Merge pull request #489 from remlapmot/march-2024-suggestions

TwoSampleMR 0.5.11

DESCRIPTION

@@ -1,6 +1,6 @@
 Package: TwoSampleMR
 Title: Two Sample MR Functions and Interface to MR Base Database
-Version: 0.5.10
+Version: 0.5.11
 Authors@R: c(
     person("Gibran", "Hemani", , "g.hemani@bristol.ac.uk", role = c("aut", "cre"),
            comment = c(ORCID = "0000-0003-0920-1055")),
@@ -32,7 +32,7 @@ Imports:
     cowplot,
     data.table,
     dplyr,
-    ggplot2,
+    ggplot2 (>= 3.4.0),
     glmnet,
     gridExtra,
     gtable,

NEWS.md

@@ -1,3 +1,18 @@
+# TwoSampleMR v0.5.11
+
+(Release date: 2024-03-21)
+
+* In `mr_leaveoneout_plot()` and `mr_forest_plot()` amended `size` argument to `linewidth` as per ggplot2 version 3.4.0.
+* Add some datasets such that tests, continuous integration services, and creation of the vignettes don't rely on the availability of the OpenGWAS server.
+* Various improvements to helpfiles.
+
+# TwoSampleMR v0.5.10
+
+(Release date: 2024-02-20)
+
+* Added `bfile` and `plink_bin` arguments to `clump_data()`
+* Improvements to file reading and dataset formatting capabilities of `mv_extract_exposures_local()` to create the multiple exposure dataset
+
 # TwoSampleMR v0.5.9
 
 (Release date: 2024-02-01)
@@ -217,12 +232,6 @@ TwoSampleMR v0.4.17
 (Release date: 2018-12-03)
 
 * Added facility to harmonise indels
-
-TwoSampleMR v0.4.17
-==============
-
-(Release date: 2018-12-03)
-
 * Documentation and options added to multivariable MR
 
 TwoSampleMR v0.3.4

R/ld.R

@@ -12,14 +12,14 @@
 #' This function does put load on the OpenGWAS servers, which makes life more difficult for other users.
 #' We have implemented a method and made available the LD reference panels to perform clumping locally, see [ieugwasr::ld_clump()] and related vignettes for details.
 #'
-#' @param dat Output from [format_data()]. Must have a SNP name column (SNP), SNP chromosome column (chr_name), SNP position column (chrom_start). If id.exposure or pval.exposure not present they will be generated.
+#' @param dat Output from [format_data()]. Must have a SNP name column (`SNP`), SNP chromosome column (`chr_name`), SNP position column (`chrom_start`). If `id.exposure` or `pval.exposure` not present they will be generated.
 #' @param clump_kb Clumping window, default is `10000`.
 #' @param clump_r2 Clumping r2 cutoff. Note that this default value has recently changed from `0.01` to `0.001`.
 #' @param clump_p1 Clumping sig level for index SNPs, default is `1`.
 #' @param clump_p2 Clumping sig level for secondary SNPs, default is `1`.
-#' @param pop Super-population to use as reference panel. Default = "EUR". Options are EUR, SAS, EAS, AFR, AMR. 'legacy' also available - which is a previously used version of the EUR panel with a slightly different set of markers
-#' @param bfile If this is provided then will use the API. Default = ‘NULL’
-#' @param plink_bin If ‘NULL’ and ‘bfile’ is not ‘NULL’ then will detect packaged plink binary for specific OS. Otherwise specify path to plink binary. Default = ‘NULL’
+#' @param pop Super-population to use as reference panel. Default = `"EUR"`. Options are `"EUR"`, `"SAS"`, `"EAS"`, `"AFR"`, `"AMR"`. `'legacy'` also available - which is a previously used version of the EUR panel with a slightly different set of markers
+#' @param bfile If this is provided then will use the API. Default = `NULL`
+#' @param plink_bin If `NULL` and `bfile` is not `NULL` then will detect packaged plink binary for specific OS. Otherwise specify path to plink binary. Default = `NULL`
 #'
 #' @export
 #' @return Data frame

