diff --git a/CRAN-RELEASE b/CRAN-RELEASE
index b191737a..1e217160 100644
--- a/CRAN-RELEASE
+++ b/CRAN-RELEASE
@@ -1,2 +1,2 @@
-This package was submitted to CRAN on 2020-06-17.
-Once it is accepted, delete this file and tag the release (commit 7a1f5431a0).
+This package was submitted to CRAN on 2020-08-25.
+Once it is accepted, delete this file and tag the release (commit f750974add).
diff --git a/NAMESPACE b/NAMESPACE
index 906cf7c9..0e80236b 100644
--- a/NAMESPACE
+++ b/NAMESPACE
@@ -81,6 +81,7 @@ export(sigprofiler_extract)
export(sigprofiler_import)
export(sym)
export(syms)
+export(transform_seg_table)
export(use_color_style)
exportClasses(CopyNumber)
exportClasses(MAF)
diff --git a/NEWS.md b/NEWS.md
index 88447aca..37201e1f 100644
--- a/NEWS.md
+++ b/NEWS.md
@@ -1,5 +1,6 @@
# sigminer 1.0.12
+- Added `transform_seg_table()`.
- Added `show_cn_group_profile()`.
- Added `show_cn_freq_circos()`.
- `sig_orders` option in `show_sig_profile()` function now can select and order signatures to plot.
diff --git a/R/show_cn_circos.R b/R/show_cn_circos.R
index cc5d9733..68fd4a09 100644
--- a/R/show_cn_circos.R
+++ b/R/show_cn_circos.R
@@ -38,7 +38,7 @@
show_cn_circos <- function(data, samples = NULL,
show_title = TRUE,
chrs = paste0("chr", 1:22),
- genome_build = c("hg19", "hg38"),
+ genome_build = c("hg19", "hg38", "mm10"),
col = NULL,
side = "inside",
...) {
diff --git a/R/show_cn_freq_circos.R b/R/show_cn_freq_circos.R
index d2032fa5..b2e9bb42 100644
--- a/R/show_cn_freq_circos.R
+++ b/R/show_cn_freq_circos.R
@@ -34,7 +34,7 @@ show_cn_freq_circos <- function(data,
resolution_factor = 1L,
title = c("AMP", "DEL"),
chrs = paste0("chr", 1:22),
- genome_build = c("hg19", "hg38"),
+ genome_build = c("hg19", "hg38", "mm10"),
cols = NULL,
plot_ideogram = TRUE,
track_height = 0.5,
diff --git a/R/show_cn_group_profile.R b/R/show_cn_group_profile.R
index 8459157b..1c572135 100644
--- a/R/show_cn_group_profile.R
+++ b/R/show_cn_group_profile.R
@@ -44,7 +44,7 @@ show_cn_group_profile <- function(data,
fill_area = TRUE,
cols = NULL,
chrs = paste0("chr", c(1:22, "X")),
- genome_build = c("hg19", "hg38"),
+ genome_build = c("hg19", "hg38", "mm10"),
cutoff = 2L,
resolution_factor = 1L,
force_y_limit = TRUE,
diff --git a/R/show_cn_profile.R b/R/show_cn_profile.R
index 482160fb..bb4bc1f6 100644
--- a/R/show_cn_profile.R
+++ b/R/show_cn_profile.R
@@ -31,7 +31,7 @@
#' expect_s3_class(p, "ggplot")
show_cn_profile <- function(data, samples = NULL, show_n = NULL, show_title = FALSE,
chrs = paste0("chr", 1:22),
- genome_build = c("hg19", "hg38"),
+ genome_build = c("hg19", "hg38", "mm10"),
nrow = NULL, ncol = NULL, return_plotlist = FALSE) {
stopifnot(is.data.frame(data) | inherits(data, "CopyNumber"))
if (is.data.frame(data)) {
diff --git a/R/transform_seg_table.R b/R/transform_seg_table.R
new file mode 100644
index 00000000..b392badc
--- /dev/null
+++ b/R/transform_seg_table.R
@@ -0,0 +1,111 @@
+#' Transform Copy Number Table
+#'
+#' @inheritParams get_cn_freq_table
+#' @inheritParams tidyr::pivot_wider
+#' @param ref_type annotation data type used for constructing matrix.
