Seperated reference and query for profile dist

phac-nml · Dec 12, 2024 · f1c2fc6 · f1c2fc6
1 parent 25f22dd
commit f1c2fc6
Show file tree

Hide file tree

Showing 3 changed files with 28 additions and 7 deletions.
diff --git a/modules/local/locidex/merge/main.nf b/modules/local/locidex/merge/main.nf
@@ -10,16 +10,22 @@ process LOCIDEX_MERGE {
     'mwells14/locidex:0.2.3' }"
 
     input:
-    path input_values // [file(sample1), file(sample2), file(sample3), etc...]
+    path input_query // [file(sample1), file(sample2), file(sample3), etc...]
+    path input_reference // [file(sample1), file(sample2), file(sample3), etc...]
 
     output:
-    path("${combined_dir}/*.tsv"), emit: combined_profiles
+    path("${combined_dir1}/*.tsv"), emit: combined_profiles_query
+    path("${combined_dir2}/*.tsv"), emit: combined_profiles_reference
     path "versions.yml", emit: versions
 
     script:
-    combined_dir = "merged"
+    combined_dir1 = "merged_query"
+    combined_dir2 = "merged_reference"
     """
-    locidex merge -i ${input_values.join(' ')} -o ${combined_dir}
+    locidex merge -i ${input_query.join(' ')} -o ${combined_dir1}
+    mv ${combined_dir1}/profile.tsv ${combined_dir1}/profile_query.tsv
+    locidex merge -i ${input_reference.join(' ')} -o ${combined_dir2}
+    mv ${combined_dir2}/profile.tsv ${combined_dir2}/profile_reference.tsv
     cat <<-END_VERSIONS > versions.yml
     "${task.process}":
         locidex merge: \$(echo \$(locidex search -V 2>&1) | sed 's/^.*locidex //' )

diff --git a/modules/local/profile_dists/main.nf b/modules/local/profile_dists/main.nf
@@ -9,6 +9,7 @@ process PROFILE_DISTS{
 
     input:
     path query
+    path reference
     val mapping_format
     path mapping_file
     path columns
@@ -40,7 +41,7 @@ process PROFILE_DISTS{
     }
     prefix = "distances"
     """
-    profile_dists --query $query --ref $query $args --outfmt $mapping_format \\
+    profile_dists --query $query --ref $reference $args --outfmt $mapping_format \\
                 --force \\
                 --distm $params.pd_distm \\
                 --file_type $params.pd_file_type \\

diff --git a/workflows/fastmatchirida.nf b/workflows/fastmatchirida.nf
@@ -104,6 +104,20 @@ workflow FASTMATCH {
         meta, mlst_files -> mlst_files
     }.collect()
 
+    input_assure.result
+    .branch { meta, mlst_file ->
+        reference: meta.ref_query == "reference"
+        query: meta.ref_query == "query"
+        }.set {merged_alleles}
+
+    merged_alleles_query = merged_alleles.query.map{
+        meta, mlst_files -> mlst_files
+    }.collect()
+
+    merged_alleles_reference = merged_alleles.reference.map{
+        meta, mlst_files -> mlst_files
+    }.collect()
+
     metadata_headers = Channel.of(
         tuple(
             SAMPLE_HEADER,
@@ -119,7 +133,7 @@ workflow FASTMATCH {
         meta.metadata_5, meta.metadata_6, meta.metadata_7, meta.metadata_8)
     }.toList()
 
-    merged = LOCIDEX_MERGE(merged_alleles)
+    merged = LOCIDEX_MERGE(merged_alleles_query, merged_alleles_reference)
     ch_versions = ch_versions.mix(merged.versions)
 
     // optional files passed in
@@ -148,7 +162,7 @@ workflow FASTMATCH {
     // Options related to profile dists
     mapping_format = Channel.value("pairwise")
 
-    distances = PROFILE_DISTS(merged.combined_profiles, mapping_format, mapping_file, columns_file)
+    distances = PROFILE_DISTS(merged.combined_profiles_query, merged.combined_profiles_reference, mapping_format, mapping_file, columns_file)
     ch_versions = ch_versions.mix(distances.versions)
 
     // Append metadata to references: