Merge pull request #141 from Joon-Klaps/no-ref-hits

Add logic to allow samples with no reference hits to be analysed

CHANGELOG.md

@@ -15,6 +15,7 @@ Initial release of Joon-Klaps/viralgenie, created with the [nf-core](https://nf-
 - Add read & contig decomplexification using prinseq++  ([#133](https://github.com/Joon-Klaps/viralgenie/pull/133))
 - Add option to filter contig clusters based on cumulative read coverage ([#138](https://github.com/Joon-Klaps/viralgenie/pull/138))
 - Adding mash-screen output to result table ([#140](https://github.com/Joon-Klaps/viralgenie/pull/140))
+- Add logic to allow samples with no reference hits to be analysed ([#141](https://github.com/Joon-Klaps/viralgenie/pull/141))
 
 ### `Fixed`
 

bin/blast_filter.py

@@ -177,6 +177,15 @@ def main(argv=None):
     """Coordinate argument parsing and program execution."""
     args = parse_args(argv)
     logging.basicConfig(level=args.log_level, format="[%(levelname)s] %(message)s")
+
+    if args.blast is None and args.contigs.is_file():
+        logger.warning(f"No blast input was provide, just copying input file.")
+        with open(args.contigs, "r") as contigs_file:
+            contig_content = contigs_file.read()
+        with open(f"{args.prefix}_withref.fa", "w") as f:
+            f.write(contig_content)
+        return 0
+
     if not args.blast.is_file():
         logger.error(f"The given input file {args.blast} was not found!")
         sys.exit(2)
@@ -193,6 +202,8 @@ def main(argv=None):
 
     write_hits(df_filter, args.contigs, args.references, args.prefix)
 
+    return 0
+
 
 if __name__ == "__main__":
     sys.exit(main())

bin/custom_multiqc_tables.py

@@ -213,14 +213,24 @@ def concat_table_files(table_files, **kwargs):
     Returns:
         pandas.DataFrame: The concatenated dataframe.
     """
-    df = pd.concat(
-        [
+    try:
+        valid_dfs = [
             read_file_to_dataframe(file, **kwargs)
             for file in table_files
             if check_file_exists(file)
         ]
-    )
-    return df
+
+        if not valid_dfs:
+            logging.warning(f"Warning concatenating files: {table_files}")
+            logging.warning("No valid files found to concatenate.")
+            return pd.DataFrame()
+
+        df = pd.concat(valid_dfs)
+        return df
+
+    except ValueError as e:
+        logging.warning(f"Warning concatenating files: {table_files}")
+        return pd.DataFrame()
 
 
 def read_in_quast(table_files):

conf/modules.config

@@ -890,6 +890,7 @@ process {
             }
 
             withName: RENAME_FASTA_HEADER_SINGLETON {
+                ext.prefix = { "${meta.id}_singleton" } // DON'T CHANGE
                 publishDir = [
                     enabled: false
                 ]

conf/tests/test_fail_mapped.config

@@ -35,7 +35,7 @@ params {
     mapping_constrains           = "${projectDir}/assets/samplesheets/mapping_constrains_fail.tsv"
 
     min_mapped_reads            = 100
-    intermediate_mapping_stats      = true
+    intermediate_mapping_stats  = true
     skip_checkv                 = true
 }
 

modules/local/blast_filter/main.nf

@@ -8,13 +8,14 @@ process BLAST_FILTER {
         'biocontainers/mulled-v2-949aaaddebd054dc6bded102520daff6f0f93ce6:aa2a3707bfa0550fee316844baba7752eaab7802-0' }"
 
     input:
-    tuple val(meta), path(blast), path(contigs)
+    tuple val(meta), path(blast)
+    tuple val(meta), path(contigs)
     tuple val(meta2), path(db)
 
     output:
-    tuple val(meta), path("*.hits.txt")  , emit: hits
+    tuple val(meta), path("*.hits.txt")  , emit: hits, optional: true
     tuple val(meta), path("*.fa")        , emit: sequence
-    tuple val(meta), path("*.filter.tsv"), emit: filter
+    tuple val(meta), path("*.filter.tsv"), emit: filter, optional: true
     path "versions.yml"                  , emit: versions
 
     when:
@@ -23,10 +24,11 @@ process BLAST_FILTER {
     script:
     def args = task.ext.args ?: ''
     def prefix = task.ext.prefix ?: "${meta.id}"
+    def blast_command = blast ? "-i ${blast}" : ""
     """
     blast_filter.py \\
         $args \\
-        -i ${blast} \\
+        ${blast_command} \\
         -c ${contigs} \\
         -r ${db} \\
         -p ${prefix}

subworkflows/local/fasta_blast_refsel.nf

@@ -45,11 +45,17 @@ workflow FASTA_BLAST_REFSEL {
     // Filter out false positve hits that based on query length, alignment length, identity, e-score & bit-score
     ch_blast_txt
         .hits
-        .join(fasta, by:[0], remainder:false)
-        .set{ hits_contigs }
+        .join(fasta, by:[0], remainder:true)
+        .multiMap{
+            meta, txt, fasta ->
+            hits : [meta, txt ? txt : []]
+            contigs : [meta, fasta]
+        }
+        .set{input_blast_filter}
 
     BLAST_FILTER (
-        hits_contigs,
+        input_blast_filter.hits,
+        input_blast_filter.contigs,
         blast_db_fasta
     )
     ch_versions = ch_versions.mix(BLAST_FILTER.out.versions.first())