updated default VAR values to be more meaningful

DESCRIPTION

@@ -12,6 +12,7 @@ Encoding: UTF-8
 LazyData: false
 Suggests:
     BiocStyle,
+    curl,
     knitr,
     testthat,
     rmarkdown

NAMESPACE

@@ -107,6 +107,7 @@ export(subsetSamples)
 export(subsetValues)
 export(swapNameVariable)
 export(transformSignal)
+export(translateSSVtoCT2)
 exportMethods("[")
 exportMethods("colToRowMatCols<-")
 exportMethods("rowRanges<-")

NEWS

@@ -1,3 +1,6 @@
+# v 0.1.49
+* changed all default VAR names
+
 # v 0.1.47
 * depends on bugfix in seqsetvis in 1.25.5
 

R/class_ChIPtsne.R

@@ -48,10 +48,10 @@
 ChIPtsne2 = function(
         rowToRowMat=matrix(0,0,0),
         colToRowMatCols=list(),
-        name_VAR = "sample",
-        position_VAR = "x",
-        value_VAR = "y",
-        region_VAR = "id",
+        name_VAR = "name",
+        position_VAR = "position",
+        value_VAR = "signal",
+        region_VAR = "region",
         fetch_config = FetchConfig.null(),
         ...)
 {

R/class_ChIPtsne_alt_constructors.R

@@ -2,10 +2,10 @@
 #'
 #' @param prof_dt Profile data.table, as returned from seqsetvis::ssvFetch* functions
 #' @param query_gr The query GRanges object used to fetch prof_dt.
-#' @param name_VAR Variable name that contains sample ids/names. Links prof_dt to meta_dt. Default is "sample".
-#' @param position_VAR Variable name that contains positional information in prof_dt. Default is "x".
-#' @param value_VAR Variable name that contains signal value information in prof_dt. Default is "y".
-#' @param region_VAR Variable name that contains region ID information in prof_dt. Default is "id".
+#' @param name_VAR Variable name that contains sample ids/names. Links prof_dt to meta_dt. Default is "name".
+#' @param position_VAR Variable name that contains positional information in prof_dt. Default is "position".
+#' @param value_VAR Variable name that contains signal value information in prof_dt. Default is "signal".
+#' @param region_VAR Variable name that contains region ID information in prof_dt. Default is "region".
 #' @param sample_metadata Metadata for entries in prof_dt's name_VAR, must include name_VAR
 #' @param region_metadata Metadata to append to rowRanges, mcols of query_gr will also be used.
 #' @param auto_sample_metadata If true, additional attributes in prof_dt will used for metadata (minus certain region related attributes such as seqnames, start, end, etc.)
@@ -30,10 +30,10 @@ ChIPtsne2.from_tidy = function(prof_dt,
                                query_gr,
                                sample_metadata = NULL,
                                region_metadata = NULL,
-                               name_VAR = "sample",
-                               position_VAR = "x",
-                               value_VAR = "y",
-                               region_VAR = "id",
+                               name_VAR = "name",
+                               position_VAR = "position",
+                               value_VAR = "signal",
+                               region_VAR = "region",
                                auto_sample_metadata = TRUE,
                                obj_history = list(),
                                fetch_config = FetchConfig.null(),
@@ -135,12 +135,10 @@ ChIPtsne2.from_tidy = function(prof_dt,
                           "end",
                           "width",
                           "strand",
-                          "id",
-                          "y",
-                          "x",
-                          "cluster_id",
-                          position_VAR,
-                          value_VAR)
+                          region_VAR,
+                          value_VAR,
+                          position_VAR
+                          )
             sample_metadata = prof_dt %>%
                 # dplyr::select(all_of(c(name_VAR))) %>%
                 dplyr::select(!dplyr::any_of(c(drop_vars))) %>%
@@ -253,7 +251,7 @@ ChIPtsne2.from_FetchConfig = function(fetch_config,
 
     fetch_res = runFetchAtRegions(fetch_config, query_gr, use_cache = use_cache)
     prof_dt = fetch_res$prof_dt
-
+    prof_dt = translateSSVtoCT2(prof_dt)
     ct2 = ChIPtsne2.from_tidy(prof_dt = prof_dt,
                               name_VAR = name_VAR,
                               query_gr = query_gr,

