P281424/AMR
1
0
Fork 0
mirror of https://github.com/msberends/AMR.git synced 2025-07-09 02:03:04 +02:00
Code Releases Activity

(v1.3.0.9031) matching score update

Browse Source
This commit is contained in:
dr. M.S. (Matthijs) Berends
2020-09-26 16:51:17 +02:00
parent 050a9a04fb
commit 22f6ceb3e4
19 changed files with 85 additions and 81 deletions
Download Patch File Download Diff File Expand all files Collapse all files

16
R/mo.R
View File

@ -301,7 +301,7 @@ exec_as.mo <- function(x,
initial = initial_search,
uncertainty = actual_uncertainty,
input_actual = actual_input) {
if (!is.null(input_actual)) {
input <- input_actual
} else {
@ -318,7 +318,7 @@ exec_as.mo <- function(x,
if (NROW(res_df) > 1 & uncertainty != -1) {
# sort the findings on matching score
scores <- mo_matching_score(x = input,
fullname = res_df[, "fullname", drop = TRUE])
n = res_df[, "fullname", drop = TRUE])
res_df <- res_df[order(scores, decreasing = TRUE), , drop = FALSE]
}
res <- as.character(res_df[, column, drop = TRUE])
@ -442,7 +442,7 @@ exec_as.mo <- function(x,
# we need special treatment for very prevalent full names, they are likely!
# e.g. as.mo("Staphylococcus aureus")
x <- MO_lookup[match(tolower(x), MO_lookup$fullname_lower), property, drop = TRUE]
} else if (all(x %in% reference_data_to_use$fullname)) {
# we need special treatment for very prevalent full names, they are likely!
# e.g. as.mo("Staphylococcus aureus")
@ -1544,7 +1544,7 @@ exec_as.mo <- function(x,
# this will save the uncertain items as attribute, so they can be bound to `uncertainties` in the uncertain_fn() function
x <- structure(x, uncertainties = uncertainties)
}
if (old_mo_warning == TRUE & property != "mo") {
warning("The input contained old microorganism IDs from previous versions of this package.\nPlease use `as.mo()` on these old IDs to transform them to the new format.\nSUPPORT FOR THIS WILL BE DROPPED IN A FUTURE VERSION.", call. = FALSE)
@ -1639,7 +1639,7 @@ freq.mo <- function(x, ...) {
")"),
`No. of genera` = pm_n_distinct(mo_genus(x_noNA, language = NULL)),
`No. of species` = pm_n_distinct(paste(mo_genus(x_noNA, language = NULL),
mo_species(x_noNA, language = NULL)))))
mo_species(x_noNA, language = NULL)))))
}
#' @method print mo
@ -1773,7 +1773,7 @@ print.mo_uncertainties <- function(x, ...) {
if (x[i, ]$candidates != "") {
candidates <- unlist(strsplit(x[i, ]$candidates, ", ", fixed = TRUE))
scores <- mo_matching_score(x = x[i, ]$input,
fullname = candidates)
n = candidates)
# sort on descending scores
candidates <- candidates[order(1 - scores)]
n_candidates <- length(candidates)
@ -1799,8 +1799,8 @@ print.mo_uncertainties <- function(x, ...) {
ifelse(!is.na(x[i, ]$renamed_to), paste(", renamed to", font_italic(x[i, ]$renamed_to)), ""),
" (", x[i, ]$mo,
", matching score = ", trimws(percentage(mo_matching_score(x = x[i, ]$input,
fullname = x[i, ]$fullname),
digits = 1)),
n = x[i, ]$fullname),
digits = 1)),
") "),
uncertainty_interpretation,
candidates),

