Merge pull request #9 from mrc-ide/develop

added function to subset based on a position string

DESCRIPTION

@@ -1,7 +1,7 @@
 Package: variantstring
 Type: Package
 Title: Functions for working with variant string format
-Version: 1.5.1
+Version: 1.6.0
 Authors@R: c(
     person("Bob", "Verity", email = "r.verity@imperial.ac.uk", role = c("aut", "cre"))
     )

NAMESPACE

@@ -15,6 +15,7 @@ export(order_position_string)
 export(order_variant_string)
 export(position_from_variant_string)
 export(position_to_long)
+export(subset_position)
 export(variant_to_long)
 import(dplyr)
 importFrom(stats,na.omit)

R/main.R

@@ -7,6 +7,7 @@
 # position_to_long
 # long_to_position
 # position_from_variant_string
+# subset_position
 # order_variant_string
 # order_position_string
 # count_het_loci
@@ -730,6 +731,46 @@ position_from_variant_string <- function(x) {
 
 }
 
+#------------------------------------------------
+#' @title Subset position of a variant string
+#'
+#' @description
+#' Given a vector of variant strings and a single position string, subsets all
+#' variant strings to only the genes and codons in the position string. Retains
+#' read counts at these positions if present.
+#'
+#' @param position_string a single position string.
+#' @param variant_strings a variant string or vector of variant strings.
+#'
+#' @import dplyr
+#'
+#' @export
+
+subset_position <- function(position_string, variant_strings) {
+
+  # checks
+  check_position_string(position_string)
+  stopifnot(length(position_string) == 1)
+  check_variant_string(variant_strings)
+
+  # get position string in long form
+  df_position <- position_to_long(position_string)[[1]]
+
+  ret <- mapply(function(x) {
+    df_diff <- anti_join(df_position, x, by = c("gene", "pos"))
+    if (nrow(df_diff) == 0) {
+      df_sub <- semi_join(x, df_position, by = c("gene", "pos"))
+      ret <- long_to_variant(list(df_sub))
+    } else {
+      ret <- NA
+    }
+    ret
+  }, variant_to_long(variant_strings), SIMPLIFY = FALSE) |>
+    unlist()
+
+  return(ret)
+}
+
 #------------------------------------------------
 #' @title Reorders a variant string
 #'

README.md

@@ -12,6 +12,7 @@ In brief, this package contains functions that...
 
 - Check for correctly formatted variant and position strings.
 - Extract a position string from a variant string.
+- Subset a variant string based on a position string.
 - Compare two variant strings to look for a match (useful in numerator of prevalence calculation). Reports if this is an exact match or an ambiguous match.
 - Compare a position string against a variant string to look for a match (useful in denominator of prevalence calculation).
 - Convert between string format and a long-form data.frame format.
@@ -20,4 +21,4 @@ There are also a few more utility functions not listed here - see the package he
 
 ## Release history
 
-The current version is 1.5.1, released 16 Jan 2025.
+The current version is 1.6.0, released 16 Jan 2025.