Depictio Templates
Templates are pre-packaged project configurations that set up a complete bioinformatics analysis project, dashboards included, with a single command.
Available templates¶
16S · ITS · CO1 · 18S amplicon sequencing across Illumina, PacBio, and IonTorrent.
Viral assembly and variant calling — SARS-CoV-2 and other genomes via the nf-core reference config.
Benchmark variant callers against truth sets: precision, recall and F1 for germline small variants, somatic indels and structural variants.
DESeq2 differential expression: per-contrast statistics, gene annotation and the variance-stabilised sample space.
Functional screening of contigs: AMR genes, antimicrobial peptides, biosynthetic gene clusters and CAZymes.
B and T cell repertoire analysis: sequence processing, repertoire composition and clonal analysis.
Gene-fusion detection: caller consensus, per-caller evidence, in-silico validation and splice-junction analysis.
Bulk RNA-seq quantification: MultiQC funnel, Salmon expression overview and a per-gene explorer.
Metagenomic taxonomic profiling across many classifiers, with concordance and confidence views.
Transcription-factor and histone ChIP-seq: MACS3 peaks, HOMER annotation and one consensus set per antibody.
Chromatin accessibility: ataqv library quality, MACS2 broad peaks and the consensus set across libraries.
CUT&RUN / CUT&Tag chromatin profiling: SEACR and MACS2 side by side, their agreement and the consensus set per target.
Ancient DNA: read preprocessing, mapping, damage patterns and contamination estimates to judge authenticity per library and sample.
Hi-C: valid pair statistics, contact maps and compartment and TAD calls, from library QC to 3D genome structure.
Metagenome-assembled genomes: assembly, binning, completeness and contamination, and taxonomy of the recovered bins.
DNA methylation: bisulfite conversion QC, M-bias, CpG coverage and methylation levels compared across samples and groups.
Oxford Nanopore long reads: read length and quality, alignment, and transcript quantification per sample.
Germline and somatic variant calling: alignment QC, variant counts per caller, annotated variants and their genome positions.
Single-cell RNA-seq: barcode ranks, cells called, genes per cell and embeddings of the filtered cells per sample.
Ribosome profiling: footprint lengths, P-site periodicity, ORF calls and translational efficiency against matched RNA-seq.
Small RNA-seq: adapter trimming, read length profiles, miRNA quantification and the contaminant fractions per sample.
De novo genome assembly: contiguity, completeness and assembly QC across assemblers and polishing steps.
Immunopeptidomics: identified MHC peptides, length and anchor-motif signatures, replicate reproducibility and source proteins across runs.
Run demultiplexing: reads per lane and sample, index balance, undetermined barcodes and per-sample quality.
Alternative splicing: differential splicing events and transcript usage between conditions, with the supporting expression.
Protein families: how the input clustered, the families built or extended, and each family's alignment, conservation, structure and profile HMM.
Protein annotation: domains from HMM libraries and InterProScan, predicted secondary structure, and structures folded on demand.
Structure prediction: model confidence per engine, pLDDT along the chain, the predicted aligned error and where the engines agree.
Cancer genomics: QC, purity and ploidy, copy number, structural variants, drivers, protein changes on the structure and immune context.
Gene editing: read accounting, edit outcomes per library and per guide, and the indels and substitutions around the cut site.
Iso-Seq: CCS and primer QC, the full-length read funnel, isoform categories and structures per gene.
Pairwise genome alignment: coverage and identity of each query on the target, divergence, assembly contiguity and synteny.
RNA outlier detection: aberrant expression, splicing and mono-allelic expression in rare disease cohorts.
MHC binding prediction for variant, protein and peptide inputs.
Genome assembly and annotation QC across genomes: BUSCO, assembly statistics and orthology.
HLA class I typing with OptiType from DNA or RNA reads.
Nascent transcription time courses (GRO-seq, PRO-seq): signal along the genome and differential transcription.
Label-free DIA proteomics quantification: identifications, protein abundances and QC.
Rare disease family analysis: ranked variants and inheritance models.
RNA-seq germline variant calling: variants, annotation and alignment QC.
Most stable reference genes across public expression datasets.
Viral genome reconstruction from metagenomic reads: assemblies, consensus genomes and QC.
No template matches that search.
Status levels¶
Generated and not yet reviewed. Expect it to need fixes before it is usable.
Shared as-is. Feedback and PRs welcome.
Tested, CI passes, reviewed by the Depictio team or community.
Validated by the pipeline lead developer. Highest trust level.
How templates work¶
A template bundles:
- Project configuration: workflows, data collections and cross-DC links, with
{VAR_NAME}placeholders - Recipes: Python transforms that turn raw pipeline outputs into dashboard-ready tables
- Dashboard YAML: imported on the first run, with the template variables substituted
- Cross-DC links: interactive filtering across data collections
- Recipe seeds: each recipe's output committed as a
.tsv, so the bundled project can be explored, and re-ingested, without the pipeline run behind it (v1.6.0+)
Pipeline version and template version¶
A template carries two version numbers. They move independently.
-
Pipeline version
The pipeline release whose outputs the template reads. It is the last part of the template id:
nf-core/rnaseq/3.26.0. A new pipeline release gets a template of its own.Choose it with
--template, or with the Pipeline version picker on a template's page. -
Template version
The revision of the template itself: its data collections, recipes and dashboards, for one pipeline release. It moves when the template improves, for example
2.0.0to2.1.0.Find it beside the picker on a template's page, in the ingest log, and in the project and dashboard settings.
When the template version moves, what you need to do depends on which number changed:
Major 2.1.0 → 3.0.0
The data collections changed. A collection, a column or a template variable was added, removed or renamed.
Ingest the run again.
Minor 2.0.0 → 2.1.0
The dashboards changed. Figures, tabs or recipes moved; every collection keeps its columns.
Import the dashboards again.
Patch 2.1.0 → 2.1.1
Only text changed. Titles, captions and descriptions.
Nothing to do; import again to get the new wording.
Depictio records the template version a project was ingested with, and the one each dashboard was imported from. The viewer warns when a newer one is available.
Contributing a template¶
Want to add a template for another pipeline? The contributing guide covers the directory layout, recipe requirements, and review process.