mokume

A comprehensive proteomics quantification library for the quantms ecosystem.

The name comes from mokume-gane (木目金), a Japanese metalworking technique that fuses multiple metal layers into distinctive patterns — similar to how this library melds peptide intensities into unified protein expression profiles.

---

• Multiple Quantification Methods

  ---

  iBAQ, TopN, MaxLFQ, DirectLFQ, Sum, Ratio — choose the right method for your experiment.

  Quantification methods

• Flexible Normalization

  ---

  Feature-level, sample-level, hierarchical, and TMM normalization with a unified pipeline.

  Normalization

• Batch Correction

  ---

  Remove technical variation while preserving biological signal using ComBat.

  Batch correction

• IRS for Multi-plex TMT

  ---

  Internal Reference Scaling with automatic reference detection from SDRF.

  IRS normalization

• Preprocessing Filters

  ---

  Comprehensive QC filters configurable via YAML or CLI.

  Preprocessing

• Differential Expression

  ---

  LimROTS, DEqMS, and proDA with BH or IHW FDR correction — choose by discovery vs precision priority.

  Differential Expression

• Tissue Proteome Atlas

  ---

  Build per-dataset tissue atlases with AdaTiSS tissue-specificity scoring, AnnData outputs, and atlas plots.

  TissueMap workflow

• One-Step Pipeline

  ---

  Go from feature parquet to protein intensities in a single command.

  Quick start

---

Choose Your Workflow

• Standard LFQ / TMT quantification — start with features2proteins
• Need more control before protein summarization — use the two-step path via features2peptides and peptides2protein
• Tissue atlas analysis — use tissuemap

Quick Example

CLIPython

# MaxLFQ quantification with normalization
mokume features2proteins \
    -p features.parquet \
    -o proteins.csv \
    -s experiment.sdrf.tsv \
    --quant-method maxlfq

from mokume.pipeline import QuantificationPipeline, PipelineConfig
from mokume.pipeline.config import InputConfig, QuantificationConfig

config = PipelineConfig(
    input=InputConfig(parquet="features.parquet", sdrf="experiment.sdrf.tsv"),
    quantification=QuantificationConfig(method="maxlfq"),
)
pipeline = QuantificationPipeline(config)
proteins = pipeline.run()

---

Part of the quantms Ecosystem

mokume is a core component of the quantms proteomics analysis platform, providing the quantification engine that powers protein-level analysis, normalization, and tissue atlas workflows from mass spectrometry data.

Ecosystem Tool	Purpose
quantms	Nextflow pipeline for quantitative proteomics
qpx	Data format and conversion tools
mokume	Protein quantification, normalization, and tissue atlas analysis