feat: functions for process based features.

DESCRIPTION

@@ -1,6 +1,6 @@
 Package: dlptools
 Title: DLPTools: Handling DLP+ Data
-Version: 0.3.18
+Version: 0.3.20
 Authors@R: 
     person("Ben", "Furman", , "bfurman@bccrc.ca", role = c("aut", "cre"))
 Description: A collections of functions for basic manipulation and plotting of

NAMESPACE

@@ -24,11 +24,20 @@ export(chip_plot)
 export(compute_tip_sibling_distances)
 export(convert_dists_to_pairwise)
 export(convert_long_reads_to_wide)
+export(count_oscillations)
+export(create_chrom_arm_intervals)
+export(create_chrom_window_intervals)
 export(create_expected_bins)
 export(cust_mode)
 export(expand_length_to_bins)
+export(extract_bp_per_arm)
+export(extract_bp_per_window)
+export(extract_breakpoints)
+export(extract_changepoint)
+export(extract_oscillations)
 export(extract_ploidy_cn_feature)
 export(extract_segment_position_feature)
+export(extract_segment_sizes)
 export(extract_sigminer_wang_features)
 export(extract_wu_features)
 export(factor_column_mixedsort)

R/chromosome_layouts.R

@@ -21,7 +21,8 @@
     ) |>
     dplyr::mutate(
       chr = stringr::str_replace(chr, "chr", "")
-    )
+    ) |>
+    dplyr::select(-misc)
 
   return(chrom_info)
 }
@@ -33,7 +34,7 @@
 #' @param cn_df dataframe of copy number information
 #' @param version string. hg19 (default) or hg38
 #' @param chrom_col string. name of column with chromosome information.
 add_chromosome_length <- function(cn_df, version = c("hg19", "hg38"), chrom_col = "chr") {
   chrom_info <- load_chrom_info_file(version = version)
 
   if (is_chr_used_in_chroms(cn_df[[chrom_col]])) {
@@ -95,7 +96,7 @@
 #' @return input table, but with a boolean 'within_centro' column added (and
 #' potentially other centromere information columns, if needed)
 #' @export
 mark_bins_overlapping_centromeres <- function(
     reads_df,
     padding = 0,
     bin_start_col = "start",
@@ -221,11 +222,11 @@
 
   # col names by inspection and checking against here by looking up gap table
   # data format description
   # https://genome.ucsc.edu/cgi-bin/hgTables?hgsid=2018561924_ZAtdZC9CFEJw8BKiYeUjd2ImvhS7&clade=mammal&org=Human&db=hg19&hgta_group=allTables&hgta_track=hg19&hgta_table=gap&hgta_regionType=genome&position=chr7%3A155%2C592%2C223-155%2C605%2C565&hgta_outputType=primaryTable&hgta_outFileName=
   telos <- vroom::vroom(
     telos_file,
     col_names = c(
       "bin", "chrom", "telostart", "teloend", "ix", "n", "size", "feat_type", "bridge"
     )
   ) |>
     dplyr::filter(feat_type == "telomere") |>
@@ -338,7 +339,7 @@
 #' @export
 mark_segs_chromosome_span <- function(
     segs_df,
     min_bound_distance = 5e5, # given scale of DLP, this should probably be 1 bin width, at least
     min_span_of_chrom = 0.9,
     min_span_of_arm = 0.9,
     version = c("hg19", "hg38"),
@@ -399,7 +400,7 @@
       seg_span_event = dplyr::case_when(
         spans_chrom ~ event_labels["whole"], # "whole-chrom",
         centro_bound & telo_bound & !spans_centro ~ event_labels["arm"],
         !telo_bound & !centro_bound & !spans_chrom & spans_arm ~ event_labels["arm"],
         telo_bound & !centro_bound & !spans_chrom ~ event_labels["telo"], # "telo-bound",
         centro_bound | spans_centro ~ event_labels["centro"], # "centro-bound"
         .default = event_labels["inter"]
@@ -437,3 +438,64 @@
 
   return(segs_df)
 }
+
+#' break a chromosome up into intervals of a defined window size
+#'
+#' @param window_size integer. The size of window to split the chromosome into.
+#' @param genome_version string. "hg19" (default) or "hg38"
+#' @return list. Named by chromosome, vectors of window starts.
+#' @export
+create_chrom_window_intervals <- function(
+    window_size,
+    genome_version = c("hg19", "hg38")) {
+  genome_version <- match.arg(genome_version)
+
+  chr_info <- suppressWarnings(
+    chr_info <- load_chrom_info_file(version = genome_version)
+  )
+  intervals <- purrr::map(chr_info$total_length, \(total_length) {
+    max_end <- total_length + window_size
+    intervals <- seq(1, max_end, window_size)
+    intervals
+  })
+  names(intervals) <- chr_info$chr
+  return(intervals)
+}
+
+#' create a list of intervals spanning chromosome arms
+#'
+#' Splits a chromosome at the middle of the centromere. Sets up intervals for
+#' splitting each chromosome arm.
+#'
+#' @param genome_version string. "hg19" (default) or "hg38"
+#' @return list. Named by chromosome, vectors of how to break a chromsome into
+#' intervals of arms.
+#' @export
+create_chrom_arm_intervals <- function(genome_version = c("hg19", "hg38")) {
+  chrom_layouts <- suppressWarnings(
+    load_ucsc_centromeres(version = genome_version)
+  )
+  chrom_lengths <- suppressWarnings(
+    load_chrom_info_file(version = genome_version)
+  )
+
+  chrom_info <- dplyr::left_join(
+    chrom_lengths,
+    chrom_layouts,
+    by = dplyr::join_by("chr" == "chrom")
+  )
+
+  intervals <- purrr::pmap(
+    dplyr::select(
+      chrom_info,
+      total_length, centro_start, centro_end
+    ),
+    \(total_length, centro_start, centro_end) {
+      middle_centro <- centro_start + round(((centro_end - centro_start) / 2))
+      c(1, middle_centro, total_length + 1)
+    }
+  )
+  names(intervals) <- chrom_info$chr
+
+  return(intervals)
+}