal_b^2)),
.keep = "unused")
- # If format is set to "wide", transform the table
- if (format == "wide") {
- message("Pivoting output table to wide format")
- lr <- lr |>
- tidyr::pivot_wider(id_cols = probeset_id,
- names_from = file_name,
- values_from = lr)
- }
-
# Assign to Global Environment
assign(x = "lr", value = lr, pos = ".GlobalEnv")
diff --git a/R/merge_segments.R b/R/merge_segments.R
deleted file mode 100644
index 6ccd03f0e65f224d6b278140f21b9255895c3520..0000000000000000000000000000000000000000
--- a/R/merge_segments.R
+++ /dev/null
@@ -1,95 +0,0 @@
-#' Merge consecutive segments and recalculate parameters
-#'
-#' @param segments_file segments file
-#' @param regrouped_data_file lrn & calls file
-#'
-#' @import dplyr
-#' @import GenomicRanges
-#'
-#' @return A [tibble()].
-#' @export
-#'
-#' @examples
-#' \dontrun{
-#' merge_segments(segments_file, regrouped_data_file)
-#' }
-
-merge_segments <- function(segments_file, regrouped_data_file) {
-
- # Set NULL variables
-
- chromosome <- file_name <- group_name <- high_lrn_count <- mseg_end <-
- mseg_start <- position <- seg_nb <- value <- NULL
-
- # Merge consecutive segments
- merged_segments <- segments_file %>%
- dplyr::select(file_name, chromosome, start = seg_nb, end = seg_nb) %>%
- GenomicRanges::makeGRangesListFromDataFrame(split.field = "file_name") %>%
- GenomicRanges::reduce() %>%
- tibble::as_tibble() %>%
- dplyr::select(file_name = group_name,
- chromosome = seqnames,
- mseg_start = start,
- mseg_end = end)
-
- # Keep non-merged segments
- same_segments <- merged_segments %>%
- dplyr::filter(mseg_start == mseg_end) %>%
- dplyr::select(file_name, chromosome, seg_nb = mseg_start) %>%
- dplyr::left_join(segments_file) %>%
- dplyr::select(file_name, chromosome, start:high_lrn_count)
-
- if (nrow(same_segments) != nrow(merged_segments)) {
-
- # Extract merged segments and recalculate parameters
- diff_segments <- merged_segments %>%
- dplyr::filter(mseg_start != mseg_end) %>%
- tidyr::pivot_longer(!c(file_name, chromosome)) %>%
- dplyr::select(file_name, chromosome, seg_nb = value) %>%
- dplyr::left_join(segments_file) %>%
- dplyr::select(file_name, chromosome, start:end) %>%
- dplyr::group_by(file_name, chromosome) %>%
- dplyr::mutate(start = min(start),
- end = max(end)) %>%
- dplyr::slice(1)
-
- lrn_means <- list()
- snp_counts <- list()
- otv_counts <- list()
- low_lrn_counts <- list()
- high_lrn_counts <- list()
-
- for (i in 1:nrow(diff_segments)) {
- d1 <- regrouped_data_file %>%
- dplyr::filter(file_name == diff_segments$file_name[i] &
- chromosome == diff_segments$chromosome[i] &
- position >= diff_segments$start[i] &
- position <= diff_segments$end[i])
-
- lrn_means[[i]] <- mean(d1$lrn)
- snp_counts[[i]] <- nrow(d1)
- otv_counts[[i]] <- length(which(d1$call == "-2"))
- low_lrn_counts[[i]] <- length(which(d1$lrn <= -1))
- high_lrn_counts[[i]] <- length(which(d1$lrn >= 1))
- }
-
- diff_segments$lrn_mean <- unlist(lrn_means)
- diff_segments$snp_count <- unlist(snp_counts)
- diff_segments$otv_count <- unlist(otv_counts)
- diff_segments$low_lrn_count <- unlist(low_lrn_counts)
- diff_segments$high_lrn_count <- unlist(high_lrn_counts)
-
- # rbind same_segments & diff_segments + reorder
- final_mseg <- rbind(same_segments, diff_segments)
-
- }
-
- if (nrow(same_segments) == nrow(merged_segments)) {
-
- final_mseg <- same_segments
-
- }
-
- return(final_mseg)
-
-}
diff --git a/R/standardise_lr.R b/R/standardise_lr.R
new file mode 100644
index 0000000000000000000000000000000000000000..f7b3495a1c94f87bab49229b6c37582c18b10d99
--- /dev/null
+++ b/R/standardise_lr.R
@@ -0,0 +1,62 @@
+#' Standardise lr values
+#'
+#' @param lr lr values
+#' @param method how to perform standardisation ("ref", "mean", or "both", defaults to "mean")
+#' @param ref name of the sample to use for standardisation (must be defined if method is "ref" or "both", defaults to NULL)
