Built a small open-source tool I wanted to share: FlowProof, an MCP server that lets an AI assistant run Nextflow pipelines and returns a Workflow Run RO-Crate for each run (container digests, tool versions, SHA-256 of every input and output), so AI-driven pipeline runs are reproducible and verifiable rather than a black box. It sits above the engine (in the spirit of nf-prov, but the provenance is retrievable through the same interface that launched the run). Runs locally (uvx flowproof-mcp, your data stays on your machine) or hosted. Repo: github.com/ajibadedapo/flowproof-mcp It's early, and feedback is very welcome, especially on the provenance model and which pipeline you'd want wired next.

📣 Just released nf-prometheus — a plugin that gets your Nextflow runs onto the Prometheus + Grafana stack your HPC site already operates. Built for on-prem clusters: the default export mode is a

.prom

file for the node_exporter textfile collector, so it works with head jobs behind firewalls and on air-gapped systems. Per-process scheduler queue wait, requested-vs-used resources, live run status. Grafana dashboard included.

plugins { id 'nf-prometheus' }

and a 5-minute setup guide in the repo: github.com/mcallisto/nf-prometheus Longer write-up + discussion: community.seqera.io/t/nf-prometheus-self-hosted-prometheus-grafana-metrics-for-nextflow-on-hpc-clusters/2748 Feedback from anyone running on Slurm/PBS/LSF very welcome — especially "it doesn't fit my cluster because X", that's the most useful kind.

Hi, We updated the nft-utils plugin to 1.1.1 This new version improved the

sanitizeOutput()

function with a bug fix and new features. You can now use: •

unstableKeys

with a folder, and get all the files inside the folder (⚠️ this will change existing snapshots where you got only the folder name). •

ignoreKeys

will remove the output entirely from the snapshot (e.g., when the file name isn't stable). •

unstablePatterns

and

ignorePatterns

do the same as their key counterparts but will instead use a glob pattern approach when you want to ignore or set as unstable a particular file or a meta key from the snapshot. •

readsMD5Keys

will replace the regular md5sum by bam(...).getReadsMD5() from nft-bam. • Similarly

variantsMD5Keys

will do the same for vcf files with nft-vcf. You can find more info at nf-co.re/nft-utils#… Thanks to @Maxime Garcia, @Nicolas Vannieuwkerke and @Jim Downie for their help!

We have just launched nf-ccloud on Carolina Cloud. We benchmarked

rnaseq

against HealthOmics and AWS Batch and saw the following results (table below). Experimental details: • Eight samples, 123GiB total zipped FASTQs •

nf-core/rnaseq -profile test_full

View the public demo here. Get up an doing by pasting a single prompt into your LLM (Claude Code, ChatGPT, Pi):

Please read this: <https://console.carolinacloud.io/static/nextflow/local-quickstart.md>, and then help me run my first pipeline, starting with nf-core/rnaseq test, and then graduating to a real pipeline.

Offering generous free trials to those serious about moving their pipelines over. <5mins between signup and triggering first run. DM me if interested. Or, simply claim a cardless $50 trial by signing up at https://console.carolinacoud.io/promo/NFCORE/ to trigger your first run.

Hallo there! A small intro about me before I promote my tool. I spent two decades building software for a safety regulated industry, the last ten years on platform and compliance tooling, where every requirement change must show mechanically what was affected, re-verify it, and keep the evidence available. While learning bioinformatics pipelines, I kept tracing the challenges and the equivalent tooling for traceability and impact analysis. A container turns out to be buggy, a sample fails QC after its data was already used, a reference gets bumped. The command -resume fixes the run, but nothing answers the questions that come before and after it, like which of the past runs used the defective input, which other runs consumed their outputs, and which results already left the building. Nextflow's lineage store and nf-prov record what ran, but turning those records into a decision (what to delete, re-run, or disclose now) was something I couldn't establish. So I attempted to build Clew. It reads the lineage data Nextflow already writes (the native lineage store on 25.04+, nf-prov RO-Crates, or plain work/ symlinks on any version), merges runs into one graph, and computes what to delete, re-run, or disclose for a given trigger, for example a donor who withdrew consent, a sample that failed QC, a container with a bug, a reference genome that got updated. It even traces across pipelines, so a withdrawn sample in an rnaseq run reaches the differentialabundance run that consumed its count matrix. Every verdict cites a rule and replays offline, so the output is something you can hand to an auditor. Two minute demo on a real sarek run, stdlib only, AGPL: pip install clew-lineage then clew demo github.com/QuietFlare/clew I am still new to the domain and learning the tooling and process by building in the open, so I am happy to discuss the scope and how it could work for you. Feedback from people who run pipelines for a living is exactly what I need. Issues and PRs very welcome. Thank you in Advance!