R/leaveoneout.R

@@ -116,7 +116,7 @@ mr_leaveoneout_plot <- function(leaveoneout_results)
 		ggplot2::geom_point(ggplot2::aes(colour=as.factor(tot))) +
 		ggplot2::geom_hline(ggplot2::aes(yintercept = which(levels(SNP) %in% "")), colour="grey") +
 		ggplot2::scale_colour_manual(values=c("black", "red")) +
-		ggplot2::scale_size_manual(values=c(0.3, 1)) +
+		ggplot2::scale_linewidth_manual(values=c(0.3, 1)) +
 		# xlim(c(min(c(0, d$b), na.rm=T), max(c(0, d$b), na.rm=T))) +
 		ggplot2::theme(
 			legend.position="none", 

R/multivariable_mr.R

@@ -9,10 +9,10 @@
 #' @param access_token Google OAuth2 access token. Used to authenticate level of access to data.
 #' @param find_proxies Look for proxies? This slows everything down but is more accurate. The default is `TRUE`.
 #' @param force_server Whether to search through pre-clumped dataset or to re-extract and clump directly from the server. The default is `FALSE`.
-#' @param pval_threshold Instrument detection p-value threshold. Default = 5e-8
+#' @param pval_threshold Instrument detection p-value threshold. Default = `5e-8`
 #' @param pop Which 1000 genomes super population to use for clumping when using the server
-#' @param plink_bin If ‘NULL’ and ‘bfile’ is not ‘NULL’ then will detect packaged plink binary for specific OS. Otherwise specify path to plink binary. Default = ‘NULL’
-#' @param bfile If this is provided then will use the API. Default = ‘NULL’
+#' @param plink_bin If `NULL` and `bfile` is not `NULL` then will detect packaged plink binary for specific OS. Otherwise specify path to plink binary. Default = `NULL`
+#' @param bfile If this is provided then will use the API. Default = `NULL`
 #'
 #' @export
 #' @return data frame in `exposure_dat` format
@@ -80,13 +80,13 @@ mv_extract_exposures <- function(id_exposure, clump_r2=0.001, clump_kb=10000, ha
 #' @param id_col Optional column name to give the dataset an ID. Will be generated automatically if not provided for every trait / unit combination. The default is `"id"`.
 #' @param min_pval Minimum allowed p-value. The default is `1e-200`.
 #' @param log_pval The pval is -log10(P). The default is `FALSE`.
-#' @param pval_threshold Default=5e-8 for clumping
-#' @param plink_bin If ‘NULL’ and ‘bfile’ is not ‘NULL’ then will detect packaged plink binary for specific OS. Otherwise specify path to plink binary. Default = ‘NULL’
-#' @param bfile If this is provided then will use the API. Default = ‘NULL’
-#' @param clump_r2 Default=0.001 for clumping
-#' @param clump_kb Default=10000 for clumping
+#' @param pval_threshold Default=`5e-8` for clumping
+#' @param plink_bin If `NULL` and `bfile` is not `NULL` then will detect packaged plink binary for specific OS. Otherwise specify path to plink binary. Default = `NULL`
+#' @param bfile If this is provided then will use the API. Default = `NULL`
+#' @param clump_r2 Default=`0.001` for clumping
+#' @param clump_kb Default=`10000` for clumping
 #' @param pop Which 1000 genomes super population to use for clumping when using the server
-#' @param harmonise_strictness See action argument in harmonise_data. Default=2
+#' @param harmonise_strictness See action argument in [harmonise_data()]. Default=`2`
 #'
 #' @export
 #' @return List
@@ -252,7 +252,7 @@ mv_extract_exposures_local <- function(
 #' \item{exposure_se}{is the same as `exposure_beta`, but for standard errors.}
 #' \item{exposure_pval}{the same as `exposure_beta`, but for p-values.}
 #' \item{expname}{A data frame with two variables, `id.exposure` and `exposure` which are character strings.}
-#' \item{outcome_beta}{an array of effects for the outcome, corresponding to the SNPs in exposure_beta.}
+#' \item{outcome_beta}{an array of effects for the outcome, corresponding to the SNPs in `exposure_beta`.}
 #' \item{outcome_se}{an array of standard errors for the outcome.}
 #' \item{outcome_pval}{an array of p-values for the outcome.}
 #' \item{outname}{A data frame with two variables, `id.outcome` and `outcome` which are character strings.}

R/singlesnp.R

@@ -118,7 +118,7 @@ mr_forest_plot <- function(singlesnp_results, exponentiate=FALSE)
 		ggplot2::geom_point(ggplot2::aes(colour=as.factor(tot))) +
 		ggplot2::geom_hline(ggplot2::aes(yintercept = which(levels(SNP) %in% "")), colour="grey") +
 		ggplot2::scale_colour_manual(values=c("black", "red")) +
-		ggplot2::scale_size_manual(values=c(0.3, 1)) +
+		ggplot2::scale_linewidth_manual(values=c(0.3, 1)) +
 		# xlim(c(min(c(0, d$b), na.rm=T), max(c(0, d$b), na.rm=T))) +
 		ggplot2::theme(
 			legend.position="none", 

