P281424/AMR
1
0
Fork 0
mirror of https://github.com/msberends/AMR.git synced 2026-05-14 01:10:45 +02:00
Code Releases Activity

Revert mo_matching_score.R; fix #288 in mo.R pre-filter instead

Browse Source 
Issue #288: when the abbreviated-genus pre-filter (≤3 char genus) yields
exactly one candidate whose species (and subspecies) word(s) exactly match
the input, narrow filtr to that single candidate and set
minimum_matching_score = 0. This bypasses the automatic 0.55 cutoff that
only runs in the is.null() branch, so "S. apiospermum" resolves to
Scedosporium apiospermum without touching the validated scoring formula.

https://claude.ai/code/session_01VH4Ju4Xq9aW1AHuoVbjGEo
This commit is contained in:
Claude
2026-05-06 18:07:37 +00:00
parent b3b8d301ff
commit d23004641f
2 changed files with 14 additions and 22 deletions
Download Patch File Download Diff File Expand all files Collapse all files

12
R/mo.R
View File

@@ -352,6 +352,18 @@ as.mo <- function(x,
(MO_lookup_current$species_first == substr(x_parts[2], 1, 1) |
MO_lookup_current$subspecies_first == substr(x_parts[2], 1, 1) |
MO_lookup_current$subspecies_first == substr(x_parts[3], 1, 1)))
# Issue #288: if the species (and subspecies) word(s) in the input exactly match
# exactly one candidate, use only that candidate and bypass the 0.55 cutoff.
# This prevents prevalent bacteria from outranking a rarer organism whose species
# epithet is an unambiguous exact match, e.g. "S. apiospermum" → Scedosporium.
sp_exact <- tolower(MO_lookup_current$species[filtr]) == x_parts[2]
if (length(x_parts) == 3) {
sp_exact <- sp_exact & tolower(MO_lookup_current$subspecies[filtr]) == x_parts[3]
}
if (sum(sp_exact) == 1) {
filtr <- filtr[sp_exact]
minimum_matching_score <- 0
}
} else {
filtr <- which(MO_lookup_current$full_first == substr(x_parts[1], 1, 1) |
MO_lookup_current$species_first == substr(x_parts[2], 1, 1) |

24
R/mo_matching_score.R
View File

@@ -125,26 +125,6 @@ mo_matching_score <- function(x, n) {
# kingdom index (Bacteria = 1, Fungi = 2, Protozoa = 3, Archaea = 4, others = 5)
k_n <- AMR_env$MO_lookup[match(n, AMR_env$MO_lookup$fullname), "kingdom_index", drop = TRUE]
# base matching score
score <- (l_n - 0.5 * l_n.lev) / (l_n * p_n * k_n)
# Issue #288: when the genus is abbreviated (≤3 chars) and the species epithet of the
# candidate exactly matches the species epithet of the input, boost the score ×2.
# This prevents a prevalent bacterium (low p_n/k_n) from outranking a rarer organism
# whose species epithet is the only exact match, e.g. "S. apiospermum" → Scedosporium.
x_parts_list <- strsplit(x, " ", fixed = TRUE)
n_parts_list <- strsplit(n, " ", fixed = TRUE)
x_genus <- vapply(x_parts_list, function(w) if (length(w) >= 1) w[1L] else "", character(1L))
x_sp <- vapply(x_parts_list, function(w) if (length(w) >= 2L) tolower(w[2L]) else "", character(1L))
n_g1 <- vapply(n_parts_list, function(w) if (length(w) >= 1L) tolower(substr(w[1L], 1L, 1L)) else "", character(1L))
n_sp <- vapply(n_parts_list, function(w) if (length(w) >= 2L) tolower(w[2L]) else "", character(1L))
exact_sp <- nchar(x_genus) <= 3L &
x_sp != "" &
n_sp != "" &
tolower(substr(x_genus, 1L, 1L)) == n_g1 &
x_sp == n_sp
score[exact_sp] <- score[exact_sp] * 2
score
# matching score:
(l_n - 0.5 * l_n.lev) / (l_n * p_n * k_n)
}