Got ITSxRust (fungal ITS extraction) published in Methods in Ecology and Evolution! Just wanted to celebrate haha. Also, the tool is already integrated in nfcore #CEA7TBJGJ, thanks to @Daniel Straub for that :)) doi.org/10.1111/2041-210x.70393

Hii can I ask someone who speaks Portuguese and it's interesting in talk about nextflow?

set the channel topic: Share work, product announcements, events or other news relevant to Nextflow / nf-core.

I wrote a Tampermonkey (browser extension) script today to add some custom buttons to the GitHub writing toolbar. Some utility features to help make AI-written comments suck less. Let me know if anyone else finds it useful 🙂 github.com/ewels/gfm-tidy

Hi everyone — we’re building GenomQC, a browser-based decision layer for MultiQC RNA-seq reports. It turns MultiQC JSON, TSV or CSV exports into: • explainable PASS / WARN / FAIL sample decisions • a focused review queue • visible thresholds and evidence traces • suggested next actions • JSON and print-to-PDF exports Selected report files are processed locally in the browser and are not uploaded to a GenomQC server. We’re looking for five RNA-seq users or labs to test export compatibility and threshold clarity. We’ll provide a complimentary compatibility check for your existing MultiQC export. Live beta + example report: https://genomqc.pages.dev/ Source: https://github.com/muhammetozcelik/genomqc-rnaseq-qc Feedback and edge cases are very welcome. Research use only; no clinical interpretation or diagnostic reporting.

Hey guys, here's an article showing over 5 GiB/s (that's gigabytes per second) on Cloudflare R2 which should really accelerate the stage-in and stage-out steps of your pipelines. Also, no egress charges. carolinacloud.substack.com/p/pushing-r2-to-its-limits-stress-testing

Hi all - sharing an event in case it's relevant to folks here. We're running the London AI × Science Hackathon on 3–4 October in London with future.bio. 48 hours, teams of 3–5, working on hard open problems in science -> a good chunk of it around protein research and computational biology. Free to attend, backed by Anthropic and Hugging Face (cash prizes, compute and API credits, mentorship from sponsor labs). Figured a few people here might be into it - it leans towards folks who build the tooling rather than just use it. Applications close end of September: luma.com/3iipivod Happy to answer questions here.

🔍 Introducing `nf-diff`: Track and compare Nextflow pipeline runs! Hey everyone! 👋 When refactoring Nextflow pipelines or testing bug fixes, it’s often difficult to verify whether output files, execution metrics, or trace reports changed and by how much. I’ve been working on

nf-diff

, a dedicated Nextflow plugin that compares the output directory structure, metrics, process relationships and generated artifacts between two Nextflow runs. It helps you quickly spot unintended data drift, missing files, or unexpected changes across runs. 🚀 🛠️ What

nf-diff

gives you: • Direct comparison of output directories and execution artifacts generated by Nextflow. • Clear visual summaries of added, modified, or removed outputs and trace data. • Lightweight CLI integration for local testing or custom pipelines. It supports many features and multiple output formats. I’m attaching the HTML report of a

nf-diff

I did here to showcase the plugin capabilities 😄 ⚙️ Bonus — Use it in CI: I’ve also built the

nf-diff

GitHub Action so you can run these output comparisons automatically on Pull Requests and see the results directly as PR comments! 🕸️ Official website: mribeirodantas.github.io/nf-diff/latest 📌 Core Project: github.com/mribeirodantas/nf-diff 💡 GitHub Action Demo & Live PR Examples: github.com/mribeirodantas/demo-nf-pipeline/pulls Give it a spin, star the repo if you find it useful, and let me know if you have any feedback or edge cases to test! Open to contributions! 🙌