+#'
+#' @return a `data.table`.
+#' @export
+#'
+#' @examples
+#' load(system.file("extdata", "toy_copynumber.RData",
+#' package = "sigminer", mustWork = TRUE
+#' ))
+#' # Compute the mean segVal in each cytoband
+#' x <- transform_seg_table(cn, resolution_factor = 1)
+#' x
+#' # Compute the mean segVal in each half-cytoband
+#' x2 <- transform_seg_table(cn, resolution_factor = 2)
+#' x2
+#' @testexamples
+#' expect_is(x, "data.table")
+#' expect_is(x2, "data.table")
+transform_seg_table <- function(data,
+ genome_build = c("hg19", "hg38", "mm10"),
+ ref_type = c("cytoband", "gene"),
+ values_fill = NA,
+ values_fn = function(x, ...) {
+ round(mean(x, ...))},
+ resolution_factor = 1L) {
+
+ stopifnot(is.data.frame(data) | inherits(data, "CopyNumber"))
+ if (is.data.frame(data)) {
+ nc_cols <- c("chromosome", "start", "end", "segVal", "sample")
+ if (!all(nc_cols %in% colnames(data))) {
+ stop("Invalid input, it must contain columns: ", paste(nc_cols, collapse = " "))
+ }
+ }
+
+ genome_build <- match.arg(genome_build)
+ if (inherits(data, "CopyNumber")) {
+ genome_build <- data@genome_build
+ data <- data@data
+ } else {
+ data <- data.table::as.data.table(data)
+ }
+
+ ref_type <- match.arg(ref_type)
+
+ #data$sample <- factor(data$sample, levels = unique(data$sample))
+ data$chromosome <- ifelse(startsWith(data$chromosome, prefix = "chr"),
+ data$chromosome,
+ paste0("chr", data$chromosome))
+
+ if (ref_type == "cytoband") {
+ annot <- get_genome_annotation(
+ data_type = "cytobands",
+ genome_build = genome_build
+ )
+ annot$start <- annot$start + 1L
+ } else {
+ if (genome_build == "mm10") {
+ # Not support for now
+ annot_file <- system.file("extdata", "mouse_mm10_gene_info.rds",
+ package = "sigminer", mustWork = TRUE)
+ } else {
+ annot_file <- system.file("extdata", paste0("human_", genome_build, "_gene_info.rds"),
+ package = "sigminer", mustWork = TRUE)
+ }
+
+ annot <- readRDS(annot_file)
+ annot <- annot[, c("chrom", "start", "end", "gene_name", "gene_type")]
+ colnames(annot)[4] <- "band"
+ }
+
+
+ data.table::setDT(annot)
+ ## Control the resolution
+ if (resolution_factor > 1) {
+ f <- function(x, y, n, chrom, band) {
+ helper_create_chunks(x, y,
+ n = n,
+ chrom = chrom,
+ band = paste(band, seq_len(n), sep = "-chunk-")
+ )
+ }
+ annot <- purrr::pmap_df(
+ data.frame(
+ x = annot$start,
+ y = annot$end,
+ n = resolution_factor,
+ chrom = annot$chrom,
+ band = annot$band
+ ),
+ .f = f
+ ) %>%
+ data.table::as.data.table() %>%
+ data.table::setcolorder(c("chrom", "start", "end", "band"))
+ }
+ data.table::setkey(annot, chrom, start, end)
+ merge_dt <- data.table::foverlaps(data, annot,
+ by.x = c("chromosome", "start", "end")
+ )
+ merge_dt <- merge_dt %>%
+ dplyr::as_tibble() %>%
+ dplyr::select(-c("i.start", "i.end")) %>%
+ na.omit() %>%
+ tidyr::pivot_wider(names_from = "sample", values_from = "segVal",
+ values_fill = values_fill, values_fn = values_fn)
+ colnames(merge_dt)[4] <- "label"
+ merge_dt %>% data.table::as.data.table()
+}
diff --git a/_pkgdown.yml b/_pkgdown.yml
index aa3b7c52..1499d2f8 100644
--- a/_pkgdown.yml
+++ b/_pkgdown.yml
@@ -112,6 +112,7 @@ reference:
- title: Copy number analysis and visualization
desc: Functions for analyzing copy number data and visualization.
