WfMS Citation Analysis

Regenerates Supplementary Figure 1 from Langer et al. 2025, Genome Biology, "Empowering bioinformatics communities with Nextflow and nf-core" — per-year citation counts for bioinformatics workflow management systems.

The original figure (and methodology) is in the published supplementary file (DOCX).

Scopus (primary)

Note: Two DOIs are mis-indexed in Scopus and have been excluded (Galaxy 2012, Pachyderm — see Notes). Four further DOIs are missing (preprints, Figshare, ACM conference proceedings), so 17/23 papers contribute.

	Absolute	Percentage	Data
Without Galaxy	PNG / SVG	PNG / SVG	CSV
With Galaxy	PNG / SVG	PNG / SVG	CSV

Other sources

OpenAlex	Dimensions	iCite
OpenAlex is slow to update, so the most recent year's citation counts will typically lag behind reality for several months after publication.	Snakemake counts are underrepresented: Dimensions splits F1000Research paper revisions into separate records, and only v1 of the Snakemake 2021 paper (311 citations) has per-year data — v2 (1,643 citations) has no per-year chart. See Notes.	Limited to PubMed-indexed papers, so 4 papers without PMIDs (preprints, Figshare, some conference proceedings) are missing.
CSV: with Galaxy / no Galaxy — all plots	CSV: with Galaxy / no Galaxy — all plots	CSV: with Galaxy / no Galaxy — all plots

Quick start

# 1. Fetch citation data from OpenAlex (free API, ~30 seconds)
uv run fetch_openalex.py

# 2. Fetch citation data from Dimensions (browser scraping, ~5 minutes)
uv run fetch_dimensions.py

# 3. Fetch citation data from NIH iCite (free API, ~30 seconds)
uv run fetch_icite.py

# 4. (Optional) Fetch citation data from Scopus (requires institutional API key)
#    See scopus_instructions.md for setup. Must run from the institution's network.
export SCOPUS_API_KEY=your_key
uv run fetch_scopus.py

# 5. Generate plots and CSVs for all sources
uv run generate_plots.py

Data sources

Source	Per-year data	API key needed	Method	Coverage
OpenAlex	Yes	No	REST API	23/23 papers
Dimensions	Yes	No	Browser scraping (Highcharts extraction from badge.dimensions.ai)	22/23 papers
iCite	Yes	No	NIH iCite API (PubMed-based)	19/23 papers
Scopus	Yes	Yes (institutional)	Scopus Citation Overview API (fetch_scopus.py)	17/23 papers
CrossRef	Totals only	No	REST API	—
Google Scholar	Totals only	No	scholarly library (very slow, gets rate-limited)	—
Semantic Scholar	Unreliable	No	Free API returns incomplete citation subsets	—
Altmetric	No	Yes (403)	N/A	—

OpenAlex, Dimensions, and iCite are free and need no API key. Scopus is included for parity with the original paper but needs an institutional API key with Citation Overview access — see scopus_instructions.md. OpenAlex has the best coverage.

Papers included

Galaxy

DOI	Label
10.1093/nar/gkae410	Galaxy 2024 update
10.1093/nar/gkac247	Galaxy 2022 update
10.1093/nar/gkaa434	Galaxy 2020 update
10.1093/nar/gky379	Galaxy 2018 update
10.1093/nar/gkw343	Galaxy 2016 update
10.1186/gb-2012-13-10-r86	Galaxy 2012
10.1186/gb-2010-11-8-r86	Galaxy 2010
10.1101/gr.4086505	Galaxy 2005

Nextflow

DOI	Label
10.1038/s41587-020-0439-x	nf-core framework (2020)
10.1038/nbt.3820	Nextflow enables reproducible workflows (2017)
10.1186/s13059-025-03673-9	Empowering bioinformatics communities (2025)
10.1101/2024.05.10.592912	Same paper, preprint version

Snakemake

DOI	Label
10.12688/f1000research.29032.2	Sustainable data analysis with Snakemake (2021)
10.1093/bioinformatics/bts480	Snakemake workflow engine (2012)

CWL (shown separately from Other)

DOI	Label
10.1038/nbt.3772	Toil
10.6084/m9.figshare.3115156.v2	CWL v1.0

Other

DOI	Label
10.1016/j.jbiotec.2017.07.028	KNIME reproducible
10.1145/1656274.1656280	KNIME 2.0
10.1007/978-3-030-28954-6_1	KNIME data analysis
10.1093/bioinformatics/bts167	Bpipe
10.1093/bioinformatics/btx152	Pachyderm
10.1093/gigascience/giz044	SciPipe
10.1101/201178	Cromwell/GATK4

Plot description

• Stacked area charts with Nextflow (#2DC09C) on top, then Galaxy (blue), Snakemake, CWL, and Other in grey shades
• Two variants: with Galaxy as a separate category, and without Galaxy
• Two scales: absolute citation counts and 100% stacked (percentage)

Notes

• Source comparison: All four sources agree on the overall trend. OpenAlex has the best coverage (23/23 papers). Dimensions is missing CWL v1.0 (Figshare DOI, not indexed). iCite is missing 4 papers without PMIDs (preprints, Figshare, some conference proceedings). Scopus is missing 4 DOIs (preprints, Figshare, ACM proceedings) and has two mis-indexed records (see below), giving 17/23 effective coverage.
• F1000Research versions: Dimensions treats each F1000Research revision as a separate record with its own DOI. fetch_dimensions.py handles this by fetching all versions and summing their citations. Currently only v1 of Snakemake 2021 renders a per-year chart; v2 has the majority of citations but no chart.
• Year range: Edit YEAR_MIN / YEAR_MAX in each script to extend the range.
• Adding papers: Add new entries to the PAPERS list in all four fetch_*.py scripts, then re-run all five scripts.
• Dimensions scraping: The fetch_dimensions.py script extracts data from Highcharts charts on badge.dimensions.ai. If their page structure changes, the JS extraction (window._Highcharts) may need updating. As a fallback, use the Playwright MCP in Claude Code to manually navigate and extract data (see below).
• iCite: Only covers PubMed-indexed papers. Uses the citedByPmidsByYear field which gives exact per-year counts. Some papers need PMID overrides (DOI lookup fails) — these are hardcoded in fetch_icite.py.
• Scopus DOI mismatches: Scopus returned the wrong record for two DOIs — 10.1186/gb-2012-13-10-r86 (Galaxy 2012) resolved to a histone PTM paper, and 10.1093/bioinformatics/btx152 (Pachyderm) resolved to HiC-spector. Both have been excluded from the Scopus summary in scopus/citation_data.json (their by_year data is removed; the entries are kept with an error field). If Scopus fixes this in future, re-run fetch_scopus.py.

Manual Dimensions extraction (Claude Code fallback)

If fetch_dimensions.py breaks, use the Playwright MCP interactively:

1. Navigate to https://badge.dimensions.ai/details/doi/{DOI}
2. Accept cookies if prompted
3. Click the "Citations" tab
4. Extract data with JS: window._Highcharts.charts.filter(c=>c)[0].series[0].data.map(p=>({year:p.category,citations:p.y}))
5. Repeat for each DOI

License

Code and data in this repository are licensed under the Apache License 2.0.