+#'
+#' @return a [tibble()]
+#' @export
+#'
+#' @examples
+#' \dontrun{
+#' standardise_lr(lr, method = "mean", ref = NULL)
+#' }
+
+standardise_lr <- function(lr, method = "mean", ref = NULL) {
+
+ # Set NULL variables
+
+ file_name <- probeset_id <- NULL
+
+ # If method = "ref" or "both", check if ref is defined
+ if(method %in% c("ref", "both")){
+ print("Checking if reference sample is defined")
+ if(is.null(ref)){
+ stop("Reference sample is not defined")
+ }
+ if(!is.null(ref) & !ref %in% unique(lr$file_name)) {
+ stop("Reference sample doesn't exist in the list of samples")
+ }
+ }
+
+ # If method = "mean", calculate lr_st_mean
+ if (method == "mean") {
+ lr <- lr |>
+ dplyr::mutate(lr_mean = mean(lr, na.rm = TRUE),
+ .by = probeset_id) |>
+ dplyr::mutate(lr_st_mean = lr - lr_mean) |>
+ dplyr::select(-lr_mean)
+ }
+
+ # If method = "ref", calculate lr_st_ref
+ if (method == "ref") {
+ lr <- lr |>
+ dplyr::mutate(lr_st_ref = lr - lr[file_name == ref],
+ .by = probeset_id)
+ }
+
+ # If method = "both", calculate lr_st_ref & lr_st_mean
+ if (method == "both") {
+ lr <- lr |>
+ dplyr::mutate(lr_st_ref = lr - lr[file_name == ref],
+ lr_mean = mean(lr, na.rm = TRUE),
+ .by = probeset_id) |>
+ dplyr::mutate(lr_st_mean = lr - lr_mean,
+ .before = lr_st_ref) |>
+ dplyr::select(-lr_mean)
+ }
+
+ # Assign to Global Environment
+ assign(x = "lr", value = lr, pos = ".GlobalEnv")
+
+}
diff --git a/R/standardize_lr.R b/R/standardize_lr.R
new file mode 100644
index 0000000000000000000000000000000000000000..fa780b3b08cf115dceab5211e4643c7b1e7ce7c4
--- /dev/null
+++ b/R/standardize_lr.R
@@ -0,0 +1,62 @@
+#' Standardize lr values
+#'
+#' @param lr lr values
+#' @param method how to perform standardization ("ref", "mean", or "both", defaults to "mean")
+#' @param ref name of the sample to use for standardization (must be defined if method is "ref" or "both", defaults to NULL)
+#'
+#' @return a [tibble()]
+#' @export
+#'
+#' @examples
+#' \dontrun{
+#' standardize_lr(lr, method = "mean", ref = NULL)
+#' }
+
+standardize_lr <- function(lr, method = "mean", ref = NULL) {
+
+ # Set NULL variables
+
+ file_name <- probeset_id <- NULL
+
+ # If method = "ref" or "both", check if ref is defined
+ if(method %in% c("ref", "both")){
+ print("Checking if reference sample is defined")
+ if(is.null(ref)){
+ stop("Reference sample is not defined")
+ }
+ if(!is.null(ref) & !ref %in% unique(lr$file_name)) {
+ stop("Reference sample doesn't exist in the list of samples")
+ }
+ }
+
+ # If method = "mean", calculate lr_st_mean
+ if (method == "mean") {
+ lr <- lr |>
+ dplyr::mutate(lr_mean = mean(lr, na.rm = TRUE),
+ .by = probeset_id) |>
+ dplyr::mutate(lr_st_mean = lr - lr_mean) |>
+ dplyr::select(-lr_mean)
+ }
+
+ # If method = "ref", calculate lr_st_ref
+ if (method == "ref") {
+ lr <- lr |>
+ dplyr::mutate(lr_st_ref = lr - lr[file_name == ref],
+ .by = probeset_id)
+ }
+
+ # If method = "both", calculate lr_st_ref & lr_st_mean
+ if (method == "both") {
+ lr <- lr |>
+ dplyr::mutate(lr_st_ref = lr - lr[file_name == ref],
+ lr_mean = mean(lr, na.rm = TRUE),
+ .by = probeset_id) |>
+ dplyr::mutate(lr_st_mean = lr - lr_mean,
+ .before = lr_st_ref) |>
+ dplyr::select(-lr_mean)
+ }
+
+ # Assign to Global Environment
+ assign(x = "lr", value = lr, pos = ".GlobalEnv")
+
+}
diff --git a/man/calculate_lr.Rd b/man/calculate_lr.Rd
index 38748af93ba130eb9f5b08d9252228f9413da475..d9a21914ecd766e173db37f3ed31c666d6827bbf 100644
--- a/man/calculate_lr.Rd
+++ b/man/calculate_lr.Rd
@@ -4,12 +4,10 @@
\alias{calculate_lr}
\title{Calculate lr values from summary file}
\usage{
-calculate_lr(summary_file, format = "long")
+calculate_lr(summary_file)
}
\arguments{
\item{summary_file}{imported summary file (usually named 'summary')}
-
-\item{format}{"wide" or "long" table (defaults to "long")}
}
\value{
A \code{\link[=tibble]{tibble()}}.
