Merge pull request #666 from remlapmot/2025-12-fixes-2

TwoSampleMR 0.6.28

Author: remlapmot

DESCRIPTION

@@ -1,7 +1,7 @@
 Package: TwoSampleMR
 Title: Two Sample MR Functions and Interface to MRC Integrative
     Epidemiology Unit OpenGWAS Database
-Version: 0.6.27
+Version: 0.6.28
 Authors@R: c(
     person("Gibran", "Hemani", , "g.hemani@bristol.ac.uk", role = c("aut", "cre"),
            comment = c(ORCID = "0000-0003-0920-1055")),

NEWS.md

@@ -1,3 +1,10 @@
+# TwoSampleMR v0.6.28
+
+(Release date 2025-12-15)
+
+* Allowed the arguments for `clump_data()` to be specified from `read_exposure_data()` (thanks @al3xjwood).
+* Fixed two function calls in `mr_moe_single()`
+
 # TwoSampleMR v0.6.27
 
 (Release date 2025-12-15)

R/moe.R

@@ -174,7 +174,7 @@ get_rsq <- function(dat) {
 #' # been sorted in order from most likely to least likely to
 #' # be accurate, based on MOE prediction
 #' r[[1]]$estimates
-#'}
+#' }
 mr_moe <- function(res, rf) {
   if (!requireNamespace("randomForest", quietly = TRUE)) {
     stop(
@@ -210,8 +210,8 @@ mr_moe_single <- function(res, rf) {
     )
     return(d)
   }) %>%
-    bind_rows %>%
-    arrange(desc(MOE))
+    dplyr::bind_rows() %>%
+    dplyr::arrange(desc(MOE))
   if ("MOE" %in% names(res$estimates)) {
     message("Overwriting previous MOE estimate")
     res$estimates <- subset(res$estimates, select = -c(MOE, method2))

R/read_data.R

@@ -104,7 +104,7 @@ read_outcome_data <- function(
 #' @param log_pval The p-value is -log10(P). The default is `FALSE`.
 #' @param chr_col Optional column name for chromosome. Default is `"chr"`.
 #' @param pos_col Optional column name for genetic position Default is `"pos"`.
-#'
+#' @inheritParams clump_data
 #' @export
 #' @return data frame
 read_exposure_data <- function(
@@ -128,7 +128,13 @@ read_exposure_data <- function(
   min_pval = 1e-200,
   log_pval = FALSE,
   chr_col = "chr",
-  pos_col = "pos"
+  pos_col = "pos",
+  clump_kb = 10000,
+  clump_r2 = 0.001,
+  clump_p1 = 1,
+  pop = "EUR",
+  bfile = NULL,
+  plink_bin = NULL
 ) {
   exposure_dat <- data.table::fread(filename, header = TRUE, sep = sep)
   exposure_dat <- format_data(
@@ -156,7 +162,15 @@ read_exposure_data <- function(
   )
   exposure_dat$data_source.exposure <- "textfile"
   if (clump) {
-    exposure_dat <- clump_data(exposure_dat)
+    exposure_dat <- clump_data(
+      exposure_dat,
+      clump_kb = clump_kb,
+      clump_r2 = clump_r2,
+      clump_p1 = clump_p1,
+      pop = pop,
+      bfile = bfile,
+      plink_bin = plink_bin
+    )
   }
   return(exposure_dat)
 }
@@ -593,7 +607,7 @@ check_units <- function(x, id, col) {
 #' data(gwas_catalog)
 #' bmi <- subset(gwas_catalog, Phenotype=="Body mass index" & Year==2010 & grepl("kg", Units))
 #' bmi <- format_data(bmi)
-#'}
+#' }
 format_gwas_catalog <- function(gwas_catalog_subset, type = "exposure") {
   message("This function is now deprecated and has been replaced by 'format_data'.")
   return(NULL)

man/read_exposure_data.Rd

@@ -704,14 +704,14 @@ MR-MoE operates by taking a set of harmonised data, inferring some characteristi
 
 In order to run the analysis you must download an RData object that contains the trained random forests that are used to predict the efficacy of each method. This can be downloaded from here:
 
-[dropbox.com/s/5la7y38od95swcf](dropbox.com/s/5la7y38od95swcf)
+<https://www.dropbox.com/s/5la7y38od95swcf>
 
 **Caution: this is a large file (approx 167Mb)**
 
 Once downloaded, read in the object and use the `mr_moe()` function to perform the analysis. An example is shown here, estimating the causal effect of BMI on coronary heart disease:
 
 ```{r eval=FALSE}
-# Extact instruments for BMI
+# Extract instruments for BMI
 exposure_dat <- extract_instruments("ieu-a-2")
 
 # Get corresponding effects for CHD
@@ -728,6 +728,11 @@ res_all <- mr_wrapper(dat)
 
 # MR-MoE - predict the performance of each method
 res_moe <- mr_moe(res_all, rf)
+
+# Now you can view the estimates, and see that they have
+# been sorted in order from most likely to least likely to
+# be accurate, based on MOE prediction
+res_moe[[1]]$estimates
 ```
 
 The function does the following:
@@ -754,7 +759,7 @@ The TwoSampleMR package also provides the following functions for managing or ed
 
 ### Split outcome names
 
-The outcome column in the output of mr() combines the original outcome name with the outcome trait ID.
+The outcome column in the output of `mr()` combines the original outcome name with the outcome trait ID.
 
 ```{r}
 head(res)