Merge pull request #1055 from jiajic/suite_dev

Suite dev

.Rbuildignore

@@ -2,6 +2,7 @@
 ^renv\.lock$
 ^Giotto\.Rproj$
 ^\.Rproj\.user$
+^.lintr
 
 ^LICENSE\.md$
 ^README\.Rmd$

DESCRIPTION

@@ -26,8 +26,6 @@ URL: https://drieslab.github.io/Giotto/, https://github.com/drieslab/Giotto
 BugReports: https://github.com/drieslab/Giotto/issues
 RoxygenNote: 7.3.2
 Depends:
-    base (>= 4.4.1),
-    utils (>= 4.4.1),
     R (>= 4.4.1),
     methods,
     GiottoClass (>= 0.3.3)

NEWS.md

@@ -1,11 +1,16 @@
-# Giotto 4.1.3
+# Giotto 4.1.3 (2024/10/27)
 
 ## New
 * Add `giottoToAnndataZarr()` to create a local anndata zarr folder and interact with the vitessceR package.
+* `reduceDims()` API function for dimension reductions
+* `runNMF()` implementation that works via RcppML
 
 ## Changes
+* `screePlot()` `name` param changed to `dim_reduction_name`
 * `runWNN()` and `runIntegratedUMAP()` arguments were updated to make the function flexible to handle any number of modalities.
 * update `jackstrawPlot()` to make more flexible and efficient. Changed default params for `scaling`, `centering`, and `feats_to_use` to match `runPCA()`
+* change warning when reduction "feats" is selected in `runtSNE()` to error to avoid accidentally wiping the `giotto` object.
+
 
 # Giotto 4.1.2
 
@@ -26,7 +31,11 @@
 * move `progressr` and `jsonlite` dependencies to GiottoUtils v0.1.12
 * remove `reshape2` dependency.
 
+## Bug fixes
+* `processGiotto()` can now skip adjust step by default
+
 ## New
+* `identifyTMAcores()` for assigning IDs to tissue microarray spatial data.
 * `labelTransfer()` for transferring labels between giotto objects or subsets thereof. Supercedes `doClusterProjection()`
 
 # Giotto 4.1.1

R/clustering.R

@@ -3228,11 +3228,11 @@ getDendrogramSplits <- function(gobject,
 
     if (show_dend == TRUE) {
         # plot dendrogram
-        graphics::plot(cordend)
+        plot(cordend)
 
         # add horizontal lines
         if (!is.null(h)) {
-            graphics::abline(h = h, col = h_color)
+            abline(h = h, col = h_color)
         }
     }
 

R/dimension_reduction.R

@@ -230,11 +230,11 @@ reduceDims <- function(
     )
 
     if (set_seed) {
-        withSeed(
+        gwith_seed(
+            seed = seed_number,
             {
                 pca_res <- do.call(BiocSingular::runPCA, pca_param)
             },
-            seed = seed_number
         )
     } else {
         pca_res <- do.call(BiocSingular::runPCA, pca_param)
@@ -1757,7 +1757,7 @@ jackstrawPlot <- function(gobject,
     # create a random subset if random_subset is not NULL
     if (!is.null(random_subset)) {
         if (set_seed) {
-            withSeed(seed = seed_number, {
+            gwith_seed(seed = seed_number, {
                 random_selection <- sort(sample(
                     seq_len(ncol(expr_values)), random_subset
                 ))
@@ -1795,7 +1795,7 @@ jackstrawPlot <- function(gobject,
         }
 
         if (set_seed) {
-            withSeed(seed = seed_number, {
+            gwith_seed(seed = seed_number, {
                 jtest <- .perm_pa(
                     dat = expr_values,
                     iter = iter,

R/globals.R

@@ -1,4 +1,4 @@
-utils::globalVariables(names = c(
+globalVariables(names = c(
     ":=", ".N", ".SD", ".", "cast",
     "%--%", ".inc", # igraph
     "python_leiden", "python_louvain", "python_spatial_genes",

R/image_registration.R

@@ -705,10 +705,10 @@ registerGiottoObjectListRvision <- function(gobject_list = gobject_list,
 
     ## 4. Compute transformations
     # Choose reference image
-    refImage <- unreg_images[[base::floor(length(unreg_images) / 2)]]
+    refImage <- unreg_images[[floor(length(unreg_images) / 2)]]
 
     # Compute ECC transforms
-    transfs <- base::vector(mode = "list", length = length(unreg_images))
+    transfs <- vector(mode = "list", length = length(unreg_images))
     for (i in seq_along(unreg_images)) {
         transfs[[i]] <- Rvision::findTransformECC(
             refImage, unreg_images[[i]],

R/normalize.R

@@ -199,7 +199,7 @@ normalizeGiotto <- function(
 .lib_norm_giotto <- function(mymatrix, scalefactor) {
     libsizes <- colSums_flex(mymatrix)
 
-    if (any(libsizes == 0)) {
+    if (0 %in% libsizes) {
         warning(wrap_txt("Total library size or counts for individual spat
                     units are 0.
                     This will likely result in normalization problems.