fun_proteomics

Analyze raw Bruker DDA-PASEF, DIA-PASEF and conventional DDA BAF data, or import a protein quantification table. Get protein and gene matrices, all-versus-all comparisons, quality control, and service detection benchmarks. Every matrix is available as CSV and TSV.

Install

Linux, Python 3.11 or newer:

pip install --upgrade fun_proteomics
fun_proteomics install --conda
fun_proteomics doctor

Installation and existing runtimes · PyPI · Walkthrough videos

Quick start with real data

Download a small, published ubiquitin-affinity LFQ dataset: 12 samples, four conditions, three experimental replicates per condition. The archive includes protein quantities, sample assignments, original source files, checksums and data attribution.

wget https://github.com/cfarkas/fun_proteomics/releases/download/v0.9.0/ubiquitin-lfq-example.tar.gz
wget https://github.com/cfarkas/fun_proteomics/releases/download/v0.9.0/ubiquitin-lfq-example.tar.gz.sha256
sha256sum -c ubiquitin-lfq-example.tar.gz.sha256
tar -xzf ubiquitin-lfq-example.tar.gz

Play the interactive walkthrough below. Start the browser once with fun_proteomics setup, then choose the extracted files and a new output/project folder, such as ubiquitin-study. The video shows file selection, condition assignment, all-pairs analysis, CSV/TSV downloads and continuation.

This recording uses the downloaded real data and runs the complete matrix analysis. It needs no raw spectrum search. Step-by-step quick start, terminal and scripted versions.

Set up your analysis

Choose one of the following launch methods. --project is your output folder; setup --outdir is an equivalent option. Use a new project for a new analysis.

A) Interactive settings

Follow the video above. Open the browser with fun_proteomics setup, then set your output folder in the page. Keep the terminal open. In the local browser:

1. Choose a matrix, raw-data folder or previous run, and review the output folder.
2. Click Discover samples. Drag sample cards into condition columns. Add or rename conditions; move excluded samples to Unassigned. A sample table can prefill assignments.
3. Choose independent samples or paired matrix/BAF comparisons with donor IDs. Keep all-versus-all comparisons; four conditions produce six pairs at each feature level.
4. Click Save configuration and check, then Run / continue analysis.
5. Open Results or Matrices · CSV & TSV in the same page.

Use the Browse buttons to select files on the computer running the analysis. Input formats, technical injections and SSH browser access.

B) Terminal questions

fun_proteomics setup --terminal --project /data/my-study --run

Answer the questions about input files, conditions, replicate design and output folder. This uses the same analysis as the browser and works over SSH. Omit --run to save and review the configuration first.

C) Full command-line options

Supply every input explicitly to run without questions:

fun_proteomics setup --non-interactive --project /data/my-study \
  --mode matrix --matrix /data/proteins.csv --samples /data/samples.csv \
  --independent-samples --normalization none \
  --cpus 8 --memory-gib 32 --run

Example input files (illustrative intensity values):

samples.csv — one row per sample:

sample_id,condition
C1,Control
C2,Control
T1,Treatment
T2,Treatment

proteins.csv — one row per protein group:

Protein.Group,Genes,C1,C2,T1,T2
P04637,TP53,120000,135000,260000,245000
P60709,ACTB,480000,505000,490000,510000
P02768,ALB,90000,85000,,100000

Each sample_id must exactly match a quantity column in proteins.csv. Use non-log-transformed intensities; leave missing measurements blank. Genes is optional: include gene symbols here or supply a matching FASTA with GN= annotations for gene-level results.

For paired matrix/BAF data, add a donor column to the sample table and replace --independent-samples with --subject-column donor. Add --service-matrix /data/provider.tsv to benchmark a matching service table. All command-line options and raw DIA/DDA/BAF examples.

provider.tsv — optional service quantities using the same sample names:

Protein.Group	Genes	C1	C2	T1	T2
P04637	TP53	120000	135000	260000	245000
P60709	ACTB	480000	505000	490000	510000

To check and run an existing saved configuration:

fun_proteomics check --config /data/my-study/config/run.json
fun_proteomics run --config /data/my-study/config/run.json

All three methods save config/run.json. Browser sample assignments are also saved as CSV and TSV.

{
  "project": "/data/my-study",
  "mode": "matrix",
  "matrix": "/data/proteins.csv",
  "samples": "/data/samples.csv"
}

Results and continuation

Open my-study/START_HERE.html for results or my-study/MATRICES.html for direct CSV/TSV downloads. Detailed outputs live in my-study/results/final_results/.

fun_proteomics --continue /data/my-study

Completed native results are verified and reused. Native BAF checks converted spectra and complete searches. Raw PASEF workflows resume their Nextflow session. Integrity checks and recovery.

Guides

MIT licensed. Maintained by cfarkas. The public example retains its upstream data license and citation.