The packaged analysis backend supports Windows through WSL2 and native Linux. macOS CI exercises the GUI and configuration layer, but there is no supported macOS installer or verified native analysis backend.
Before you start, identify your operating system and arrange enough local storage for tools, reference files, reads and results. Storage requirements depend on the study; the tool installation alone is not a complete disk budget.
1. Download a release#
Open the official release page. The versioned release lists x86-64 Windows installer and portable ZIP packages plus a Linux x86-64 AppImage, its zsync companion and a portable tarball. Select the package for your platform and confirm that its version is published before installing it.
The Windows GUI installer accepts Windows 10 version 1809 (build 17763) or later, but its automatic wsl --install setup needs Windows 10 version 2004 (build 19041) or later, or Windows 11, as stated in Microsoft’s WSL installation requirements. The GUI installation floor does not mean WSL2 local analysis works on Windows 10 version 1809; see Microsoft’s manual WSL2 requirements for older Windows builds.
On Linux, make the downloaded AppImage executable with chmod +x FILENAME.AppImage. The AppImage needs glibc 2.38 or newer: its bundled graphics and audio libraries require GLIBC_2.38 symbols, so Ubuntu 22.04, Debian 12 and Red Hat Enterprise Linux 9 are too old for it.
2. Complete Environment setup#
Open the app and let the first-run Environment setup check the GUI packages, WSL2 on Windows, core bioinformatics environment and R/DESeq2 stack. It names the phase it is running — the WSL distribution, the bioinformatics tools, then the R and Bioconductor load test, which can take about two minutes on the first run after an update. Work through Action needed entries, then click Re-check. Show details / log exposes installation failures. Its persistent setup log is stored in your user application-data directory rather than the installed or portable application folder, so it remains available when bundled files are read-only.
Setup reports its result in Show details / log as WSL bioinformatics installer finished with exit code N. Exit codes and statuses says what each code means and what to do.
Choices that matter#
| Setting | What it changes | When to change it / example |
|---|---|---|
| Installer or portable package | Changes how the application is installed, not your scientific model. | Use portable when you need an extracted application directory; keep projects separate from application files. |
| Full analysis environment | Adds the R/Bioconductor analysis and figure tools. | A ready interface alone does not establish that differential-expression tools are ready. |
| WSL2 on Windows | Runs Linux analysis tools behind the desktop window. | Enabling the Windows feature may require administrator rights and a reboot. Later environment installation runs in your WSL user account. |
The full environment specification declares six fewer R packages than 0.30.1 — biomaRt, ComplexHeatmap, EnhancedVolcano, ggtext, ggraph and tidygraph — because no workflow script loads them. A new installation is that much smaller; an environment installed earlier keeps them until it is rebuilt, and no analysis output depends on either state. The running application version is shown at the right of the status bar, so a bug report can name it without opening About.
3. Create a project#
Continue to Project and sample sheet. You may prepare metadata while tools install, but starting a pipeline requires its applicable tools.
Advanced: Windows resource limits
WSL2 may have a lower RAM cap than Windows. Use Edit WSL2 memory / CPU limits on Compute resources and apply the requested restart only after other WSL work is safely stopped. See compute resources; do not copy a large memory allocation from another computer.
Terms on this page: WSL2.