First release of nf-core/genomeqc, which compares the quality of multiple genomes and their annotations. The pipeline runs in two modes depending on the inputs provided: Genome only (FASTA files) and Genome and Annotation (FASTA plus GTF/GFF files).

Added

• Input genomes and annotations from local files or NCBI accessions, downloaded with NCBI genome download.
• Genome completeness assessment with BUSCO, including a BUSCO Ideogram plotting the location of markers on the assembly.
• Read-based genome completeness evaluation with Merqury (optional).
• Telomeric repeat identification and visualisation with tidk (optional).
• Assembly contiguity and integrity statistics (N50, N90, GC%, number of sequences) with QUAST.
• Contamination screening with FCS-GX, FCS-adaptor, and Tiara, shown as contamination stats in the tree summary plot.
• Annotation statistics with AGAT, including a Gene Overlaps analysis counting overlapping genes.
• Extraction of the longest protein isoform with GffRead.
• Orthologous gene inference with OrthoFinder (v2 or v3).
• Repeat/transposable element annotation with RepeatModeler and RepeatMasker (or HITE).
• BUSCO marker-based and orthology-based phylogenetic Tree Summary plots combining assembly and annotation summary statistics, with multiple layout styles including a circular layout with configurable quality-threshold rings.
• An executable Shiny app for interactively adjusting the tree plot and summary statistics, with PNG/SVG export.
• HTML and Excel summary reports.
• Aggregated quality-control report with MultiQC.
• Genome-mode and protein-mode BUSCO/validated-GFF outputs are now published to separate paths so a run using both modes no longer has one mode’s files silently overwrite the other’s.

Fixed

• Fixed BUSCO not showing in the tree plot.
• Fixed the Shiny app launcher pulling its container from Docker Hub instead of quay.io, which caused a pull access denied error.
• Circular layout: quality statistics (sequence count, N50, BUSCO complete/duplicated) are now scored as a colour-vision-safe traffic light (Good/Warn/Poor) against phylogenetic-group thresholds (--quality_preset), rather than a sequential colour ramp that misleadingly implied “dark = good”. --show_ring_values can print each value on its ring.
• Renamed ORTHOLOGOUS_CHROMOSOMES to ORTHO_SEQ_COUNT (#196), and its outputs accordingly, since the input assembly need not be chromosome-level.
• Fixed NCBIGENOMEDOWNLOAD failing under singularity/apptainer with a container-image error.
• Fixed ORTHOFINDER silently succeeding when it failed to produce Orthogroups/Orthogroups.tsv.
• Fixed ORTHO_SEQ_COUNT mapping zero genes for GFFs using a transcript feature instead of mRNA (common in AUGUSTUS output) (#174).
• Fixed ORTHOFINDERV2 crashing under -profile conda with a misleading “out of RAM” message.
• Fixed -profile conda failing to resolve for RM_DOWNLOAD_DB and REPEATMODELER_REPEATMODELER (HITE remains docker/singularity-only under conda).
• Fixed HTML_REPORT/EXCEL_REPORT crashing on samplesheets that mix genome-only and annotated assemblies.
• Fixed gene_overlaps.R silently dropping genes with unresolved strand, and never detecting overlaps between genes on opposite strands, undercounting overlap statistics.
• Fixed AGAT sp_statistics results being effectively absent from both reports; added a feature-first “Annotation stats” tab/sheet and an “Assembly stats” tab to the HTML report.
• Fixed a stale filename reference in the Shiny app that silently broke the ortho-seqs panel.
• Fixed the circular tree layout’s rings (--tree_style circular) rendering with wildly uneven thickness, and --show_ring_values’ printed values landing on the wrong ring.
• Fixed the conventional tree layout’s TE column title rendering below its panel instead of above it.
• Fixed the FCS-GX panel/ring being drawn against the wrong species, and occasionally showing a fabricated contamination grade for species with no FCS-GX data, whenever FCS-GX only ran for some samples.
• Fixed SVG export failing in the interactive Shiny app.
• Fixed --RM_download_db true never actually downloading a DFAM partition, so the parameter had no effect.
• Fixed GENOMEANNOTATIONBUSCOIDEOGRAM overwriting outputs when a species is assessed against more than one BUSCO lineage (--busco_lineage auto).
• Fixed Merqury outputs (.qv, .completeness.stats, spectra plots) never being published to results/merqury/<species>/.

Dependencies

Deprecated