nf-core/longraredisease

What nf-core/longraredisease is

nf-core/longraredisease is a bioinformatics pipeline engineered for the detection and clinical interpretation of structural variants using long-read sequencing technologies, specifically Oxford Nanopore Technologies and PacBio platforms. It is primarily targeted at rare disease diagnostics and identifying pathogenic variants in patients with neurodevelopmental disorders.

The pipeline was developed within the nf-core framework with support from clinical scientists and researchers including Chipo Mashayamombe-Wolfgarten, Hannah Titheradge, Lorraine Hartles-Spencer, Professor Andrew Beggs, Friederike Hanssen, Ken Brewer, Nicolas Vannieuwkerke, and Maxime U Garcia. It addresses the challenge of resolving complex genomic variants in rare disease cohorts by combining consensus variant calling with clinical annotation and phenotype-driven ranking.

By incorporating pedigree relationships and clinical features codified as Human Phenotype Ontology terms, the workflow enables researchers and clinical diagnostic teams to triage genomic variants effectively and pinpoint candidate de novo or inherited alterations.

What the pipeline does

The pipeline processes long-read alignments provided in BAM format. Structural variants are detected across multiple callers—Sniffles, CuteSV, and SVIM—and merged into a high-confidence consensus set using JASMINE. Variant calls are phased across haplotypes using LongPhase. For family trios, the workflow carries out joint calling and identifies de novo mutations based on sample pedigree structures specified in the input metadata.

Once structural variants are identified, they undergo clinical annotation through AnnotSV integrated with disease databases. Phenotype-driven variant prioritization is then executed using SVANNA, which ranks candidates based on user-supplied Human Phenotype Ontology terms. In addition to core structural variant discovery, the workflow can optionally be configured to detect single nucleotide variants with Clair3 or DeepVariant, copy number variants via Spectre or HiFiCNV, short tandem repeats using Straglr, and DNA methylation patterns with Modkit.

Key facts

What you can use it for

Neurodevelopmental disorder diagnosis

Identify structural variants and rank their clinical relevance in patients presenting with neurodevelopmental phenotypes.

Trio-based de novo variant discovery

Perform joint family analysis on parent-child trios to isolate de novo structural variations underlying rare genetic conditions.

Phenotype-prioritised variant screening

Score and rank genomic alterations against clinical manifestations using patient-specific Human Phenotype Ontology terms via SVANNA.

Comprehensive long-read variant profiling

Optionally evaluate single nucleotide variants, copy number changes, short tandem repeats, and base methylation in tandem with structural variants.

Run it here

The research agent runs nf-core/longraredisease with Nextflow on its own machine — up to 32 cores, 256 GiB of memory and 24 hours per run, which covers the test profile and real single-sample runs. It builds the samplesheet, checks it against the pipeline's schema and reads the results back. When a run is too big for one machine, it gives you the exact command for your own cluster instead.

Strengths and limits

Questions

How to cite it

10.5281/zenodo.20935122