Raw Data Expansion Roadmap

Published

Jun 2026

This guide has focused on a results-first proteomics workflow.

The starting point was a processed proteomics result table, not raw mass spectrometry files.

This chapter explains how the system can be expanded in the future to support raw data processing, from mass spectrometry files to protein-level differential abundance results.

Current Scope

The current edition starts here:

processed proteomics result table
        ↓
table inspection
        ↓
metadata and comparison design
        ↓
differential abundance understanding
        ↓
quality control
        ↓
filtering and ranking
        ↓
identifier cleaning
        ↓
functional enrichment preparation
        ↓
STRING preparation
        ↓
visualization
        ↓
biological interpretation
        ↓
reproducible report

This workflow is useful when the analyst receives result tables from:

proteomics core facility
collaborator
vendor
published supplementary data
previous internal analysis

Future Scope

A raw-data edition would begin earlier.

raw mass spectrometry files
        ↓
raw-data organization
        ↓
metadata validation
        ↓
database search or DIA processing
        ↓
peptide-spectrum matching
        ↓
protein inference
        ↓
protein quantification
        ↓
normalization
        ↓
differential abundance analysis
        ↓
results-first workflow

The current system can become the downstream interpretation layer of a larger proteomics workflow.

Raw Proteomics Data Types

Raw proteomics workflows may include several file types.

Examples include:

.raw
.mzML
.mzXML
.mgf
.tsv
.txt
.csv

Vendor-specific raw files may need conversion before open workflows can use them.

For example:

Thermo .raw
        ↓
mzML
        ↓
search or quantification workflow

Common Processing Stages

A raw-data proteomics system may include the following stages.

raw data inventory
        ↓
file conversion
        ↓
database preparation
        ↓
search engine execution
        ↓
peptide identification
        ↓
protein inference
        ↓
protein quantification
        ↓
quality control
        ↓
statistical analysis
        ↓
differential protein abundance table

The final output from this upstream workflow would feed directly into the current Results-First Edition.

Possible Software Tools

A future raw-data expansion could support tools such as:

MaxQuant
Proteome Discoverer
FragPipe
MSFragger
DIA-NN
OpenMS
MSstats
limma
DEP
Perseus-style tables

The system does not need to support all tools at once.

A practical expansion should begin with one clear workflow and one example dataset.

Data-Independent and Data-Dependent Acquisition

Raw proteomics processing depends on the acquisition strategy.

Common strategies include:

DDA
DIA
TMT or other labeling workflows
label-free quantification
spectral-library workflows
library-free workflows

Each approach has different processing requirements.

For that reason, raw-data expansion should be modular rather than one-size-fits-all.

Proposed Raw-Data Module Structure

A future raw-data module could use this structure:

raw-proteomics-processing/
├── 01-raw-data-organization.qmd
├── 02-mass-spec-metadata.qmd
├── 03-file-conversion.qmd
├── 04-database-and-search-setup.qmd
├── 05-peptide-identification.qmd
├── 06-protein-inference.qmd
├── 07-protein-quantification.qmd
├── 08-normalization-and-qc.qmd
├── 09-differential-abundance-analysis.qmd
└── 10-handoff-to-results-first-workflow.qmd

The handoff output would be a table compatible with this guide.

Handoff Table

The most important connection between raw processing and results-first interpretation is a standardized handoff table.

A minimal handoff table should contain:

protein_id
gene_symbol
protein_name
log2fc
p_value
adjusted_p_value
comparison

Additional useful columns include:

sample intensities
group means
peptide counts
unique peptide counts
protein coverage
missingness statistics
contaminant flags
protein group information
normalization method
statistical model

The goal is to make upstream processing and downstream interpretation communicate clearly.

Raw-Data QC Expansion

A raw-data edition should add QC before differential abundance analysis.

Possible QC outputs include:

raw file inventory
sample metadata validation
MS run summary
identification rate
peptide counts
protein counts
missing value profile
intensity distributions
PCA or clustering
batch effect checks
contaminant review
normalization diagnostics

These QC steps are upstream of the current result-table QC chapter.

Reproducibility Requirements

A raw-data workflow should record:

software versions
database version
FASTA file
search parameters
enzyme specificity
missed cleavages
fixed modifications
variable modifications
FDR thresholds
quantification method
normalization method
statistical model
contrast definitions

These details are essential for reproducibility.

Containerization and Environment Control

Raw proteomics processing is more software-sensitive than result-table interpretation.

A future edition should consider:

conda environments
Docker or Apptainer containers
version-pinned tools
workflow managers
checksums
locked reference databases

This would make the workflow easier to rerun later.

Possible Workflow Managers

For larger datasets, raw processing may benefit from a workflow manager.

Possible options include:

Nextflow
Snakemake
WDL/Cromwell
bash-based staged workflows

For CDI-style teaching systems, a staged bash workflow may be easiest at first.

A scalable production workflow could later be implemented in Nextflow or Snakemake.

Recommended Expansion Strategy

A practical expansion strategy is:

start with one raw-data type
        ↓
choose one processing tool
        ↓
use one small public dataset
        ↓
produce a standardized differential abundance table
        ↓
handoff to the current results-first workflow

This keeps the expansion manageable.

Phase 1: Results-First Edition

The current guide completes Phase 1.

processed tables
        ↓
interpretation-ready outputs
        ↓
reproducible report

This is valuable because many real projects begin with processed output tables.

Phase 2: Quantification Table Edition

The next expansion could start from a protein quantification matrix.

protein intensity matrix
        ↓
sample metadata
        ↓
normalization and QC
        ↓
statistical modeling
        ↓
differential abundance table
        ↓
results-first workflow

This would sit between raw processing and the current guide.

Phase 3: Raw Mass Spectrometry Edition

The full expansion would start from raw files.

raw mass spectrometry files
        ↓
identification and quantification
        ↓
protein abundance matrix
        ↓
differential abundance analysis
        ↓
results-first workflow

This would make the proteomics system end-to-end.

Relationship to CDI Systems

This roadmap follows the same CDI system logic:

clear inputs
        ↓
structured workflow
        ↓
reproducible outputs
        ↓
interpretation
        ↓
reporting

The Results-First Edition is the interpretation layer.

The raw-data edition would add upstream processing.

Final Workflow Summary

The completed Results-First Edition can be summarized as:

Proteomics result tables
        ↓
Sample metadata and comparison design
        ↓
Differential abundance result understanding
        ↓
Result-table quality control
        ↓
Filtering and ranking
        ↓
Identifier cleaning and annotation preparation
        ↓
Functional enrichment preparation
        ↓
STRING network preparation
        ↓
Visualization
        ↓
Biological interpretation
        ↓
Reproducible report

Final Note

This guide provides a practical foundation for proteomics interpretation when processed result tables are already available.

That is a common real-world situation.

The raw-data expansion can be added later without weakening the current system.

Instead, the current system becomes the stable downstream layer that future raw-data workflows can hand off to.