From b2402ea9972430a7b30fdc6bcc8daa9b0be7e49b Mon Sep 17 00:00:00 2001
From: Elias Projahn <elias@johrpan.de>
Date: Wed, 20 Oct 2021 11:20:20 +0200
Subject: [PATCH] Compute n_species dynamically

---
 data.R   |  1 -
 server.R | 11 ++++++++++-
 2 files changed, 10 insertions(+), 2 deletions(-)

diff --git a/data.R b/data.R
index 4b47863..71ecb11 100644
--- a/data.R
+++ b/data.R
@@ -65,7 +65,6 @@ genes <- geposan::genes[, .(
     id,
     name,
     chromosome,
-    n_species,
     suggested = name %chin% genes_suggested_tpe_old,
     verified = name %chin% genes_verified_tpe_old
 )]
diff --git a/server.R b/server.R
index a102e15..df95a8e 100644
--- a/server.R
+++ b/server.R
@@ -69,8 +69,17 @@ server <- function(input, output, session) {
             by.y = "id"
         )
 
+        # Count included species from the preset per gene.
+        genes_n_species <- distances[
+            species %chin% preset$species_ids,
+            .(n_species = .N),
+            by = "gene"
+        ]
+
+        setkey(genes_n_species, gene)
+
         # Exclude genes with too few species.
-        results[n_species >= input$n_species]
+        results[genes_n_species[gene, n_species] >= input$n_species]
     })
 
     # Rank the results.