R/class_ChIPtsne_no_rowRanges.R

@@ -49,10 +49,10 @@
 ChIPtsne2_no_rowRanges = function(
         rowToRowMat=matrix(0,0,0),
         colToRowMatCols=list(),
-        name_VAR = "sample",
-        position_VAR = "x",
-        value_VAR = "y",
-        region_VAR = "id",
+        name_VAR = "name",
+        position_VAR = "position",
+        value_VAR = "signal",
+        region_VAR = "region",
         fetch_config = FetchConfig.null(),
         ...)
 {

R/functions_examples.R

@@ -44,6 +44,7 @@ exampleQueryGR = function(){
 exampleProfDT = function(){
     CTCF_in_10a_profiles_dt = NULL
     utils::data(list = "CTCF_in_10a_profiles_dt", package = "seqsetvis", overwrite = TRUE, envir = environment())
+    CTCF_in_10a_profiles_dt = translateSSVtoCT2(CTCF_in_10a_profiles_dt)
     CTCF_in_10a_profiles_dt[]
 }
 
@@ -56,7 +57,8 @@ exampleProfDT = function(){
 #' exampleChIPtsne2()
 exampleChIPtsne2 = function(){
     query_gr = exampleQueryGR()
-    prof_dt = exampleProfDT()
+    prof_dt = exampleProfDT() %>%
+        translateSSVtoCT2
 
     ChIPtsne2.from_tidy(prof_dt, query_gr)
 }
@@ -73,9 +75,10 @@ exampleChIPtsne2.with_meta = function(){
     query_gr = exampleQueryGR()
     prof_dt = exampleProfDT()
     meta_dt = prof_dt %>%
-        dplyr::select(sample) %>%
+        translateSSVtoCT2 %>%
+        dplyr::select(name) %>%
         unique %>%
-        tidyr::separate(sample, c("cell", "mark"), sep = "_", remove = FALSE)
+        tidyr::separate(name, c("cell", "mark"), sep = "_", remove = FALSE)
 
     ChIPtsne2.from_tidy(prof_dt, query_gr, sample_metadata = meta_dt)
 }
@@ -169,3 +172,36 @@ exampleDataPaths = function(){
     url_dt[, narrowPeak_file := paste0(cell, "_", mark, ".narrowPeak")]
     url_dt[]
 }
+
+#' translateSSVtoCT2
+#'
+#' *seqsetvis* creates data.table with certain default variables that *chiptsne2* uses more user-friendly values for. This function checks for *seqsetvis* defaults and converts them to *chiptsne2*.
+#'
+#' @param prof_dt A data.table generated by *seqsetvis* from ssvFetchGRanges
+#'
+#' @return
+#' @export
+#'
+#' @examples
+#' prof_dt = exampleProfDT()
+#' prof_dt = translateSSVtoCT2(prof_dt)
+#' ct2 = ChIPtsne2.from_tidy(prof_dt, exampleQueryGR())
+translateSSVtoCT2 = function(prof_dt){
+    #convert to data.table if needed
+    if(!is(prof_dt, "data.table")){
+        prof_dt = data.table::as.data.table(prof_dt)
+    }
+    if("id" %in% colnames(prof_dt)){
+        data.table::setnames(prof_dt, "id", "region")
+    }
+    if("x" %in% colnames(prof_dt)){
+        data.table::setnames(prof_dt, "x", "position")
+    }
+    if("sample" %in% colnames(prof_dt)){
+        data.table::setnames(prof_dt, "sample", "name")
+    }
+    if("y" %in% colnames(prof_dt)){
+        data.table::setnames(prof_dt, "y", "signal")
+    }
+    prof_dt[]
+}

tests/testthat/test_ChIPtsne_class.R

@@ -6,25 +6,26 @@ library(testthat)
 query_gr = exampleQueryGR()
 prof_dt = exampleProfDT()
 