@@ -19,6 +17,6 @@ Calculate lr values from summary file
}
\examples{
\dontrun{
-calculate_lr(summary_file, format = "long")
+calculate_lr(summary_file)
}
}
diff --git a/man/calculate_maf.Rd b/man/calculate_maf.Rd
deleted file mode 100644
index 0b7ba2cb890d1fac8f99e3819f5f4ba85a1829da..0000000000000000000000000000000000000000
--- a/man/calculate_maf.Rd
+++ /dev/null
@@ -1,22 +0,0 @@
-% Generated by roxygen2: do not edit by hand
-% Please edit documentation in R/calculate_maf.R
-\name{calculate_maf}
-\alias{calculate_maf}
-\title{Calculate MAF from allele counts}
-\usage{
-calculate_maf(allele_count)
-}
-\arguments{
-\item{allele_count}{allele count file}
-}
-\value{
-A \code{\link[=tibble]{tibble()}}.
-}
-\description{
-Calculate MAF from allele counts
-}
-\examples{
-\dontrun{
-calculate_maf(allele_count)
-}
-}
diff --git a/man/count_alleles.Rd b/man/count_alleles.Rd
deleted file mode 100644
index 647e301c9c6109fea51294d67d7235b1bb35c6b7..0000000000000000000000000000000000000000
--- a/man/count_alleles.Rd
+++ /dev/null
@@ -1,22 +0,0 @@
-% Generated by roxygen2: do not edit by hand
-% Please edit documentation in R/count_alleles.R
-\name{count_alleles}
-\alias{count_alleles}
-\title{Count alleles for each SNP in AximGT1.calls.txt data}
-\usage{
-count_alleles(calls_file)
-}
-\arguments{
-\item{calls_file}{calls file}
-}
-\value{
-A \code{\link[=tibble]{tibble()}}.
-}
-\description{
-Count alleles for each SNP in AximGT1.calls.txt data
-}
-\examples{
-\dontrun{
-count_alleles(calls_file)
-}
-}
diff --git a/man/count_confidences.Rd b/man/count_confidences.Rd
deleted file mode 100644
index 0a1210afb0fe84ff0461e449d341d1ed02e105ac..0000000000000000000000000000000000000000
--- a/man/count_confidences.Rd
+++ /dev/null
@@ -1,24 +0,0 @@
-% Generated by roxygen2: do not edit by hand
-% Please edit documentation in R/count_confidences.R
-\name{count_confidences}
-\alias{count_confidences}
-\title{Count number of values above confidence score threshold for each SNP in AxiomGT1.confidences.txt data}
-\usage{
-count_confidences(confidences_file, threshold = 0.15)
-}
-\arguments{
-\item{confidences_file}{confidences file}
-
-\item{threshold}{confidence score threshold (defaults to 0.15)}
-}
-\value{
-A \code{\link[=tibble]{tibble()}}.
