One open-source ecosystem, from raw spectra to biological answers.
The search space defines everything downstream. Conduit builds it, then carries the run the rest of the way.
Bring your DIA runs and a sample sheet.
Decide which proteins could be there.
Identify, quantify, annotate. Same result on any machine.
Taxonomy, function and statistics, code-free.
Eight ways to answer the first question: what proteins could be here?
Only for the command line. Basecamp runs on macOS, Windows, and Linux and manages the Linux environment Ascent needs — natively, in a VM, or on a remote workstation over SSH. Running Ascent directly needs a Linux workstation, lab server, or SLURM cluster.
A minimum of 128 GB RAM and 24 cores. Broad peptidotyping passes or big genome catalogs can need far more, so plan on an HPC node for those. Raw Astral files are around 7 GB each, and reference databases for some strategies reach 50–200 GB.
Snakemake (≥ 7.0) and Apptainer (≥ 1.1). Everything else — R, Python, MetaPhlAn, Bakta, eggNOG-mapper — ships in pinned containers. DIA-NN is downloaded separately under its own licence: free for academic use, while commercial use needs a licence from its authors.
The genome strategies can pull a curated MGnify genome catalog — genomes and taxonomy included — by setting genome_download_source: mgnify and mgnify_catalog: human-gut/v2.0.2.
Every Conduit component is released under the MIT License; third-party tools such as DIA-NN keep their own (notices). Conduit is supported by the UCSD Health Sciences Collaborative Center for Proteomics.