It's that time again: Australian researchers 🇦🇺 🐨 , join us for the "Workflows Community Writes nf-tests hacky session" where a small group of us get together to nf-tests. 📅 25/09/2026, 1-4pm (AEST) 💻 Location: Online (Zoom) 🔗 Register with your Australian institutional email address: lnkd.in/gN9U-F24

Hi everyone! I recently decided to start blogging about everything I've been learning/have learnt in bioinformatics, and my first blog's here: priyalt.github.io/2026/10/04/rna-seq-tutorial.html the linkedin post to the same is here: lnkd.in/p/gMczNxns i'd love to hear feedback! please let me know what i can improve upon, thank you. :))

I built a Korean-language guide to Nextflow Summit 2026, including the full agenda in Korea Standard Time. If you have any Korean colleagues, feel free to share it with them! jhlee0637.github.io/nextflow-summit-website-kr lnkd.in/p/gYpmQmtw

Hi everyone! Bioconductor hosts a monthly Developer Forum, and this month we're having Louis Le Nézet presenting on: Bridging R and Nextflow in nf-core modules Integrating R scripts into Nextflow modules can present several challenges: parsing configuration arguments, validating input files, normalizing variables, and ensuring reproducible execution across environments. Within nf-core, these tasks are commonly handled using a template and with the new r-nfcore-utils R package. This session will discuss current nf-core practices for integrating R into workflows, address common challenges, gather developer feedback, and identify ways to improve reproducibility and collaboration. The October Developer Forum will be held on October 29th at 4pm UTC. Registration Link: us06web.zoom.us/meeting/register/… We'd love to see you there! Bioconductor is trying to push for greater interoperability with other scientific software stacks. A recent hackathon effort here has included trying to formalize Nextflow and Galaxy interfacing directly in submitted R packages.

depictio Depictio update: 28 nf-core pipelines now have templates Hi all! 👋 An update on Depictio, which I have posted about here before: an open-source platform from EMBL that turns the output of a workflow run into an interactive dashboard you can share, so nobody has to write a Shiny or Streamlit app for each project. For nf-core pipelines it comes as a dashboard template: point it at any run and you get the same dashboard. There are now templates for 28 nf-core pipelines. 🌐 Showcases (live dashboards) · 📚 Documentation · 💻 GitHub repository ✨ What it does • Your data side by side, with linked filters (filter one, all follow): multiqc MultiQC reports (aggregated across runs), tables (CSV, Excel, Parquet), images, maps (GeoJSON), phylogenetic trees (Newick, Nexus) and 3D structures (PDB, mmCIF), with genome tracks in JBrowse 2 (BAM, VCF, bigWig, BED) and bioimages & spatial omics (OME-Zarr, OME-TIFF, SpatialData) coming soon • No code: build and edit dashboards by drag and drop in the GUI, or in YAML • Reusable: point the template at any new run of the same pipeline and you get the same dashboard • Shared with labs, reviewers or the public, versioned, with comments and annotations on any component • Runs as a Python package, with Docker Compose or on Kubernetes, and can serve as a companion web app for a publication nextflow Run it at the end of a pipeline: already today, a Nextflow config snippet (a

workflow.onComplete

handler) ingests the results of any successful run into Depictio, with nothing to change in the pipeline: documentation. An nf-core module is planned as well; if you maintain a pipeline and would like Depictio as a standard option, tell me and I will open a PR for it. I will post the updates in #C0AFS06G9HR. 🙌 We need your help! Please join #C0AFS06G9HR: pick the template of a pipeline you use, review it and tell me what is useful, what is misleading and what is missing. The dashboards of all the templates, the discussions, the details and my updates are all there. Even one line helps!