R/zzz.R

@@ -4,7 +4,7 @@
                 paste("TwoSampleMR version", utils::packageVersion("TwoSampleMR"), "\n"),
                 "[>] New: Option to use non-European LD reference panels for clumping etc\n",
                 "[>] Some studies temporarily quarantined to verify effect allele\n",
-                "[>] See news(package='TwoSampleMR') and https://gwas.mrcieu.ac.uk for further details\n"
+                "[>] See news(package = 'TwoSampleMR') and https://gwas.mrcieu.ac.uk for further details\n"
         )
 
 	a <- suppressWarnings(try(readLines("https://raw.githubusercontent.com/MRCIEU/TwoSampleMR/master/DESCRIPTION"), silent=TRUE))

tests/testthat/test_add_metadata.r

@@ -13,61 +13,70 @@ context("add metadata")
 # d9 <- extract_instruments("bbj-a-1") 
 # d10 <- extract_instruments("ieu-b-109")
 
-# save(d1, d2, d3, d4, d5, d6, d7, d8, d9, d10, file="inst/extdata/test_add_metadata.RData")
+# save(d1, d2, d3, d4, d5, d6, d7, d8, d9, d10, file="inst/extdata/test_add_metadata.RData", compress = "xz")
 
 load(system.file("extdata", "test_add_metadata.RData", package="TwoSampleMR"))
 
 test_that("exposure data 1", {
+  skip("Skip unless you have good access to the API.")
 	d1 <- d1 %>% add_metadata()
 	expect_true("units.exposure" %in% names(d1))
 })
 
 test_that("exposure data 2", {
+  skip("Skip unless you have good access to the API.")
 	d2 <- d2 %>% add_metadata()
 	expect_true("units.exposure" %in% names(d2))
 })
 
 test_that("outcome data 1", {
+  skip("Skip unless you have good access to the API.")
 	d3 <- d3 %>% add_metadata()
 	expect_true("units.outcome" %in% names(d3))
 })
 
 test_that("outcome data 2", {
+  skip("Skip unless you have good access to the API.")
 	d4 <- d4 %>% add_metadata()
 	expect_true("units.outcome" %in% names(d4))
 })
 
 test_that("dat 2", {
+  skip("Skip unless you have good access to the API.")
 	d5 <- d5 %>% add_metadata()
 	expect_true("units.outcome" %in% names(d5) & "units.exposure" %in% names(d5))
 })
 
 test_that("no id1", {
+  skip("Skip unless you have good access to the API.")
 	d6$id.exposure <- "not a real id"
 	d6 <- add_metadata(d6)
 	expect_true(!"units.exposure" %in% names(d6))
 })
 
 test_that("no id2", {
+  skip("Skip unless you have good access to the API.")
 	d7$id.outcome <- "not a real id"
 	d7 <- add_metadata(d7)
 	expect_true(!"units.outcome" %in% names(d7))
 })
 
 test_that("ukb-d", {
+  skip("Skip unless you have good access to the API.")
 	d8 <- add_metadata(d8)
 	expect_true("units.outcome" %in% names(d8))
 })
 
 test_that("bbj-a-1", {
+  skip("Skip unless you have good access to the API.")
 	d9 <- d9 %>% add_metadata()
 	expect_true("samplesize.exposure" %in% names(d9))
 	expect_true(all(!is.na(d9$samplesize.exposure)))
 })
 
 test_that("ieu-b-109", {
+  skip("Skip unless you have good access to the API.")
 	d10 <- d10 %>% add_metadata()
 	expect_true("samplesize.exposure" %in% names(d10))
 	expect_true(all(!is.na(d10$samplesize.exposure)))
 })
-

tests/testthat/test_create_test_data.r

@@ -1,8 +1,10 @@
-skip()
+test_that("Create local data", {
+skip("Skip local data creation unless you have good access to the API.")
 
 exp_dat <- extract_instruments("ieu-a-2")
 out_dat <- extract_outcome_data(exp_dat$SNP, "ieu-a-7")
 dat <- make_dat("ieu-a-2", "ieu-a-7") %>% add_metadata()
 dat2 <- make_dat()
 
