Introduction Usage Parameters Output

Introduction

This document describes the output produced by ncrnannotator. All paths are relative to the --outdir directory specified at runtime.

Pipeline overview

ncrnannotator runs the following steps:

  1. FILTER_RFAM_CM — filter Rfam covariance models to clade-specific accessions (skipped in mgnify-assembly and full modes)
  2. GENOME_CHUNK — split genome into non-overlapping windows for parallel processing
  3. CMSEARCH — search each chunk against Rfam covariance models using Infernal
  4. PARSE_RFAM — consolidate results, remove overlapping hits, apply GA score thresholds
  5. RFAM_TO_FORMATS — convert hits to GTF, GFF3, and BED annotation files
  6. MultiQC — aggregate software versions and pipeline metrics into a report

Output directories

rfam/

Intermediate Rfam files published for inspection and reuse.

Output files
  • rfam/
    • rfam_filtered.cm — Rfam covariance model file filtered to clade-specific accessions (not present in mgnify-assembly or full mode)
    • rfam_hits.tsv — Tab-separated table of all final ncRNA hits after overlap removal and score filtering

rfam_hits.tsv format

Column Description
seqname Sequence name from the input FASTA
start Hit start position (1-based)
end Hit end position (1-based, inclusive)
strand Strand (+ or -)
score Infernal bit score
evalue E-value
query_name Rfam model name (e.g. U1, 5S_rRNA)
accession Rfam accession (e.g. RF00003)
biotype Ensembl-style biotype (e.g. snRNA, snoRNA, rRNA, lncRNA)

annotation/

Final annotation files ready for use in genome browsers and downstream analysis.

Output files
  • annotation/
    • annotation.gtf — GTF v2.2 annotation with gene/transcript/exon features
    • annotation.gff3 — GFF3 annotation with gene/ncRNA/exon hierarchy
    • annotation.bed — 6-column BED file (0-based start coordinates)

GTF format

Ensembl-style GTF with three feature types per hit: gene, transcript, exon. Key attributes:

  • gene_id — sequential identifier (e.g. rfam_gene_000001)
  • gene_name — Rfam model name
  • gene_biotype — Ensembl biotype (e.g. snRNA, snoRNA, rRNA)
  • rfam_accession — Rfam family accession

GFF3 format

Standard GFF3 with gene → ncRNA (or biotype-specific type) → exon hierarchy. Feature types follow Sequence Ontology conventions.

BED format

6-column BED (chrom, chromStart, chromEnd, name, score, strand). Start is 0-based. Score is clamped to 0–1000.

multiqc/

Output files
  • multiqc/
    • multiqc_report.html — standalone HTML report viewable in any browser
    • multiqc_data/ — parsed statistics from all pipeline steps
    • multiqc_plots/ — static images from the report

MultiQC aggregates software versions and pipeline metrics into a single report.

pipeline_info/

Output files
  • pipeline_info/
    • execution_report.html — Nextflow execution report (resource usage per task)
    • execution_timeline.html — timeline of all tasks
    • execution_trace.txt — tab-separated trace of all tasks
    • nf_core_ncrnannotator_software_mqc_versions.yml — software versions used

Biotype reference

The following biotypes are assigned based on Rfam model names and seed classifications:

Biotype Examples
snRNA U1, U2, U4, U5, U6, U11, U12
snoRNA SNORD, SNORA families
scaRNA scaRNA families
rRNA 5S_rRNA, 5_8S_rRNA, SSU_rRNA_eukarya, LSU_rRNA_eukarya
rRNA (prokaryotic, mgnify-assembly only) SSU_rRNA_bacteria, LSU_rRNA_archaea, etc.
tRNA tRNA families
pre_miRNA miRNA precursors
lncRNA Long non-coding RNA
SRP_RNA Signal recognition particle RNA
RNase_P_RNA RNase P families
vault_RNA Vault RNA
Y_RNA Y RNA families
ribozyme Ribozyme families
antisense_RNA Antisense RNA
ncRNA Other non-coding RNA (default)
On this page