Hi everyone! 👋 One of the challenges in computational genomics is moving beyond raw sequence and variant data toward structured representations that researchers can interpret, explore, and evaluate. At Z Genome Lab™, we're exploring a symbolic approach to genomic modeling, combining structured genomic representations, interactive visualization, and traceable research workflows. We're particularly interested in understanding how these approaches could complement existing bioinformatics pipelines and support researchers working with complex genomic data. Our platform is currently open to scientific evaluation, and we'd genuinely value feedback from researchers and bioinformatics core facilities. For teams interested in exploring the approach in a real research setting, we're also offering a no-cost institutional pilot, focused on a jointly defined use case and independent scientific feedback. We're especially interested in hearing your perspectives on: 🔬 The practical value of symbolic representations in genomic analysis. 🧬 Challenges in interpreting and visualizing complex genomic information. 🔄 Requirements for reproducibility and traceability in scientific workflows. 🌐 https://zgenomelab.com⁠� I'd be happy to discuss the methodology and hear your thoughts here

Hi everyone! I've recently built an R package to visualize single-cell GO over-representation analysis results as gene overlap networks. This helps layout your GO results for celltypes/clusters of interest in a clean way that allows you to identify biological themes / hub genes in a streamlined way. It works with Seurat 4 and 5 and integrates seamlessly with clusterProfiler. Feel free to have a look through the repo/vignette. I'm open to any and all feedback if you think this would be useful for your own analyses! github.com/mikemartinez99/scoranet

Dear all, I am happy to share the app - JClub I have been developing for researchers to explore scientific literature quickly in a swipe-up manner. It is available for both iOS and Android. What it does: - A short AI summary for each paper, covering why it matters, what the authors did and what they found, so you can decide in seconds whether to read it. - One tap full-text access through your library, so papers open with your institution's subscription where it has one. - Zotero and Mendeley sync, so papers you save land in your reference manager automatically. - A personal feed of new papers from the journals you choose, ranked by your interests and your own publications via ORCID connect. - Follow authors and articles for new papers matching your keywords, new work from authors you follow, and new citations of papers you follow or of your own work. - Reading groups (Labs) for journal clubs: invite your group with one code, pin the papers you want to discuss, and set a weekly meeting. It's free and you can use it without signing up. Signing in with Google or Apple keeps your journals, saves and alerts in sync across devices. Apple : iOS and Apple Silicon Google: Android i would love to hear feedbacks or a feature request at chirag.parsania@centenary.org.au

👋 Hi all! Sharing something we built, and looking for feedback. bioinformatics-claude-skills: a set of skills for Claude Code (Anthropic's CLI agent) that help you set up and run RNA-seq analyses on a cluster or a single machine. • nf-core setup wizards: rnaseq, scrnaseq, rnavar and rnasplice. They walk you through the options one question at a time and write the samplesheet,

nextflow.config

and launch script. • Downstream analysis: bulk RNA-seq (DESeq2/edgeR/GSEA), Seurat single-cell, and rnasplice reports, as R Markdown + job scripts. • Allele-specific expression pipeline (SLURM only for now). • Cluster settings (queue, modules, container paths, walltime caps) live in one config file, so nothing is hard-coded. SLURM and a plain "local" mode are tested; PBS, SGE and LSF are translated from the docs but not yet run on a real cluster. ⚠️ These are AI-agent instructions, unofficial and not endorsed by nf-core. They generate scripts and configs, so please review everything before you submit anything. What I'd love feedback on: 1. If you run PBS, SGE or LSF: do the generated headers and submit commands work for you? 2. Which pipelines or failure modes should be covered next? 3. Do the default resource requests look sane to you? Repo: github.com/KochInstitute-Bioinformatics/bioinformatics-claude-skills Issues and PRs very welcome. Thanks! 🙏

Hi all! I'm one of the developers of datavzrd, which fits nicely at the end of pretty much any NF analysis: it turns your final tables into interactive and visually appealing HTML reports that you can easily customize with a short YAML config, from plots and heatmaps per column to links between tables. Just like a MultiQC report, the result is self-contained, so one can zip it and send it to collaborators or attach it to a publication as a supplement. I just updated the existing nf-core module, so it's easy to add to your pipelines. If you want to go deeper, here's the paper: doi.org/10.1371/journal.pone.0323079 Happy about any feedback!