-metadata = prof_dt %>% dplyr::select(sample) %>% unique
-metadata = metadata %>% tidyr::separate(sample, c("cell", "mark"), sep = "_", remove = FALSE)
+metadata = prof_dt %>% dplyr::select(name) %>% unique
+metadata = metadata %>% tidyr::separate(name, c("cell", "mark"), sep = "_", remove = FALSE)
 
-map_dt = prof_dt %>% dplyr::select(sample, x) %>% unique %>%
-    dplyr::mutate(cn = paste(sample, x, sep = "_")) %>%
-    dplyr::mutate(nr = seq_along(x))
-map_list = split(map_dt$n, map_dt$sample)
+map_dt = prof_dt %>% dplyr::select(name, position) %>% unique %>%
+    dplyr::mutate(cn = paste(name, position, sep = "_")) %>%
+    dplyr::mutate(nr = seq_along(position))
 
-tmp_wide = tidyr::pivot_wider(prof_dt, names_from = c("sample", "x"), values_from = "y", id_cols = "id")
+map_list = split(map_dt$nr, map_dt$name)
+
+tmp_wide = tidyr::pivot_wider(prof_dt, names_from = c("name", "position"), values_from = "signal", id_cols = "region")
 prof_mat = as.matrix(tmp_wide[, -1])
-rownames(prof_mat) = tmp_wide$id
+rownames(prof_mat) = tmp_wide$region
 
 
 prof_max = prof_dt %>%
-    dplyr::group_by(id, sample) %>%
-    dplyr::summarise(y = max(y)) %>%
-    tidyr::pivot_wider(names_from = "sample", id_cols = "id", values_from = "y")
+    dplyr::group_by(region, name) %>%
+    dplyr::summarise(signal = max(signal)) %>%
+    tidyr::pivot_wider(names_from = "name", id_cols = "region", values_from = "signal")
 prof_max_mat = as.matrix(prof_max[, -1])
-rownames(prof_max_mat) = prof_max$id
+rownames(prof_max_mat) = prof_max$region
 
 ct = ChIPtsne2(assay = list(max = prof_max_mat[names(query_gr),]),
                rowRanges = query_gr,
@@ -34,14 +35,18 @@ ct = ChIPtsne2(assay = list(max = prof_max_mat[names(query_gr),]),
                metadata = list(time = date()))
 
 
-prof_dt = seqsetvis::ssvSignalClustering(prof_dt, nclust = 4)
-region_metadata = prof_dt %>% dplyr::select(id, cluster_id) %>% unique
+clust_dt = seqsetvis::ssvSignalClustering(prof_dt, nclust = 4, facet_ = "name", row_ = "region", column_ = "position", fill_ = "signal")
+# prof_dt = translateSSVtoCT2(prof_dt)
+# clust_dt = translateSSVtoCT2(clust_dt)
+
+region_metadata = clust_dt %>% dplyr::select(region, cluster_id) %>% unique
+
 
 ct2 = ChIPtsne2.from_tidy(prof_dt, query_gr, region_metadata = region_metadata)
 
 test_that("Constructors - valid", {
     expect_true(validObject(ct2))
-    expect_true(validObject(.ChIPtsne2())) # internal
+    expect_true(validObject(chiptsne2:::.ChIPtsne2())) # internal
     expect_true(validObject(ChIPtsne2())) # exported
 
     se = as(ct2, "SummarizedExperiment")
@@ -56,7 +61,7 @@ test_that("Constructors - invalid", {
 })
 
 test_that("Gettters", {
-    expect_identical(rowToRowMat(ct2), prof_mat[levels(prof_dt$id),])
+    expect_identical(rowToRowMat(ct2), prof_mat[unique(prof_dt$region),])
     expect_identical(rownames(rowToRowMat(ct2)), rownames(ct2))
     expect_identical(names(colToRowMatCols(ct2)), colnames(ct2))
 })
@@ -69,6 +74,7 @@ query_gr = exampleQueryGR()[1:10]
 suppressWarnings({
     ct2.cfg = ChIPtsne2.from_FetchConfig(fetch_config, query_gr)
 })
+
 test_that("Constructor FetchConfig", {
     expect_setequal(rownames(rowToRowMat(ct2.cfg)), names(query_gr))
     expect_equal(ncol(rowToRowMat(ct2.cfg)), 400)