-}
-\description{
-Count number of values above confidence score threshold for each SNP in AxiomGT1.confidences.txt data
-}
-\examples{
-\dontrun{
-count_confidences(confidences_file, threshold)
-}
-}
diff --git a/man/extract_ranges.Rd b/man/extract_ranges.Rd
deleted file mode 100644
index 555151e014f82c84da70e983697cd6d6fe6b46d7..0000000000000000000000000000000000000000
--- a/man/extract_ranges.Rd
+++ /dev/null
@@ -1,22 +0,0 @@
-% Generated by roxygen2: do not edit by hand
-% Please edit documentation in R/extract_ranges.R
-\name{extract_ranges}
-\alias{extract_ranges}
-\title{Extract ranges}
-\usage{
-extract_ranges(snps)
-}
-\arguments{
-\item{snps}{list of SNPs}
-}
-\value{
-a table with minimum and maximum SNP positions for each chromosome
-}
-\description{
-Extract ranges
-}
-\examples{
-\dontrun{
-extract_ranges(snps)
-}
-}
diff --git a/man/find_cpts.Rd b/man/find_cpts.Rd
deleted file mode 100644
index a51851778ad4291f2d95dddb3de43d4ef9a2db91..0000000000000000000000000000000000000000
--- a/man/find_cpts.Rd
+++ /dev/null
@@ -1,46 +0,0 @@
-% Generated by roxygen2: do not edit by hand
-% Please edit documentation in R/find_cpts.R
-\name{find_cpts}
-\alias{find_cpts}
-\title{Find changepoints in lrn data}
-\usage{
-find_cpts(
- lr_st,
- calc = "mean",
- var = "lr_st_ref",
- method = "BinSeg",
- penalty = "MBIC",
- pen_value = 0,
- max_cpts = 3,
- count_otvs = TRUE
-)
-}
-\arguments{
-\item{lr_st}{standardized lr values}
-
-\item{calc}{changepoints will be calculated w.r.t. this value ("mean", "var" or "mean_var")}
-
-\item{var}{variable to be used for changepoint detection}
-
-\item{method}{changepoint detection method (defaults to "BinSeg")}
-
-\item{penalty}{changepoint detection penalty (defaults to "MBIC")}
-
-\item{pen_value}{penalty value (defaults to 0)}
-
-\item{max_cpts}{maximum number of changepoints (defaults to 3)}
-
-\item{count_otvs}{whether or not to add number of OTVs in detected segments (defaults to TRUE)}
-}
-\value{
-a \code{\link[=tibble]{tibble()}}
-}
-\description{
-Find changepoints in lrn data
-}
-\examples{
-\dontrun{
-find_cpts(lr_st, calc = "mean", var = "lr_st_ref", method = "BinSeg",
-penalty = "MBIC", pen_value = 0, max_cpts = 3, count_otvs = TRUE)
-}
-}
diff --git a/man/merge_segments.Rd b/man/merge_segments.Rd
deleted file mode 100644
index f4276c81830f41442a778e5b449756c1d9740d9b..0000000000000000000000000000000000000000
--- a/man/merge_segments.Rd
+++ /dev/null
@@ -1,24 +0,0 @@
-% Generated by roxygen2: do not edit by hand
-% Please edit documentation in R/merge_segments.R
-\name{merge_segments}
-\alias{merge_segments}
-\title{Merge consecutive segments and recalculate parameters}
-\usage{
-merge_segments(segments_file, regrouped_data_file)
-}
-\arguments{
-\item{segments_file}{segments file}
-
-\item{regrouped_data_file}{lrn & calls file}
-}
-\value{
-A \code{\link[=tibble]{tibble()}}.
-}
-\description{
-Merge consecutive segments and recalculate parameters
-}
-\examples{
-\dontrun{
-merge_segments(segments_file, regrouped_data_file)
-}
-}
diff --git a/man/normalise_lr.Rd b/man/normalise_lr.Rd
deleted file mode 100644
index 8c07644da28f291d6c501e12233c8e88440c12be..0000000000000000000000000000000000000000
--- a/man/normalise_lr.Rd
+++ /dev/null
@@ -1,24 +0,0 @@
-% Generated by roxygen2: do not edit by hand
-% Please edit documentation in R/normalise_lr.R
-\name{normalise_lr}
-\alias{normalise_lr}
-\title{Normalise lr values}
-\usage{
-normalise_lr(lr, norm)
-}
-\arguments{
-\item{lr}{lr values}
-
-\item{norm}{sample used for normalisation}
-}
-\value{
-a \code{\link[=tibble]{tibble()}}
-}
-\description{
-Normalise lr values
-}
-\examples{
-\dontrun{
-normalise_lr(lr, norm)
-}
-}
diff --git a/man/regroup_data.Rd b/man/regroup_data.Rd
deleted file mode 100644
index 48b1de2edb732509914a70d127cc138afa33a8f7..0000000000000000000000000000000000000000
--- a/man/regroup_data.Rd
+++ /dev/null
@@ -1,24 +0,0 @@
-% Generated by roxygen2: do not edit by hand
-% Please edit documentation in R/regroup_data.R
-\name{regroup_data}
-\alias{regroup_data}
-\title{Regroup lrn and calls data}
-\usage{
-regroup_data(lrn_file, calls_file)
-}
-\arguments{
-\item{lrn_file}{lrn file}
-
-\item{calls_file}{calls file}
-}
-\value{
-A \code{\link[=tibble]{tibble()}}.