Hi everyone! 👋 I wanted to share StageNest Studio, a visual editor for Nextflow pipelines that we’ve been working on. Some of you may have seen an earlier version of this project under a different name. We’ve since rebranded to StageNest Studio, following trademark feedback. Thanks to the community for helping us get that right! 🙏 The idea came from wanting an easier way to build new pipelines and understand/modify existing ones without constantly jumping between files, and to bring analysis scripts together while keeping track of how everything connects. 🔗 You can import a public GitHub repository or ZIP file and explore its processes, subworkflows, and connections. You can also add and configure stages visually, inspect the source, and export the project. Pipeline execution stays on your own infrastructure. So far, it works well with mid-sized projects, including read-processing workflows and modular viral-analysis pipelines such as ViromeXplore. There’s also documentation with screenshots and code examples explaining how the visual elements relate to Nextflow code. 🌐 Website: stagenest.studio 🚀 Getting started: stagenest.studio/docs/getting-started 📖 Documentation: stagenest.studio/docs 💬 I’d be interested to hear how it handles your pipelines, especially anything it gets wrong or makes harder than it should be. Comments here or a DM are welcome!

For those willing to go to Brussels, we once again got a devroom to talk about bioinformatics at FOSDEM ! https://fosdem-bioinformatics.github.io/cfp_2027.html Last year, we got some nice talks from @Nicolas Vannieuwkerke and Jose Espinosa-Carrasco

Hi all! I'm building a tool that looks at a failed Nextflow/Slurm run and tells you the likely cause, the log lines that show it, and a suggested fix. To make it work on real problems, I'm collecting failed runs. If you have one you wouldn't mind sharing, there's a short form here: https://forms.gle/cTrvUdbiw56649G19. Please remove secrets, usernames and internal paths first, and no patient data. Even one or two examples help a lot. Thanks so much! 🙏

📅 Two chances left to join our nf-test community sessions (30th of October and 27th of November)! 🕵️ Testing doesn't have to be a solo activity. Register here: lnkd.in/gN9U-F24

🎨 nf arts nf-arts 1.0.0 is out 🎆 It's a Nextflow plugin that turns each run into generative art. When a run ends, it draws the run's tasks as an SVG or PNG, from their timing, peak memory and status only. Try it:

plugins { id 'nf-arts@1.0.0' }

The run writes

art-<runName>.svg

next to where you launched it. What's in it: • 8 styles: radial, hilbert, ridgeline, phyllotaxis, flowfield, mondrian, tree and ginkgo • 7 palettes, or your own colors •

nextflow plugin nf-arts:render

paints runs that already finished, from your local history or from a trace file, including nf-core megatests on S3 • A caption with the pipeline, version and run name, which

title = false

turns off • No pipeline data in the image: no task names, sample IDs, inputs, outputs or paths Gallery of real megatests: <https://github.com/maxulysse/nf-arts/blob/main/docs/gallery.md> Registry: <https://registry.nextflow.io/plugins/nf-arts> Repo: <https://github.com/maxulysse/nf-arts> Feedback and style ideas welcome, as issues or PRs 🌻

I hope your Halloween is filled with more treats than tricks, and more laughs than scares!

Ligase, type-safe computational biology pipelines I'd like to share an open-source project I've been working on. Ligase is a computational biology pipeline engine built around a Rust core, with Python bindings. The idea is to address a class of problems in scientific computing where pipelines execute successfully but produce unreliable results because of inconsistencies in their biological data. Ligase introduces explicit biological data types and compatibility checks to help catch these issues before computation. Some of its key features include: • Type-safe biological data: Explicit compatibility constraints across pipeline operations. • Reproducible execution: Content-aware caching and tool-version tracking. • Provenance: Traceable computational workflows and method documentation. • Scientific validation: Documented benchmarking against established reference results. The project is currently pre-alpha, and I'm looking to get feedback from people working in bioinformatics, genomics, computational biology, and research software engineering. Links: • GitHub: https://github.com/Neoforge-Labs/ligase • Documentation: https://ligase.neoforgelabs.tech/ I'd appreciate feedback, questions, and perspectives on the architecture and design decisions. Always interested in learning how others approach these challenges.