update docs

.Rbuildignore

@@ -23,3 +23,4 @@ paper/
 ^vignettes/articles$
 ^optimize-figures.sh$
 ^R/GranthamDist/
+^inst/extdata/

DESCRIPTION

@@ -8,7 +8,7 @@ Description: Fetch sequence alignment data from the IMGTHLA database by Robinson
 License: GPL (>= 3)
 Encoding: UTF-8
 Roxygen: list(markdown = TRUE)
-RoxygenNote: 7.2.3
+RoxygenNote: 7.3.1
 URL: https://github.com/slowkow/hlabud, https://slowkow.github.io/hlabud/
 BugReports: https://github.com/slowkow/hlabud/issues
 Depends: 

NAMESPACE

@@ -46,4 +46,5 @@ importFrom(utils,download.file)
 importFrom(utils,head)
 importFrom(utils,read.delim)
 importFrom(utils,read.table)
+importFrom(utils,tail)
 importFrom(utils,untar)

R/hla_divergence.R

@@ -67,21 +67,27 @@ V	64	96	133	152	192	96	121	109	84	29	32	97	21	50	68	124	69	88	55	0")
 #' @param alleles A character vector of comma-delimited alleles for each individual. We usually expect two alleles per individual, but it is possible to have more (or fewer) copies due to copy number alterations. This function still works when each individual has a different number of alleles.
 #' @param method A pairwise amino acid matrix, or a method name: `"grantham"` or `"uniform"` to indicate which pairwise amino acid distance matrix to use. If you choose to pass a matrix, then it should be a 20x20 symmetric matrix with zeros on the diagonal, and the rownames and colnames should be the one-letter amino acid codes `A R N D C Q E G H I L K M F P S T W Y V`.
 #' @param release Default is "latest". Should be a release name like "3.51.0".
+#' @param positions Only use these positions in the calculation. Should be column names from the `hla_alignments()$alleles` matrix.
 #' @param verbose If TRUE, print messages along the way.
 #' @seealso [hla_releases()] to get a complete list of all release names.
 #' @seealso [amino_distance_matrix()] to get a amino acid distance matrix that you can use with `hla_divergence()`.
 #' @return A dataframe with divergence for each individual.
 #' @examples
-#' my_genos <- c("A*23:01:12,A*24:550", "A*25:12N,A*11:27",
-#'               "A*24:381,A*33:85", "A*01:01:,A*01:01,A*02:01")
+#' my_genos <- c("A*23:01:12,A*24:550",
+#'               "A*25:12N,A*11:27",
+#'               "A*24:381,A*33:85",
+#'               "A*01:01:,A*01:01,A*02:01") # this individual has 3 alleles
 #' hla_divergence(my_genos, method = "grantham")
 #'
 #' # This is equivalent
 #' hla_divergence(my_genos, method = amino_distance_matrix("grantham"))
+#'
+#' # Compute divergence on a subset of positions:
+#' hla_divergence(my_genos, positions = c("n1", "1", "2", "3", "4"))
 #' @export
 hla_divergence <- function(
-  alleles = c("A*01:01,A*02:01"), method = "grantham", release = "latest", verbose = FALSE,
-  positions = NULL
+  alleles = c("A*01:01,A*02:01"), method = "grantham", release = "latest",
+  positions = NULL, verbose = FALSE
 ) {
 
   aminos <- c("A", "R", "N", "D", "C", "Q", "E", "G", "H", "I", "L", "K", "M",
@@ -165,7 +171,6 @@ hla_divergence <- function(
         }
       }
     }
-    allelePairs
 
     my_combn <- combn(my_alleles, 2)
     my_dist <- 0

R/hlabud.R

@@ -450,6 +450,7 @@ read_alignments <- function(file) {
 #' @param drop_identical Drop positions where all alleles have the same amino acid. Default is false.
 #' @param verbose Print messages along the way.
 #' @keywords internal
+#' @importFrom utils tail
 get_onehot <- function(sequences, n_pre, drop_identical = FALSE, verbose = FALSE) {
 
   # Split each string into a vector of individual characters