-}
-\description{
-Regroup lrn and calls data
-}
-\examples{
-\dontrun{
-regroup_data(lrn_file, calls_file)
-}
-}
diff --git a/man/standardise_lr.Rd b/man/standardise_lr.Rd
new file mode 100644
index 0000000000000000000000000000000000000000..48f2038be8fad0d10e3d5f85a004c2bb9885cf19
--- /dev/null
+++ b/man/standardise_lr.Rd
@@ -0,0 +1,26 @@
+% Generated by roxygen2: do not edit by hand
+% Please edit documentation in R/standardise_lr.R
+\name{standardise_lr}
+\alias{standardise_lr}
+\title{Standardise lr values}
+\usage{
+standardise_lr(lr, method = "mean", ref = NULL)
+}
+\arguments{
+\item{lr}{lr values}
+
+\item{method}{how to perform standardisation ("ref", "mean", or "both", defaults to "mean")}
+
+\item{ref}{name of the sample to use for standardisation (must be defined if method is "ref" or "both", defaults to NULL)}
+}
+\value{
+a \code{\link[=tibble]{tibble()}}
+}
+\description{
+Standardise lr values
+}
+\examples{
+\dontrun{
+standardise_lr(lr, method = "mean", ref = NULL)
+}
+}
diff --git a/man/standardize_lr.Rd b/man/standardize_lr.Rd
new file mode 100644
index 0000000000000000000000000000000000000000..87925587c6c142d9e9989cd9c54c406c736398f0
--- /dev/null
+++ b/man/standardize_lr.Rd
@@ -0,0 +1,26 @@
+% Generated by roxygen2: do not edit by hand
+% Please edit documentation in R/standardize_lr.R
+\name{standardize_lr}
+\alias{standardize_lr}
+\title{Standardize lr values}
+\usage{
+standardize_lr(lr, method = "mean", ref = NULL)
+}
+\arguments{
+\item{lr}{lr values}
+
+\item{method}{how to perform standardization ("ref", "mean", or "both", defaults to "mean")}
+
+\item{ref}{name of the sample to use for standardization (must be defined if method is "ref" or "both", defaults to NULL)}
+}
+\value{
+a \code{\link[=tibble]{tibble()}}
+}
+\description{
+Standardize lr values
+}
+\examples{
+\dontrun{
+standardize_lr(lr, method = "mean", ref = NULL)
+}
+}
diff --git a/man/summarise_segments.Rd b/man/summarise_segments.Rd
deleted file mode 100644
index 7644dd730c8eb1519d0367792b25652169e1a15d..0000000000000000000000000000000000000000
--- a/man/summarise_segments.Rd
+++ /dev/null
@@ -1,26 +0,0 @@
-% Generated by roxygen2: do not edit by hand
-% Please edit documentation in R/summarise_segments.R
-\name{summarise_segments}
-\alias{summarise_segments}
-\title{Summarise segments : count OTVs}
-\usage{
-summarise_segments(raw_segments_file, regrouped_data_file, snps_file)
-}
-\arguments{
-\item{raw_segments_file}{raw segments file}
-
-\item{regrouped_data_file}{file containing lrn values and calls}
-
-\item{snps_file}{list of SNPs}
-}
-\value{
-A \code{\link[=tibble]{tibble()}}.
-}
-\description{
-Summarise segments : count OTVs
-}
-\examples{
-\dontrun{
-summarise_segments(raw_segments_file, regrouped_data_file, snps_file)
-}
-}
diff --git a/man/transform_lr.Rd b/man/transform_lr.Rd
deleted file mode 100644
index 384864ab9da65f81ff1ca6d765e48f37bf62f7b3..0000000000000000000000000000000000000000
--- a/man/transform_lr.Rd
+++ /dev/null
@@ -1,26 +0,0 @@
-% Generated by roxygen2: do not edit by hand
-% Please edit documentation in R/transform_lr.R
-\name{transform_lr}
-\alias{transform_lr}
-\title{Transform lr values}
-\usage{
-transform_lr(lr, method = "ref", ref)
-}
-\arguments{
-\item{lr}{lr values}
-
-\item{method}{how to perform standardization ("ref", "mean", or "both")}
-
-\item{ref}{sample used for standardization}
-}
-\value{
-a \code{\link[=tibble]{tibble()}}
-}
-\description{
-Transform lr values
-}
-\examples{
-\dontrun{
-transform_lr(lr, method = "ref", ref)
-}
-}
diff --git a/R/calculate_maf.R b/misc/calculate_maf.R
similarity index 100%
rename from R/calculate_maf.R
rename to misc/calculate_maf.R
diff --git a/R/count_alleles.R b/misc/count_alleles.R
similarity index 100%
rename from R/count_alleles.R
rename to misc/count_alleles.R
diff --git a/R/count_confidences.R b/misc/count_confidences.R
similarity index 100%
rename from R/count_confidences.R
rename to misc/count_confidences.R
diff --git a/R/extract_ranges.R b/misc/extract_ranges.R
similarity index 100%
rename from R/extract_ranges.R
rename to misc/extract_ranges.R
diff --git a/R/find_cpts.R b/misc/find_cpts.R
similarity index 100%
rename from R/find_cpts.R
rename to misc/find_cpts.R
diff --git a/misc/merge_segments.R b/misc/merge_segments.R
index 249ce8a1bcadf2577408ba5fedefa9812fa0bd92..6ccd03f0e65f224d6b278140f21b9255895c3520 100644
--- a/misc/merge_segments.R
+++ b/misc/merge_segments.R
@@ -1,198 +1,95 @@
-#' Filter and merge segments obtained from changepoint detection
+#' Merge consecutive segments and recalculate parameters
#'
-#' @description
-#' This function filters and merges segments detected using the \code{detect_cpts()} function.