17
R/mo_matching_score.R
View File

@ -24,20 +24,21 @@
#' This helper function is used by [as.mo()] to determine the most probable match of taxonomic records, based on user input.
#' @param x Any user input value(s)
#' @param n A full taxonomic name, that exists in [`microorganisms$fullname`][microorganisms]
#' @param uncertainty The level of uncertainty set in [as.mo()], see `allow_uncertain` in that function (here, it defaults to 1, but is automatically determined in [as.mo()] based on the number of transformations needed to get to a result)
#' @section Matching score for microorganisms:
#' With ambiguous user input in [as.mo()] and all the [`mo_*`][mo_property()] functions, the returned results are chosen based on their matching score using [mo_matching_score()]. This matching score \eqn{m} is calculated as:
#' With ambiguous user input in [as.mo()] and all the [`mo_*`][mo_property()] functions, the returned results are chosen based on their matching score using [mo_matching_score()]. This matching score \eqn{m}, ranging from 0 to 100%, is calculated as:
#'
#' \deqn{m_{(x, n)} = \frac{l_{n} - 0.5 \times \min \begin{cases}l_{n} \\ \operatorname{lev}(x, n)\end{cases}}{l_{n} p k}}{m(x, n) = ( l_n * min(l_n, lev(x, n) ) ) / ( l_n * p * k )}
#' \deqn{m_{(x, n)} = \frac{l_{n} - 0.5 \cdot \min \begin{cases}l_{n} \\ \operatorname{lev}(x, n)\end{cases}}{l_{n} \cdot p_{n} \cdot k_{n}}}{m(x, n) = ( l_n * min(l_n, lev(x, n) ) ) / ( l_n * p_n * k_n )}
#'
#' where:
#'
#' * \eqn{x} is the user input;
#' * \eqn{n} is a taxonomic name (genus, species and subspecies);
#' * \eqn{l_{n}}{l_n} is the length of the taxonomic name;
#' * \eqn{\operatorname{lev}}{lev} is the [Levenshtein distance](https://en.wikipedia.org/wiki/Levenshtein_distance) function;
#' * \eqn{p} is the human pathogenic prevalence, categorised into group \eqn{1}, \eqn{2} and \eqn{3} (see *Details* in `?as.mo`), meaning that \eqn{p = \{1, 2 , 3\}}{p = {1, 2, 3}};
#' * \eqn{k} is the kingdom index, set as follows: Bacteria = \eqn{1}, Fungi = \eqn{2}, Protozoa = \eqn{3}, Archaea = \eqn{4}, and all others = \eqn{5}, meaning that \eqn{k = \{1, 2 , 3, 4, 5\}}{k = {1, 2, 3, 4, 5}}.
#' * \eqn{n} is a taxonomic name (genus, species and subspecies) as found in [`microorganisms$fullname`][microorganisms];
#' * \eqn{l_{n}}{l_n} is the length of \eqn{n};
#' * \eqn{\operatorname{lev}}{lev} is the [Levenshtein distance function](https://en.wikipedia.org/wiki/Levenshtein_distance);
#' * \eqn{p_{n}}{p_n} is the human pathogenic prevalence of \eqn{n}, categorised into group \eqn{1}, \eqn{2} and \eqn{3} (see *Details* in `?as.mo`), meaning that \eqn{p = \{1, 2 , 3\}}{p = {1, 2, 3}};
#' * \eqn{k_{n}}{k_n} is the kingdom index of \eqn{n}, set as follows: Bacteria = \eqn{1}, Fungi = \eqn{2}, Protozoa = \eqn{3}, Archaea = \eqn{4}, and all others = \eqn{5}, meaning that \eqn{k = \{1, 2 , 3, 4, 5\}}{k = {1, 2, 3, 4, 5}}.
#'
#' This means that the user input `x = "E. coli"` gets for *Escherichia coli* a matching score of `r percentage(mo_matching_score("E. coli", "Escherichia coli"), 1)` and for *Entamoeba coli* a matching score of `r percentage(mo_matching_score("E. coli", "Entamoeba coli"), 1)`.
#'
#' All matches are sorted descending on their matching score and for all user input values, the top match will be returned.
#' @export