contents:
+ - transform_seg_table
- get_cn_ploidy
- scoring
- show_cn_profile
diff --git a/data-raw/mouse_genome.R b/data-raw/mouse_genome.R
index fd0d1906..2db49ed8 100644
--- a/data-raw/mouse_genome.R
+++ b/data-raw/mouse_genome.R
@@ -101,4 +101,19 @@ mm10$width <- NULL
transcript.mm10 <- mm10
usethis::use_data(transcript.mm10, overwrite = TRUE)
-## Currently, I don't use gene location data, so don't generate it for now.
+# Gene --------------------------------------------------------------------
+
+## mm10 gene
+gtf_mm10 <- data.table::fread("data-raw/mm10.annotation.gtf.gz", skip = 5, sep = "\t", header = FALSE)
+
+gtf_mm10[, `:=`(
+ gene_name = extract_col(V9, "gene_name"),
+ gene_id = extract_col(V9, "gene_id"),
+ gene_type = extract_col(V9, "gene_type")
+)]
+
+gene_mm10 <- gtf_mm10[V3 == "gene", .(V1, V4, V5, V7, gene_name, gene_id, gene_type)]
+colnames(gene_mm10)[1:4] <- c("chrom", "start", "end", "strand")
+
+## Save to extdata
+saveRDS(gene_mm10, file = "inst/extdata/mouse_mm10_gene_info.rds")
diff --git a/docs/reference/cosine.html b/docs/reference/cosine.html
index 870159fe..d76355db 100644
--- a/docs/reference/cosine.html
+++ b/docs/reference/cosine.html
@@ -151,8 +151,8 @@ Value
a numeric value or matrix.
Examples
- x <- c( 1, 1, 1, 0, 0, 0, 0, 0, 0, 0, 0, 0 )
-y <- c( 0, 0, 1, 1, 1, 1, 1, 0, 1, 0, 0, 0 )
+ x <- c(1, 1, 1, 0, 0, 0, 0, 0, 0, 0, 0, 0)
+y <- c(0, 0, 1, 1, 1, 1, 1, 0, 1, 0, 0, 0)
z1 <- cosine(x, y)
z1
z2 <- cosine(matrix(x), matrix(y))
diff --git a/docs/reference/get_cn_freq_table.html b/docs/reference/get_cn_freq_table.html
index 78ff7c4c..3d0130a3 100644
--- a/docs/reference/get_cn_freq_table.html
+++ b/docs/reference/get_cn_freq_table.html
@@ -131,7 +131,12 @@ Get CNV Frequency Table
Get CNV Frequency Table
- get_cn_freq_table(data, genome_build = "hg19", cutoff = 2L)
+ get_cn_freq_table(
+ data,
+ genome_build = "hg19",
+ cutoff = 2L,
+ resolution_factor = 1L
+)
Arguments
@@ -150,6 +155,12 @@ Arg
copy number value cutoff for splitting data into AMP and DEL.
The values equal to cutoff are discarded. Default is 2, you can also set
a length-2 vector, e.g. c(2, 2). |
+
+
+ | resolution_factor |
+ an integer to control the resolution.
+When it is 1 (default), compute frequency in each cytoband.
