BulkSeq Studio — FAQ & cite

Help

FAQ & troubleshooting

Short answers to the questions that come up most often. Each one points to a tab or button in the app.

Do I need to know how to code?

No. The whole pipeline runs from a tabbed graphical interface, from project setup through metadata, reference selection, and the Outputs browser. No command line is required.

What do I need on Windows?

Windows 10 or 11 (x64) and WSL2, the Linux subsystem the pipeline runs inside. The app can enable WSL2 and install the bioinformatics environment for you from its Environment setup screen. Enabling the WSL feature is the one step that needs administrator rights (Windows requires it) and may ask for a reboot; everything after that runs without elevation.

The first-run environment setup is slow. Is that normal?

Yes, and it is one-time. The core bioinformatics environment downloads over several minutes, and the R / DESeq2 stack is the largest, slowest install. You can keep creating projects, editing metadata, and generating configs while it installs; only starting a pipeline run needs the tools in place.

Does it work on Linux without WSL?

Yes. On Linux the same GUI drives the same Snakemake pipeline natively in a local micromamba environment, with no WSL involved, and produces identical results. There is no WSL setup step and no memory cap below the host's RAM.

How do I update to a new version?

On Windows, run the newer installer: it detects the existing install and offers to update (remove the old version and install the new one) or to uninstall first. The portable ZIP has nothing to update — unzip the new one. On Linux, download the newer AppImage and delete the old one, or run AppImageUpdate on the AppImage; the build carries zsync update information, so it fetches only the changed chunks from the latest release.

Which trimmer, rRNA tool, and DE engine should I use?

The defaults are chosen to work for a standard study: fastp for trimming, SortMeRNA if you turn on rRNA removal, and DESeq2 for differential expression. Change them only when you have a reason. Prefer Trim Galore or Trimmomatic if your lab standardizes on one; use RiboDetector when no rRNA reference is available for your organism; and run limma-voom or edgeR quasi-likelihood as a cross-check, especially on larger designs. All the alternatives produce the same tables and figures, so downstream steps do not change.

My organism has no Bioconductor annotation package (a crop or fungus). Can I still get enrichment and networks?

Yes. Without an OrgDb you still get KEGG pathway ORA and GSEA for any organism with a KEGG code, plus GO through g:Profiler (gprofiler2), and the STRING protein-interaction network is unaffected. An organism is not skipped for lacking an OrgDb.

Can I run single-end reads?

Yes. Single-end and paired-end input both run the full alignment route through every trimmer, aligner, and rRNA tool; the app detects the layout from the sample sheet. A single run must use one layout — a mixed-layout sample sheet is rejected with guidance so the two are not counted together.

How much disk space and memory do I need?

About 10 GB of free disk for the toolchain and reference indices, and 16 GB or more of RAM is recommended. STAR alignment is the memory-intensive step; for very large genomes use the HISAT2 or Salmon aligner, which need far less.

My custom gene sets returned nothing. Why?

Almost always an identifier-namespace mismatch: the gene IDs in your GMT or annotation table must match the run's ID format. A mismatch is flagged, with examples of the run's IDs. Use Ensembl gene IDs for human and mouse, FlyBase for Drosophila, and NCBI GeneIDs (bare or LOC<GeneID>) for NCBI-RefSeq crops such as rice.

Where do my results go, and how do I re-make just the figures?

Everything lands in the project folder; open it from Open Project Folder on the Run Monitor tab. To restyle figures, use the built-in Figure Style editor (palette, fonts, sizes, dimensions, DPI) and click Regenerate figures — this re-renders them without re-running alignment or DESeq2.

A console window flashed on screen during a run. Should I close it?

No. On Windows a WSL/terminal window may briefly appear because the app launches Snakemake inside WSL2. Closing it kills the running pipeline. Use the Stop button in the app to cancel a run instead.

If a run pauses and asks to stop or continue That is the low-mapping safeguard: a sample aligned poorly (uniquely-mapped rate below the threshold), usually a wrong reference or contamination. Check your reference and sample sheet before choosing to continue.

Cite & reuse

Cite & license

BulkSeq Studio is free and open-source software released under the MIT License. You can install it, use it, study the code, and share it at no cost. The accompanying manuscript is in preparation for submission to a peer-reviewed journal.

How to cite

If you use BulkSeq Studio or its benchmark archive in your work, please cite the software and the reproduction archive deposited on Zenodo. The archive (DOI 10.5281/zenodo.20955660) holds the benchmark scripts, the pinned scoring environment, the per-benchmark result tables and figures, and a step-by-step reproduction guide for the validation suite.

Birgün, Tuna (2026). BulkSeq Studio: benchmark apparatus and
  reproduction archive. Version 0.16.0. Zenodo.
  https://doi.org/10.5281/zenodo.20955660

BibTeX The repository ships a CITATION.cff file. GitHub renders a ready-to-copy citation from it, and most reference managers can import the same file directly.

@software{birgun_bulkseq_studio_2026,
  author  = {Birg{\"u}n, Tuna},
  title   = {BulkSeq Studio: benchmark apparatus and
             reproduction archive},
  version = {0.16.0},
  year    = {2026},
  doi     = {10.5281/zenodo.20955660},
  url     = {https://doi.org/10.5281/zenodo.20955660},
  license = {MIT}
}