tests/testthat/test_ChIPtsne_getsetters.R

@@ -5,25 +5,25 @@ library(testthat)
 ct2 = exampleChIPtsne2.with_meta()
 
 test_that("NameVariable", {
-    expect_equal(getNameVariable(ct2), "sample")
+    expect_equal(getNameVariable(ct2), "name")
     ct2.new = setNameVariable(ct2, "Name")
     expect_equal(getNameVariable(ct2.new), "Name")
 })
 
 test_that("RegionVariable", {
-    expect_equal(getRegionVariable(ct2), "id")
+    expect_equal(getRegionVariable(ct2), "region")
     ct2.new = setRegionVariable(ct2, "Region")
     expect_equal(getRegionVariable(ct2.new), "Region")
 })
 
 test_that("PositionVariable", {
-    expect_equal(getPositionVariable(ct2), "x")
+    expect_equal(getPositionVariable(ct2), "position")
     ct2.new = setPositionVariable(ct2, "Position")
     expect_equal(getPositionVariable(ct2.new), "Position")
 })
 
 test_that("ValueVariable", {
-    expect_equal(getValueVariable(ct2), "y")
+    expect_equal(getValueVariable(ct2), "signal")
     ct2.new = setValueVariable(ct2, "Value")
     expect_equal(getValueVariable(ct2.new), "Value")
 })
@@ -79,7 +79,7 @@ test_that("Switch name variable to smaller set", {
     ct2.10a = swapNameVariable(ct2.by_cell$MCF10A, new_VAR = "mark")
     expect_equal(getNameVariable(ct2.10a), "mark")
     expect_equal(getSampleMetaData(ct2.10a)$mark, factor("CTCF"))
-    expect_equal(getSampleMetaData(ct2.10a)$sample, "MCF10A_CTCF")
+    expect_equal(getSampleMetaData(ct2.10a)$name, "MCF10A_CTCF")
     expect_equal(colnames(rowToRowMat(ct2.10a))[1], "CTCF_-325")
 })
 
@@ -89,7 +89,7 @@ test_that("When names match for operator", {
     ct2.diff = ct2.by_cell$MCF10A - ct2.by_cell$MCF10AT1
     expect_equal(colnames(rowToRowMat(ct2.diff))[1], "CTCF_-325")
     ct2.diff1 = swapNameVariable(ct2.diff, new_VAR = "cell")
-    ct2.diff2 = swapNameVariable(ct2.diff, new_VAR = "sample")
+    ct2.diff2 = swapNameVariable(ct2.diff, new_VAR = "name")
     expect_equal(colnames(ct2.diff1), "MCF10A - MCF10AT1")
     expect_equal(colnames(rowToRowMat(ct2.diff1))[1], "MCF10A - MCF10AT1_-325")
     expect_equal(colnames(ct2.diff2), "MCF10A_CTCF - MCF10AT1_CTCF")
@@ -101,7 +101,7 @@ test_that("Switch name variable same set", {
     expect_equal(getNameVariable(ct2.name_cell), "cell")
     expect_equal(getSampleMetaData(ct2.name_cell)$mark, rep("CTCF", 3))
     expect_equal(colnames(rowToRowMat(ct2.name_cell))[1], "MCF10A_-325")
-    expect_equal(getSampleMetaData(ct2.name_cell)$sample, c("MCF10A_CTCF", "MCF10AT1_CTCF", "MCF10CA1_CTCF"))
+    expect_equal(getSampleMetaData(ct2.name_cell)$name, c("MCF10A_CTCF", "MCF10AT1_CTCF", "MCF10CA1_CTCF"))
 })