Drop duplicate cell types

.github/ISSUE_TEMPLATE/bug_report.yml

@@ -43,10 +43,14 @@ body:
         If that's not possible, please at least include a screenshot of your data and other relevant details.
         
         See [here](https://www.r-bloggers.com/2020/10/how-to-make-a-reprex/) for an intro to making a reproducible example (i.e. reprex) and why they're important! 
+        
+        Surrounding your code in triple backticks will render it nicely on GitHub.
       placeholder: |
         # For example:
+        ```R
         df <- data.frame(list(a=rep(2,3), b=rep(1,3)))
         df["bTYPO"]
+        ```
     validations:
       required: true
 
@@ -58,7 +62,9 @@ body:
         Copy and paste the R console output. 
       placeholder: |
         # For example:
+        ```R
         Error in `[.data.frame`(df, "bTYPO") : undefined columns selected
+        ```
     validations:
       required: true
 
@@ -82,6 +88,7 @@ body:
       description: |
         Copy and paste the output of the `sessionInfo() R function.
       placeholder: |
+        ```R
         R version 4.4.1 (2024-06-14)  
         Platform: aarch64-apple-darwin20  
         Running under: macOS 15.0  
@@ -101,4 +108,5 @@ body:
         ...
         ...
         ...
+        ```
  

DESCRIPTION

@@ -1,7 +1,7 @@
 Package: MAGMA.Celltyping
 Type: Package
 Title: Find Causal Cell-Types Underlying Complex Trait Genetics
-Version: 2.0.13
+Version: 2.0.14
 Authors@R:
     c(person(given = "Brian",
            family = "Schilder",

NEWS.md

@@ -1,3 +1,14 @@
+# MAGMA.Celltyping 2.0.14
+
+## New features
+
+* Make sure `get_example_gwas` caches data.
+
+## Bug fixes
+
+* `create_gene_covar_file`: Add check to drop duplicate cell types (after ignoring case)
+* `load_magma_results_file`: Update message to include scenario where cell types dropped due to be duplicated.
+
 # MAGMA.Celltyping 2.0.13
 
 ## New features

R/calculate_celltype_associations.r

@@ -104,7 +104,6 @@ calculate_celltype_associations <- function(ctd,
         path <- sprintf("%s.%s.gsa.out",
                         sumstatsPrefix2,
                         analysis_name)
-
         path <- get_actual_path(path)
         #### Use existing results, or compute new ones ####
         if ((!file.exists(path)) || isTRUE(force_new)) {

R/check_quantiles.R

@@ -16,7 +16,8 @@ check_quantiles <- function(ctd,
                               function(lvl,
                                        .matrix_name=matrix_name,
                                        .metric=metric[1]){
-        messager("Checking CTD: level ",lvl,v=TRUE, parallel = TRUE)
+        messager("Checking CTD ",matrix_name," matrix: ",
+                 "level ",lvl,v=TRUE, parallel = TRUE)
         ctd_1lvl <- ctd[[lvl]]
         if(.metric=="n"){
             col_unique <- apply(ctd_1lvl[[.matrix_name]], 2,

R/create_gene_covar_file.r

@@ -86,9 +86,21 @@ create_gene_covar_file <- function(genesOutFile,
     dup_genes <- duplicated(quantDat2$entrez)
     if(sum(dup_genes, na.rm = TRUE)>0){
         messager("Removing", sum(dup_genes, na.rm = TRUE),
-                 "genes with duplicate entrez IDs.",v=verbose)
+                 "gene(s) with duplicate entrez IDs.",v=verbose)
         quantDat2 <- quantDat2[!dup_genes,]
     }
+    #### Ensure there are no duplicate cell types when ignoring case (which MAGMA does) ####
+    dup_ct <- duplicated(toupper(colnames(quantDat2)))
+    if(sum(dup_ct)>0){
+      messager("Removing", sum(dup_ct, na.rm = TRUE),
+               "cell type(s) with duplicate cell types (when ignoring upper/lowercase):",
+               paste(shQuote(colnames(quantDat2)[dup_ct]), collapse = ","),
+               "Please consider regenerating your CTD object with",
+               "unique cell type names (when ignoring upper/lowercase)",
+               "for more accurate results.",
+               v=verbose)
+      quantDat2 <- quantDat2[,!dup_ct]
+    }
     #### Write genes covar file to disk ####
     geneCovarFile <- tempfile()
     utils::write.table(

R/get_example_gwas.R

@@ -61,7 +61,8 @@ get_example_gwas <- function(trait = c(
                                  "prospective_memory"
                              ),
                              munged = TRUE,
-                             storage_dir = tempdir(),
+                             storage_dir = tools::R_user_dir("MAGMA.Celltyping",
+                                                             which="cache"),
                              timeout = 60 * 5,
                              verbose = TRUE) {
     trait <- tolower(trait)[1]

R/load_magma_results_file.r

@@ -93,7 +93,8 @@ load_magma_results_file <- function(path,
                 messager(numCTinCTD,
                 "celltypes in ctd but",numCTinRes,"in results file.",
                 "Some celltypes may have been dropped due if the variance",
-                "being too low (i.e. set contains only one gene used in analysis).")
+                "being too low (i.e. set contains only one gene used in analysis)",
+                "OR duplicate celltype names (after ignoring case).")
                 messager(
                     "<50% of celltypes missing.",
                     "Attemping to fix by removing missing cell-types:\n",

R/prepare_quantile_groups.r

@@ -46,7 +46,7 @@ prepare_quantile_groups <- function(ctd,
         ctd <- lapply(ctd, EWCE::bin_specificity_into_quantiles,
                       numberOfBins = numberOfBins, 
                       matrix_name = "specificity_quantiles") 
-    }
+    } 
     #### Compute specificity deciles ####
     ctd <- lapply(ctd, EWCE::bin_specificity_into_quantiles,
                   numberOfBins = 10,