+#' @param segments_file segments file
+#' @param regrouped_data_file lrn & calls file
#'
-#' @import data.table
-#' @importFrom plyr join
-#' @importFrom GenomicRanges makeGRangesListFromDataFrame reduce
-#'
-#' @param array name of the array to be analyzed.
-#' @param raw_seg_file path to the \code{raw_segments} file.
-#' @param mean_cutoff threshold lrn_mean value (absolute value) used to determine whether a segment should
-#' be kept (defaults to 0.25).
-#' @param size_cutoff threshold segment size value used to determine whether a segment should be kept (defaults to 500kb).
-#' @param output_path where the results will be saved
-#'
-#' @details
-#' The merging is done in two consecutive steps :
-#' - merge consecutive segments with a lrn_mean value regardless of gaps
-#' - merge segments separated by another segment with a size inferior to size_cutoff
-#'
-#' @usage merge_segments(array, raw_seg_file, mean_cutoff = 0.25,
-#' size_cutoff = 500e3, output_path)
-#'
-#' @return A \code{merged_segments} data.table containing the following columns :
-#' Returns a table containing the following columns :
-#' - species
-#' - array_type
-#' - array_name
-#' - unique_id
-#' - sample_name
-#' - chromosome
-#' - start
-#' - end
-#' - sv_type
+#' @import dplyr
+#' @import GenomicRanges
#'
+#' @return A [tibble()].
#' @export
#'
#' @examples
#' \dontrun{
-#' merge_segments(array, raw_seg_file, mean_cutoff, size_cutoff, output_path)
+#' merge_segments(segments_file, regrouped_data_file)
#' }
-merge_segments <- function(array, raw_seg_file, mean_cutoff = 0.25,
- size_cutoff = 500e3, output_path) {
-
- lrn_mean <- end <- start <- nb_cpts <- seg_nb <-
- . <- array_name <- analysis_group <- unique_id <-
- chromosome <- head <- size <- width <- lrn_cutoff <-
- raw_seg_file <- NULL
-
- # Check arguments ----
-
- if (missing(array)) stop("array is required")
- if (missing(raw_seg_file)) stop("raw_seg_file is required")
- if (missing(output_path)) stop("output_path is required")
-
- # Create folders for saving results ----
-
- if (!dir.exists(paste0(output_path, "results/"))) {
- dir.create(paste0(output_path, "results/"))
- }
-
- if (!dir.exists(paste0(output_path, "results/merged_segments"))) {
- dir.create(paste0(output_path, "results/merged_segments/"))
- }
-
- # 1. Import data and prepare files ----
-
- # 1.1. import segmentation file
- raw_seg <- data.table::fread(raw_seg_file)
- rm(raw_seg_file)
-
- # 1.3. extract list of samples
- smp <- raw_seg[, 1:5]
- smp <- smp[!duplicated(smp), ]
-
- # 1.4. split segments into seg_neg & seg_pos + filter
- seg_neg <- raw_seg[lrn_mean <= -mean_cutoff][, c(4, 7, 9:11)]
- seg_pos <- raw_seg[lrn_mean >= mean_cutoff][, c(4, 7, 9:11)]
- rm(raw_seg)
-
- # 2. Merge consecutive segments ----
-
- # 2.1. create GRangesLists
-
- seg_nb_neg <- seg_neg[, c(1:3, 3)]
- seg_nb_pos <- seg_pos[, c(1:3, 3)]
-
- names(seg_nb_neg) <- c("unique_id", "chromosome", "start", "end")
- names(seg_nb_pos) <- c("unique_id", "chromosome", "start", "end")
+merge_segments <- function(segments_file, regrouped_data_file) {
+
+ # Set NULL variables
+
+ chromosome <- file_name <- group_name <- high_lrn_count <- mseg_end <-
+ mseg_start <- position <- seg_nb <- value <- NULL
+
+ # Merge consecutive segments
+ merged_segments <- segments_file %>%
+ dplyr::select(file_name, chromosome, start = seg_nb, end = seg_nb) %>%