-save(exp_dat, out_dat, dat, dat2, file=file.path("inst", "extdata", "test_commondata.RData"))
+save(exp_dat, out_dat, dat, dat2, file=file.path("inst", "extdata", "test_commondata.RData"), compress = "xz")
+})

tests/testthat/test_instruments.R

@@ -2,6 +2,8 @@ context("Instruments")
 
 test_that("server and mrinstruments", {
 
+  skip("Skip unless you have good access to the API.")
+
 	# no no
 	exp_dat <- extract_instruments(outcomes=c("ieu-a-1032"))
 	expect_true(length(unique(exp_dat$id)) == 0)

tests/testthat/test_ld.R

@@ -1,11 +1,15 @@
 context("ld")
 
-
-a <- extract_instruments(2, clump=FALSE)
-out <- clump_data(a)
-
+test_that("extract some data", {
+  skip("Skip unless you have good access to the API.")
+  skip_on_ci()
+  skip_on_cran()
+  a <- extract_instruments(2, clump=FALSE)
+  out <- clump_data(a)
+})
 
 test_that("clump", {
+  skip("Skip unless you have good access to the API.")
 	skip_on_ci()
 	skip_on_cran()
 	expect_equal(ncol(a), ncol(out))
@@ -15,6 +19,7 @@ test_that("clump", {
 
 
 test_that("matrix", {
+  skip("Skip unless you have good access to the API.")
 	skip_on_ci()
 	skip_on_cran()
 	b <- ld_matrix(out$SNP)
@@ -24,6 +29,7 @@ test_that("matrix", {
 
 
 test_that("clump multiple", {
+  skip("Skip unless you have good access to the API.")
 	skip_on_ci()
 	skip_on_cran()
 	a <- extract_instruments(c("ieu-a-2", "ieu-a-1001"), clump=FALSE)
@@ -32,9 +38,9 @@ test_that("clump multiple", {
 })
 
 test_that("clump local", {
+  skip("Skip unless you're GH running this test locally.")
 	skip_on_ci()
 	skip_on_cran()
 	skip_if_not(file.exists("/Users/gh13047/repo/opengwas-api-internal/opengwas-api/app/ld_files/EUR.bim"))
 	aclump <- clump_data(a, bfile="/Users/gh13047/repo/opengwas-api-internal/opengwas-api/app/ld_files/EUR", plink_bin="plink")
 })
-

tests/testthat/test_mvmr.R

@@ -1,6 +1,7 @@
 context("mvmr")
 
 test_that("control", {
+  skip("Skip unless you have good access to the API.")
 	skip_on_ci()
 	skip_on_cran()
 	lipids <- mv_extract_exposures(c("ieu-a-299","ieu-a-300","ieu-a-302"))
@@ -18,6 +19,7 @@ test_that("control", {
 
 
 test_that("dat", {
+  skip("Skip unless you have good access to the API.")
 	skip_on_ci()
 	skip_on_cran()
 	a <- mv_extract_exposures(c("ukb-b-5238", "ieu-a-1001"))
@@ -37,6 +39,7 @@ test_that("dat", {
 
 
 test_that("ordering", {
+  skip("Skip unless you have good access to the API.")
 	skip_on_ci()
 	skip_on_cran()
 	lipids1 <- mv_extract_exposures(c("ieu-a-299","ieu-a-300","ieu-a-302"))

tests/testthat/test_mvmr_local.R

@@ -5,11 +5,12 @@ context("mvmr local")
 # rsid <- unique(c(a$rsid, b$rsid))
 # a1 <- ieugwasr::associations(rsid, "ieu-a-2")
 # a2 <- ieugwasr::associations(rsid, "ieu-a-1001")
-# save(a, b, rsid, a1, a2, file="inst/extdata/test_add_mvmr_local.RData")
+# save(a, b, rsid, a1, a2, file="inst/extdata/test_add_mvmr_local.RData", compress = "xz")
 
 load(system.file("extdata", "test_add_mvmr_local.RData", package="TwoSampleMR"))
 
 test_that("mv exposure local", {
+  skip("Skip unless you have good access to the API.")
 	skip_on_ci()
 	skip_on_cran()
     f1 <- tempfile()
@@ -42,4 +43,3 @@ test_that("mv exposure local", {
     expect_true(nrow(exposure_dat) > 100)
     expect_true(all.equal(exposure_dat, exposure_dat2))
 })
-