SeqDesk · original analysis
A curated cohort of nf-core (Nextflow) and Snakemake workflows, each tracked by its live GitHub usage, alongside the citation adoption curves of the workflow technologies themselves (Nextflow, Snakemake, Galaxy, Toil, CWL, Bpipe). Counts and usage come from the nf-core and GitHub APIs; citation histories from the OpenAlex graph. Re-counted weekly.
Each line is the cumulative citations of a workflow technology’s paper — an adoption proxy that, unlike GitHub stars, runs back over a decade. Numbered pins mark when each engine’s paper was published (keyed to the list below). Nextflow now leads with 4,401 citations and nf-core is rising fastest (+2.8K in the last three years), while Galaxy — long among the most-cited engines — has plateaued on this citation measure. nf-core is dashed because it is a curated framework layered on Nextflow rather than a standalone engine. The current year (2026) is still accruing citations and is excluded. Hover a line or legend chip to isolate it.
Each bar is a pipeline’s GitHub stargazers — the one usage signal available for every workflow in both ecosystems. nf-core lists 156 pipelines and the Snakemake catalogue draws on 822 tagged repositories; this board profiles the 25 most prominent. Stars approximate visibility, not the number of pipeline runs.
How many registered workflows each engine carries, counted comparably on WorkflowHub (2026-09) — the one registry that indexes workflows of every type. This is why the engines above have no GitHub-stars leaderboard: unlike nf-core and Snakemake, whose pipelines are individual starred GitHub repositories, Galaxy and CWL pipelines are registered as WorkflowHub workflows. Galaxy dominates here because its community registers on WorkflowHub, while most nf-core pipelines live on nf-co.re — so these counts reflect registration habits, not total ecosystem size.
| Technology | Engine family | Paper | Lifetime citations | Peak year |
|---|---|---|---|---|
| Nextflow | Nextflow | 2017 | 4,401 | 2024 (1.1K/yr) |
| nf-core | Nextflow (framework) | 2020 | 4,319 | 2025 (1.3K/yr) |
| Galaxy | Galaxy | 2010 | 3,560 | 2016 (458/yr) |
| Snakemake | Snakemake | 2012 | 3,207 | 2021 (587/yr) |
| Toil | Toil | 2017 | 1,476 | 2023 (284/yr) |
| CWL | CWL | 2022 | 191 | 2024 (48/yr) |
| Bpipe | Bpipe | 2012 | 182 | 2017 (26/yr) |
scripts/check-pipeline-trends.mjs| Pipeline | Ecosystem | Domain | Stars | Forks | Citations |
|---|---|---|---|---|---|
| nf-core/rnaseq | nf-core | RNA-seq | 1.4K | 893 | — |
| nf-core/sarek | nf-core | Variant calling | 599 | 546 | 270 |
| rna-seq-star-deseq2 | Snakemake | RNA-seq | 370 | 213 | — |
| nf-core/scrnaseq | nf-core | Single-cell RNA-seq | 350 | 229 | — |
| nf-core/mag | nf-core | Metagenomics | 320 | 155 | 120 |
| nf-core/ampliseq | nf-core | Amplicon / metabarcoding | 260 | 156 | 400 |
| nf-core/chipseq | nf-core | ChIP-seq | 242 | 182 | — |
| nf-core/nanoseq | nf-core | Nanopore | 228 | 108 | — |
| nf-core/atacseq | nf-core | ATAC-seq | 226 | 140 | — |
| nf-core/eager | nf-core | Ancient DNA | 213 | 91 | 176 |
| nf-core/fetchngs | nf-core | Data retrieval | 199 | 94 | — |
| nf-core/methylseq | nf-core | Methylation | 197 | 181 | — |
| nf-core/taxprofiler | nf-core | Metagenomics (taxonomy) | 194 | 68 | — |
| nf-core/rnafusion | nf-core | RNA fusion | 174 | 125 | — |
| nf-core/viralrecon | nf-core | Viral genomics | 165 | 157 | — |
| nf-core/raredisease | nf-core | Rare disease | 123 | 70 | — |
| nf-core/oncoanalyser | nf-core | Cancer genomics | 120 | 38 | — |
| grenepipe | Snakemake | Variant calling | 120 | 26 | — |
| benchpress | Snakemake | Benchmarking | 84 | 22 | — |
| tibanna | Snakemake | Cloud execution | 73 | 29 | — |
| snakemake-econ-r | Snakemake | Reproducible research | 64 | 14 | — |
| pypsa-fes | Snakemake | Energy modelling | 32 | 3 | — |
| MeSS | Snakemake | Metagenomics (simulation) | 32 | 3 | — |
| LyRic | Snakemake | Long-read RNA-seq | 23 | 6 | — |
| amoebae | Snakemake | Comparative genomics | 21 | 2 | — |
The pipeline counterpart to the tool-trends card: not which method paper wins by citations, but which end-to-end workflow the community runs and which engine it is adopting — re-counted weekly from the nf-core, GitHub, OpenAlex and WorkflowHub APIs. Only aggregate counts are published. · back to all research data