+ GenomicRanges::makeGRangesListFromDataFrame(split.field = "file_name") %>%
+ GenomicRanges::reduce() %>%
+ tibble::as_tibble() %>%
+ dplyr::select(file_name = group_name,
+ chromosome = seqnames,
+ mseg_start = start,
+ mseg_end = end)
+
+ # Keep non-merged segments
+ same_segments <- merged_segments %>%
+ dplyr::filter(mseg_start == mseg_end) %>%
+ dplyr::select(file_name, chromosome, seg_nb = mseg_start) %>%
+ dplyr::left_join(segments_file) %>%
+ dplyr::select(file_name, chromosome, start:high_lrn_count)
+
+ if (nrow(same_segments) != nrow(merged_segments)) {
+
+ # Extract merged segments and recalculate parameters
+ diff_segments <- merged_segments %>%
+ dplyr::filter(mseg_start != mseg_end) %>%
+ tidyr::pivot_longer(!c(file_name, chromosome)) %>%
+ dplyr::select(file_name, chromosome, seg_nb = value) %>%
+ dplyr::left_join(segments_file) %>%
+ dplyr::select(file_name, chromosome, start:end) %>%
+ dplyr::group_by(file_name, chromosome) %>%
+ dplyr::mutate(start = min(start),
+ end = max(end)) %>%
+ dplyr::slice(1)
+
+ lrn_means <- list()
+ snp_counts <- list()
+ otv_counts <- list()
+ low_lrn_counts <- list()
+ high_lrn_counts <- list()
+
+ for (i in 1:nrow(diff_segments)) {
+ d1 <- regrouped_data_file %>%
+ dplyr::filter(file_name == diff_segments$file_name[i] &
+ chromosome == diff_segments$chromosome[i] &
+ position >= diff_segments$start[i] &
+ position <= diff_segments$end[i])
+
+ lrn_means[[i]] <- mean(d1$lrn)
+ snp_counts[[i]] <- nrow(d1)
+ otv_counts[[i]] <- length(which(d1$call == "-2"))
+ low_lrn_counts[[i]] <- length(which(d1$lrn <= -1))
+ high_lrn_counts[[i]] <- length(which(d1$lrn >= 1))
+ }
+
+ diff_segments$lrn_mean <- unlist(lrn_means)
+ diff_segments$snp_count <- unlist(snp_counts)
+ diff_segments$otv_count <- unlist(otv_counts)
+ diff_segments$low_lrn_count <- unlist(low_lrn_counts)
+ diff_segments$high_lrn_count <- unlist(high_lrn_counts)
+
+ # rbind same_segments & diff_segments + reorder
+ final_mseg <- rbind(same_segments, diff_segments)
- seg_nb_neg <- GenomicRanges::makeGRangesListFromDataFrame(df = seg_nb_neg, split.field = "unique_id")
- seg_nb_pos <- GenomicRanges::makeGRangesListFromDataFrame(df = seg_nb_pos,split.field = "unique_id")
-
- # 2.2. reduce GRangesLists
-
- mseg_nb_neg <- GenomicRanges::reduce(seg_nb_neg)
- mseg_nb_pos <- GenomicRanges::reduce(seg_nb_pos)
-
- # 2.3. create tables
-
- mseg_nb_neg_list <- list()
- for (i in 1:length(mseg_nb_neg)) {
- mseg_nb_neg_list[[i]] <- data.table::as.data.table(mseg_nb_neg[[i]])
- mseg_nb_neg_list[[i]]$unique_id <- names(mseg_nb_neg[i])
- }
- mseg_nb_neg_tb <- data.table::rbindlist(mseg_nb_neg_list)[, c(1:3, 6)]
-
- mseg_nb_pos_list <- list()
- for (i in 1:length(mseg_nb_pos)) {
- mseg_nb_pos_list[[i]] <- data.table::as.data.table(mseg_nb_pos[[i]])
- mseg_nb_pos_list[[i]]$unique_id <- names(mseg_nb_pos[i])
- }
- mseg_nb_pos_tb <- data.table::rbindlist(mseg_nb_pos_list)[, c(1:3, 6)]
-
- # 2.4. extract physical positions
-
- mseg_nb_neg <- list()
- for (i in 1:nrow(mseg_nb_neg_tb)) {
- d1 <- seg_neg[unique_id == mseg_nb_neg_tb$unique_id[i] & chromosome == mseg_nb_neg_tb$seqnames[i]]
- mseg_nb_neg[[i]] <- data.table::data.table(
- chromosome = unique(d1$chromosome),
- unique_id = unique(d1$unique_id),
- start = d1$start[d1$seg_nb == mseg_nb_neg_tb$start[i]],
- end = d1$end[d1$seg_nb == mseg_nb_neg_tb$end[i]]
- )
- }
- mseg_nb_neg_dt <- data.table::rbindlist(mseg_nb_neg)
