From ecf1d146ab3c3a60fdc5daf0fee18a26c9105c24 Mon Sep 17 00:00:00 2001
From: jasmezz <jasmin.frangenberg@hki-jena.de>
Date: Thu, 9 Jun 2022 16:59:00 +0200
Subject: [PATCH] Add full user options for AMPir

---
 modules/ampir/main.nf        | 15 +++++++++------
 modules/ampir/meta.yml       |  8 ++++++--
 tests/modules/ampir/main.nf  |  8 +++++---
 tests/modules/ampir/test.yml |  4 ++--
 4 files changed, 22 insertions(+), 13 deletions(-)

diff --git a/modules/ampir/main.nf b/modules/ampir/main.nf
index 412317b7..11fc88cf 100644
--- a/modules/ampir/main.nf
+++ b/modules/ampir/main.nf
@@ -9,8 +9,9 @@ process AMPIR {
 
     input:
     tuple val(meta), path(faa)
-    val cut_off
     val model
+    val min_length
+    val min_probability
 
     output:
     tuple val(meta), path("*.faa"), emit: amps_faa
@@ -23,15 +24,17 @@ process AMPIR {
     script:
     def args = task.ext.args ?: ''
     def prefix = task.ext.prefix ?: "${meta.id}"
+    min_length = ("${min_length}" == "[]") ? "": " min_len = as.integer(${min_length})," // Fall back to AMPir default value if none specified
     """
     #!/usr/bin/env Rscript
     library(ampir)
 
-    protein_seqs <- read_faa('${faa}')
-    prediction <- predict_amps(protein_seqs, model = '${model}')
-    prediction <- protein_seqs[which(prediction\$prob_AMP >= as.integer(${cut_off})), ]
-    df_to_faa(protein_seqs, "${prefix}.faa")
-    write.table(prediction, file = "${prefix}.csv", row.names = FALSE, quote = FALSE, dec = '.')
+    input_seqs <- read_faa('${faa}')
+    prediction <- predict_amps(input_seqs,${min_length} model = '${model}')
+    prediction <- prediction[which(prediction\$prob_AMP >= as.numeric(${min_probability})), ]
+    output_seqs <- input_seqs[row.names(prediction), ]
+    write.table(prediction, file = "${prefix}.csv", row.names = FALSE, sep = ";", quote = FALSE, dec = '.')
+    df_to_faa(output_seqs, "${prefix}.faa")
 
     version_file_path <- "versions.yml"
     version_ampir <- paste(unlist(packageVersion("ampir")), collapse = ".")
diff --git a/modules/ampir/meta.yml b/modules/ampir/meta.yml
index 30e55e99..0b3850c0 100644
--- a/modules/ampir/meta.yml
+++ b/modules/ampir/meta.yml
@@ -25,9 +25,13 @@ input:
       pattern: "*.{faa,fasta}"
   - model:
       type: value
-      description: Model for AMP prediction
+      description: Built-in model for AMP prediction
       pattern: "{precursor,mature}"
-  - cut_off:
+  - min_length:
+      type: value
+      description: Minimum protein length for which predictions will be generated
+      pattern: "[0-9]+"
+  - min_probability:
       type: value
       description: Cut-off for AMP prediction
       pattern: "[0-9][0-9]"
diff --git a/tests/modules/ampir/main.nf b/tests/modules/ampir/main.nf
index cc7b07c8..bee2ff90 100644
--- a/tests/modules/ampir/main.nf
+++ b/tests/modules/ampir/main.nf
@@ -10,9 +10,11 @@ workflow test_ampir {
               file(params.test_data['candidatus_portiera_aleyrodidarum']['genome']['proteome_fasta'], checkIfExists: true),
     ]
 
-    cut_off = "80"
-
     model = "precursor"
 
-    AMPIR ( fasta, cut_off, model )
+    min_length = []
+
+    min_probability = "0.7"
+
+    AMPIR ( fasta, model, min_length, min_probability )
 }
diff --git a/tests/modules/ampir/test.yml b/tests/modules/ampir/test.yml
index 40bedd37..4077ef14 100644
--- a/tests/modules/ampir/test.yml
+++ b/tests/modules/ampir/test.yml
@@ -4,8 +4,8 @@
     - ampir
   files:
     - path: output/ampir/test.csv
-      md5sum: 063fd8866f6a669457d7a49404fb449f
+      md5sum: a3b3152373a8add3613ace8c71073e68
     - path: output/ampir/test.faa
-      md5sum: e605d38752fd90261c924d51f7007189
+      md5sum: 0435609144022c55ac196db053f0df89
     - path: output/ampir/versions.yml
       md5sum: 4a11d25b8a904a7ffb34ae88f6826888