Add flip_frq_as_biallelic parameter

DESCRIPTION

@@ -1,7 +1,7 @@
 Package: MungeSumstats
 Type: Package
 Title: Standardise summary statistics from GWAS
-Version: 1.9.15
+Version: 1.9.16
 Authors@R:
     c(person(given = "Alan",
            family = "Murphy",

NEWS.md

@@ -1,3 +1,12 @@
+## CHANGES IN VERSION 1.9.16
+
+### New features
+* flip_frq_as_biallelic parameter added enabling frequencies of non-bi-allelic
+SNPs to be flipped as if they were bi-allelic (1 - frequency) i.e. ignoring the
+frequencies of other alternative alleles (assuming these will be negligible). 
+Note this will not be done as default as it is not fully correct but may be 
+useful for some users.
+
 ## CHANGES IN VERSION 1.9.15
 
 ### Bug fix

R/check_allele_flip.R

@@ -31,6 +31,7 @@ check_allele_flip <- function(sumstats_dt, path,
                               allele_flip_z,
                               allele_flip_frq,
                               bi_allelic_filter,
+                              flip_frq_as_biallelic,
                               imputation_ind,
                               log_folder_ind,
                               check_save_out,
@@ -158,7 +159,7 @@ check_allele_flip <- function(sumstats_dt, path,
             sumstats_dt[match_type == FALSE, A1 := tmp]
             sumstats_dt[, tmp := NULL]
 
-            # flip effect column(s) - BETA, OR, z, log_odds, SIGNED_SUMSTAT, FRQ
+            # flip effect column(s) - BETA, OR, Z, log_odds, SIGNED_SUMSTAT, FRQ
             effect_columns <- c("BETA", "OR", "LOG_ODDS", "SIGNED_SUMSTAT")
             if (allele_flip_z) {
                 effect_columns <- c(effect_columns, "Z")
@@ -178,13 +179,33 @@ check_allele_flip <- function(sumstats_dt, path,
                 "bi_allelic_filter to TRUE so\nnon-bi-allelic SNPs are",
                 " removed. Otherwise, set allele_flip_frq to FALSE to ",
                 "not flip the FRQ column but note\nthis could lead to ",
-                "incorrect FRQ values."
+                "incorrect FRQ values.\nA new option added is to flip ",
+                "non-bi-allelic SNPs as if they were bi-allelic (1-FRQ), ",
+                "to do this set\nflip_frq_as_biallelic to TRUE but note ",
+                "these values will not be exact down to the missing ",
+                "frequencies of the other\nalternative allele(s)."
             )
             if (nrow(sumstats_dt[match_type == FALSE, ]) > 0 &&
                 "FRQ" %in% effect_columns &&
-                !bi_allelic_filter) {
+                !bi_allelic_filter && !flip_frq_as_biallelic) {
                 stop(stp_msg)
             }
+            #if set flip_frq_as_biallelic = TRUE, let them know
+            if (nrow(sumstats_dt[match_type == FALSE, ]) > 0 &&
+                "FRQ" %in% effect_columns &&
+                !bi_allelic_filter && flip_frq_as_biallelic) {
+              print_msg <- paste0(
+                "Note: You have set flip_frq_as_biallelic to TRUE meaning of ",
+                "the ",
+                formatC(nrow(sumstats_dt[match_type == FALSE, ]),
+                        big.mark = ","),
+                " SNPs to be flipped, ",
+                "any non-bi-allelic SNPs\nwill have their frequencies flipped ",
+                "as (1-FRQ) essentially ignoring the frequencies of any other ",
+                "alternative alleles."
+              )
+              message(print_msg)
+            }
             for (eff_i in effect_columns) { # set updates quicker for DT
                 # conversion done in case, VCF beta column may not be numeric
                 if (eff_i == "FRQ") {

R/format_sumstats.R

@@ -140,6 +140,10 @@
 #' along with effect and z-score columns like Beta? Default TRUE.
 #' @param bi_allelic_filter Binary Should non-biallelic SNPs be removed. Default
 #' is TRUE.
+#' @param flip_frq_as_biallelic Binary Should non-bi-allelic SNPs frequency 
+#' values be flipped as 1-p despite there being other alternative alleles? 
+#' Default is FALSE but if set to TRUE, this allows non-bi-allelic SNPs to be 
+#' kept despite needing flipping.
 #' @param snp_ids_are_rs_ids Binary Should the supplied SNP ID's be assumed to
 #' be RSIDs. If not, imputation using the SNP ID for other columns like
 #' base-pair position or chromosome will not be possible. If set to FALSE, the
@@ -249,6 +253,7 @@ format_sumstats <- function(path,
                             allele_flip_z = TRUE,
                             allele_flip_frq = TRUE,
                             bi_allelic_filter = TRUE,
+                            flip_frq_as_biallelic = FALSE,
                             snp_ids_are_rs_ids = TRUE,
                             remove_multi_rs_snp = FALSE,
                             frq_is_maf = TRUE,
@@ -336,6 +341,7 @@ format_sumstats <- function(path,
             allele_flip_z = allele_flip_z,
             allele_flip_frq = allele_flip_frq,
             bi_allelic_filter = bi_allelic_filter,
+            flip_frq_as_biallelic = flip_frq_as_biallelic,
             snp_ids_are_rs_ids = snp_ids_are_rs_ids,
             remove_multi_rs_snp = remove_multi_rs_snp,
             frq_is_maf = frq_is_maf,
@@ -702,6 +708,7 @@ format_sumstats <- function(path,
                 allele_flip_z = allele_flip_z,
                 allele_flip_frq = allele_flip_frq,
                 bi_allelic_filter = bi_allelic_filter,
+                flip_frq_as_biallelic = flip_frq_as_biallelic,
                 imputation_ind = imputation_ind,
                 log_folder_ind = log_folder_ind,
                 check_save_out = check_save_out,

R/validate_parameters.R

@@ -27,6 +27,7 @@ validate_parameters <- function(path,
                                 allele_flip_z,
                                 allele_flip_frq,
                                 bi_allelic_filter,
+                                flip_frq_as_biallelic,
                                 snp_ids_are_rs_ids,
                                 remove_multi_rs_snp,
                                 frq_is_maf,
@@ -242,6 +243,9 @@ validate_parameters <- function(path,
     if (!is.logical(bi_allelic_filter)) {
         stop("bi_allelic_filter must be either TRUE or FALSE")
     }
+    if (!is.logical(flip_frq_as_biallelic)) {
+      stop("flip_frq_as_biallelic must be either TRUE or FALSE")
+    }
     if (!is.logical(snp_ids_are_rs_ids)) {
         stop("snp_ids_are_rs_ids must be either TRUE or FALSE")
     }

vignettes/MungeSumstats.Rmd

@@ -369,6 +369,10 @@ conducted by *MungeSumstats* are:
     Z-score columns (e.g. Beta). Default is TRUE.
 -   **bi_allelic_filter** Binary, should non-biallelic SNPs be removed.
     Default is TRUE
+-   **flip_frq_as_biallelic** Binary, Should non-bi-allelic SNPs frequency 
+    values be flipped as 1-p despite there being other alternative alleles? 
+    Default is FALSE but if set to TRUE, this allows non-bi-allelic SNPs to be 
+    kept despite needing flipping.    
 -   **snp_ids_are_rs_ids** Binary, should the SNP IDs inputted be
     inferred as RS IDs or some arbitrary ID. Default is TRUE.\
 -   **remove_multi_rs_snp** Binary Sometimes summary statistics can have