-
- mseg_nb_pos <- list()
- for (i in 1:nrow(mseg_nb_pos_tb)) {
- d1 <- seg_pos[unique_id == mseg_nb_pos_tb$unique_id[i] & chromosome == mseg_nb_pos_tb$seqnames[i]]
- mseg_nb_pos[[i]] <- data.table::data.table(
- chromosome = unique(d1$chromosome),
- unique_id = unique(d1$unique_id),
- start = d1$start[d1$seg_nb == mseg_nb_pos_tb$start[i]],
- end = d1$end[d1$seg_nb == mseg_nb_pos_tb$end[i]]
- )
}
- mseg_nb_pos_dt <- data.table::rbindlist(mseg_nb_pos)
-
- rm(d1, mseg_nb_neg, mseg_nb_neg_tb, mseg_nb_pos, mseg_nb_pos_tb,
- seg_nb_neg, seg_nb_pos, seg_neg, seg_pos, i, mean_cutoff, mseg_nb_neg_list, mseg_nb_pos_list)
-
- # 3. Merge segments according to gaps ----
- # 3.1. create GRangesLists
+ if (nrow(same_segments) == nrow(merged_segments)) {
- seg_neg <- GenomicRanges::makeGRangesListFromDataFrame(df = mseg_nb_neg_dt, split.field = "unique_id")
- seg_pos <- GenomicRanges::makeGRangesListFromDataFrame(df = mseg_nb_pos_dt, split.field = "unique_id")
+ final_mseg <- same_segments
- # 3.2. reduce GRangesLists
-
- mseg_neg <- GenomicRanges::reduce(seg_neg, min.gapwidth = size_cutoff)
- mseg_pos <- GenomicRanges::reduce(seg_pos, min.gapwidth = size_cutoff)
-
- merged_segments_neg <- list()
- for (i in 1:length(mseg_neg)) {
- merged_segments_neg[[i]] <- data.table::as.data.table(mseg_neg[[i]])
- merged_segments_neg[[i]]$unique_id <- names(mseg_neg[i])
- merged_segments_neg[[i]]$sv_type <- "negative"
- }
- merged_segments_neg_table <- data.table::rbindlist(merged_segments_neg)
- merged_segments_neg_table <- merged_segments_neg_table[width >= size_cutoff][, c(1:3, 6:7)]
-
- merged_segments_pos <- list()
- for (i in 1:length(mseg_pos)) {
- merged_segments_pos[[i]] <- data.table::as.data.table(mseg_pos[[i]])
- merged_segments_pos[[i]]$unique_id <- names(mseg_pos[i])
- merged_segments_pos[[i]]$sv_type <- "positive"
}
- merged_segments_pos_table <- data.table::rbindlist(merged_segments_pos)
- merged_segments_pos_table <- merged_segments_pos_table[width >= size_cutoff][, c(1:3, 6:7)]
-
- # 3.3. join negative and positive tables
-
- merged_segments <- rbind(merged_segments_neg_table, merged_segments_pos_table)
- names(merged_segments) <- c("chromosome", names(merged_segments)[-1])
-
- rm(merged_segments_neg, merged_segments_neg_table, merged_segments_pos, merged_segments_pos_table,
- mseg_nb_neg_dt, mseg_nb_pos_dt, mseg_neg, mseg_pos, seg_neg, seg_pos, i, size_cutoff)
-
- # 3.4. join merged segments with samples
- mseg_final <- plyr::join(merged_segments, smp)
- mseg_final <- mseg_final[, c(6:8, 4, 9, 1:3, 5)]
-
- rm(merged_segments, smp)
-
- # 4. Save output ----
-
- data.table::fwrite(mseg_final,
- paste0(output_path, "results/merged_segments/", array, "_merged_segments.txt"),
- quote = FALSE, row.names = FALSE, sep = "\t")
- rm(mseg_final, array, raw_seg_file, output_path)
+ return(final_mseg)
}
diff --git a/R/normalise_lr.R b/misc/normalise_lr.R
similarity index 100%
rename from R/normalise_lr.R
rename to misc/normalise_lr.R
diff --git a/R/regroup_data.R b/misc/regroup_data.R
similarity index 100%
rename from R/regroup_data.R
rename to misc/regroup_data.R
diff --git a/R/summarise_segments.R b/misc/summarise_segments.R
similarity index 100%
rename from R/summarise_segments.R
rename to misc/summarise_segments.R
diff --git a/R/transform_lr.R b/misc/transform_lr.R
similarity index 100%
rename from R/transform_lr.R
rename to misc/transform_lr.R