Clean up wilcox; minor bug fixes

mvfki · Oct 30, 2023 · 8a8e4bf · 8a8e4bf
1 parent c9a584e
commit 8a8e4bf
Show file tree

Hide file tree

Showing 11 changed files with 280 additions and 637 deletions.
diff --git a/CMakeLists.txt b/CMakeLists.txt
@@ -13,7 +13,7 @@ set(CMAKE_CXX_FLAGS_RELEASE "")
 set(CMAKE_VERBOSE_MAKEFILE TRUE)
 
 set(srclist ModularityOptimizer.cpp RModularityOptimizer.cpp RcppExports.cpp
-data_processing.cpp fast_wilcox.cpp feature_mat.cpp quantile_norm.cpp snn.cpp)
+data_processing.cpp wilcoxon.cpp feature_mat.cpp quantile_norm.cpp snn.cpp)
 list(TRANSFORM srclist PREPEND "${PROJECT_SOURCE_DIR}/src/")
 add_library(rliger2 SHARED ${srclist})
 set_target_properties(rliger2 PROPERTIES PREFIX "")

diff --git a/R/RcppExports.R b/R/RcppExports.R
@@ -45,50 +45,6 @@ sample_cpp <- function(x, size) {
     .Call(`_rliger2_sample_cpp`, x, size)
 }
 
-cpp_sumGroups_dgc <- function(x, p, i, ncol, groups, ngroups) {
-    .Call(`_rliger2_cpp_sumGroups_dgc`, x, p, i, ncol, groups, ngroups)
-}
-
-cpp_sumGroups_dgc_T <- function(x, p, i, ncol, nrow, groups, ngroups) {
-    .Call(`_rliger2_cpp_sumGroups_dgc_T`, x, p, i, ncol, nrow, groups, ngroups)
-}
-
-cpp_sumGroups_dense <- function(X, groups, ngroups) {
-    .Call(`_rliger2_cpp_sumGroups_dense`, X, groups, ngroups)
-}
-
-cpp_sumGroups_dense_T <- function(X, groups, ngroups) {
-    .Call(`_rliger2_cpp_sumGroups_dense_T`, X, groups, ngroups)
-}
-
-cpp_nnzeroGroups_dense <- function(X, groups, ngroups) {
-    .Call(`_rliger2_cpp_nnzeroGroups_dense`, X, groups, ngroups)
-}
-
-cpp_nnzeroGroups_dense_T <- function(X, groups, ngroups) {
-    .Call(`_rliger2_cpp_nnzeroGroups_dense_T`, X, groups, ngroups)
-}
-
-cpp_nnzeroGroups_dgc <- function(p, i, ncol, groups, ngroups) {
-    .Call(`_rliger2_cpp_nnzeroGroups_dgc`, p, i, ncol, groups, ngroups)
-}
-
-cpp_in_place_rank_mean <- function(v_temp, idx_begin, idx_end) {
-    .Call(`_rliger2_cpp_in_place_rank_mean`, v_temp, idx_begin, idx_end)
-}
-
-cpp_rank_matrix_dgc <- function(x, p, nrow, ncol) {
-    .Call(`_rliger2_cpp_rank_matrix_dgc`, x, p, nrow, ncol)
-}
-
-cpp_rank_matrix_dense <- function(X) {
-    .Call(`_rliger2_cpp_rank_matrix_dense`, X)
-}
-
-cpp_nnzeroGroups_dgc_T <- function(p, i, ncol, nrow, groups, ngroups) {
-    .Call(`_rliger2_cpp_nnzeroGroups_dgc_T`, p, i, ncol, nrow, groups, ngroups)
-}
-
 #' Fast calculation of feature count matrix
 #'
 #' @param bedmat A feature count list generated by bedmap
@@ -127,3 +83,23 @@ DirectSNNToFile <- function(nn_ranked, prune, display_progress, filename) {
     .Call(`_rliger2_DirectSNNToFile`, nn_ranked, prune, display_progress, filename)
 }
 
+cpp_rank_matrix_dgc <- function(x, p, nrow, ncol) {
+    .Call(`_rliger2_cpp_rank_matrix_dgc`, x, p, nrow, ncol)
+}
+
+rowAggregateSum_sparse <- function(X, groups, ngroups) {
+    .Call(`_rliger2_rowAggregateSum_sparse`, X, groups, ngroups)
+}
+
+colAggregateSum_sparse <- function(X, groups, ngroups) {
+    .Call(`_rliger2_colAggregateSum_sparse`, X, groups, ngroups)
+}
+
+colNNZAggr_sparse <- function(X, groups, ngroups) {
+    .Call(`_rliger2_colNNZAggr_sparse`, X, groups, ngroups)
+}
+
+rowNNZAggr_sparse <- function(X, groups, ngroups) {
+    .Call(`_rliger2_rowNNZAggr_sparse`, X, groups, ngroups)
+}
+
diff --git a/R/embedding.R b/R/embedding.R
@@ -23,8 +23,8 @@
 #' \code{NULL} uses all factors.
 #' @param nDims Number of dimensions to reduce to. Default \code{2}.
 #' @param distance Character. Metric used to measure distance in the input
-#' space. Default \code{"euclidean"}, alternative options include:
-#' \code{"cosine"}, \code{"manhattan"} and \code{"hamming"}.
+#' space. Default \code{"cosine"}, alternative options include:
+#' \code{"euclidean"}, \code{"manhattan"} and \code{"hamming"}.
 #' @param nNeighbors Number of neighboring points used in local approximations
 #' of manifold structure. Default \code{10}.
 #' @param minDist Numeric. Controls how tightly the embedding is allowed
@@ -47,7 +47,7 @@ runUMAP <- function(
         useRaw = FALSE,
         useDims = NULL,
         nDims = 2,
-        distance = c("euclidean", "cosine", "manhattan", "hamming"),
+        distance = c("cosine", "euclidean", "manhattan", "hamming"),
         nNeighbors = 20,
         minDist = 0.1,
         dimredName = "UMAP",

diff --git a/R/preprocess.R b/R/preprocess.R
@@ -638,7 +638,7 @@ selectGenes.liger <- function(
                                   unshared.datasets = "useUnsharedDatasets",
                                   unshared.thresh = "unsharedThresh"),
                    defunct = c("tol", "do.plot", "cex.use"))
-    lig <- recordCommand(lig)
+    object <- recordCommand(object)
     datasetShared <- .checkUseDatasets(object, useDatasets)
     if (!is.null(useUnsharedDatasets))
         datasetUnshared <- .checkUseDatasets(object, useUnsharedDatasets)

diff --git a/R/visualization2.R b/R/visualization2.R
@@ -49,6 +49,9 @@ plotClusterDimRed <- function(
         useDimRed = NULL,
         ...) {
     useDimRed <- useDimRed %||% object@uns$defaultDimRed
+    if (is.null(useDimRed)) {
+        stop("No `useDimRed` given or default dimRed not set.")
+    }
     useCluster <- useCluster %||% object@uns$defaultCluster
     xVar <- paste0(useDimRed, ".1")
     yVar <- paste0(useDimRed, ".2")
@@ -64,6 +67,9 @@ plotDatasetDimRed <- function(
         useDimRed = NULL,
         ...) {
     useDimRed <- useDimRed %||% object@uns$defaultDimRed
+    if (is.null(useDimRed)) {
+        stop("No `useDimRed` given or default dimRed not set.")
+    }
     xVar <- paste0(useDimRed, ".1")
     yVar <- paste0(useDimRed, ".2")
     plotCellScatter(object, x = xVar, y = yVar, colorBy = "dataset",