+When it is 2, use compute frequency in each half cytoband. |
diff --git a/docs/reference/index.html b/docs/reference/index.html
index b23b279a..ca702fcd 100644
--- a/docs/reference/index.html
+++ b/docs/reference/index.html
@@ -659,6 +659,12 @@
+ |
+ transform_seg_table()
+ |
+ Transform Copy Number Table |
+
+
|
get_cn_ploidy()
|
diff --git a/docs/reference/show_cn_circos.html b/docs/reference/show_cn_circos.html
index 3e6edade..03473bec 100644
--- a/docs/reference/show_cn_circos.html
+++ b/docs/reference/show_cn_circos.html
@@ -136,7 +136,7 @@ Show Copy Number Profile in Circos
samples = NULL,
show_title = TRUE,
chrs = paste0("chr", 1:22),
- genome_build = c("hg19", "hg38"),
+ genome_build = c("hg19", "hg38", "mm10"),
col = NULL,
side = "inside",
...
diff --git a/docs/reference/show_cn_freq_circos.html b/docs/reference/show_cn_freq_circos.html
index e85858cf..287f8375 100644
--- a/docs/reference/show_cn_freq_circos.html
+++ b/docs/reference/show_cn_freq_circos.html
@@ -135,9 +135,10 @@ Show Copy Number Variation Frequency Profile with Circos
data,
groups = NULL,
cutoff = 2L,
+ resolution_factor = 1L,
title = c("AMP", "DEL"),
chrs = paste0("chr", 1:22),
- genome_build = c("hg19", "hg38"),
+ genome_build = c("hg19", "hg38", "mm10"),
cols = NULL,
plot_ideogram = TRUE,
track_height = 0.5,
@@ -162,6 +163,12 @@ Arg
copy number value cutoff for splitting data into AMP and DEL.
The values equal to cutoff are discarded. Default is 2, you can also set
a length-2 vector, e.g. c(2, 2). |
+
+
+ | resolution_factor |
+ an integer to control the resolution.
+When it is 1 (default), compute frequency in each cytoband.
+When it is 2, use compute frequency in each half cytoband. |
| title |
diff --git a/docs/reference/show_cn_group_profile.html b/docs/reference/show_cn_group_profile.html
index 9b1539e5..e2ffaff8 100644
--- a/docs/reference/show_cn_group_profile.html
+++ b/docs/reference/show_cn_group_profile.html
@@ -137,8 +137,9 @@ Show Summary Copy Number Profile for Sample Groups
fill_area = TRUE,
cols = NULL,
chrs = paste0("chr", c(1:22, "X")),
- genome_build = c("hg19", "hg38"),
+ genome_build = c("hg19", "hg38", "mm10"),
cutoff = 2L,
+ resolution_factor = 1L,
force_y_limit = TRUE,
nrow = NULL,
ncol = NULL,
@@ -178,6 +179,12 @@ Arg
copy number value cutoff for splitting data into AMP and DEL.
The values equal to cutoff are discarded. Default is 2, you can also set
a length-2 vector, e.g. c(2, 2). |
+
+
+ | resolution_factor |
+ an integer to control the resolution.
+When it is 1 (default), compute frequency in each cytoband.
+When it is 2, use compute frequency in each half cytoband. |
| force_y_limit |
@@ -212,16 +219,20 @@ Examp
ss <- unique(cn@data$sample)
p2 <- show_cn_group_profile(cn, groups = list(a = ss[1:5], b = ss[6:10]))
p2
-p3 <- show_cn_group_profile(cn, groups = list(g1 = ss[1:5], g2 = ss[6:10]),
- force_y_limit = c(-1, 1), nrow = 2)
+p3 <- show_cn_group_profile(cn,
+ groups = list(g1 = ss[1:5], g2 = ss[6:10]),
+ force_y_limit = c(-1, 1), nrow = 2
+)
p3
## Set custom cutoff for custom data
data <- cn@data
data$segVal <- data$segVal - 2L
-p4 <- show_cn_group_profile(data, groups = list(g1 = ss[1:5], g2 = ss[6:10]),
- force_y_limit = c(-1, 1), nrow = 2,
- cutoff = c(0, 0))
+p4 <- show_cn_group_profile(data,
+ groups = list(g1 = ss[1:5], g2 = ss[6:10]),
+ force_y_limit = c(-1, 1), nrow = 2,
+ cutoff = c(0, 0)
+)
p4