Make fasta optional for gtf filtering

Author: pinin4fjords

bin/filter_gtf.py

@@ -6,7 +6,7 @@
 import argparse
 import re
 import statistics
-from typing import Set
+from typing import Optional, Set
 
 # Create a logger
 logging.basicConfig(format="%(name)s - %(asctime)s %(levelname)s: %(message)s")
@@ -27,14 +27,15 @@ def tab_delimited(file: str) -> float:
         return statistics.median(line.count("\t") for line in data.split("\n"))
 
 
-def filter_gtf(fasta: str, gtf_in: str, filtered_gtf_out: str, skip_transcript_id_check: bool) -> None:
+def filter_gtf(fasta: Optional[str], gtf_in: str, filtered_gtf_out: str, skip_transcript_id_check: bool) -> None:
     """Filter GTF file based on FASTA sequence names."""
     if tab_delimited(gtf_in) != 8:
         raise ValueError("Invalid GTF file: Expected 9 tab-separated columns.")
 
-    seq_names_in_genome = extract_fasta_seq_names(fasta)
-    logger.info(f"Extracted chromosome sequence names from {fasta}")
-    logger.debug("All sequence IDs from FASTA: " + ", ".join(sorted(seq_names_in_genome)))
+    if (fasta is not None):
+        seq_names_in_genome = extract_fasta_seq_names(fasta)
+        logger.info(f"Extracted chromosome sequence names from {fasta}")
+        logger.debug("All sequence IDs from FASTA: " + ", ".join(sorted(seq_names_in_genome)))
 
     seq_names_in_gtf = set()
     try:
@@ -44,7 +45,7 @@ def filter_gtf(fasta: str, gtf_in: str, filtered_gtf_out: str, skip_transcript_i
                 seq_name = line.split("\t")[0]
                 seq_names_in_gtf.add(seq_name)  # Add sequence name to the set
 
-                if seq_name in seq_names_in_genome:
+                if fasta is None or seq_name in seq_names_in_genome:
                     if skip_transcript_id_check or re.search(r'transcript_id "([^"]+)"', line):
                         out.write(line)
                         line_count += 1
@@ -63,7 +64,7 @@ def filter_gtf(fasta: str, gtf_in: str, filtered_gtf_out: str, skip_transcript_i
 if __name__ == "__main__":
     parser = argparse.ArgumentParser(description="Filters a GTF file based on sequence names in a FASTA file.")
     parser.add_argument("--gtf", type=str, required=True, help="GTF file")
-    parser.add_argument("--fasta", type=str, required=True, help="Genome fasta file")
+    parser.add_argument("--fasta", type=str, required=False, help="Genome fasta file")
     parser.add_argument("--prefix", dest="prefix", default="genes", type=str, help="Prefix for output GTF files")
     parser.add_argument(
         "--skip_transcript_id_check", action="store_true", help="Skip checking for transcript IDs in the GTF file"

modules/local/gtf_filter/main.nf

@@ -18,11 +18,15 @@ process GTF_FILTER {
     task.ext.when == null || task.ext.when
 
     script: // filter_gtf.py is bundled with the pipeline, in nf-core/rnaseq/bin/
+    fasta_text=''
+    if (fasta){
+        fasta_text="--fasta $fasta"
+    }
     """
     filter_gtf.py \\
         --gtf $gtf \\
-        --fasta $fasta \\
-        --prefix ${fasta.baseName}
+        $fasta_text \\
+        --prefix ${gtf.baseName}
 
     cat <<-END_VERSIONS > versions.yml
     "${task.process}":