sdrf-convert

name	sdrf:convert
description	Use when the user wants to choose an analysis pipeline, check SDRF compatibility with a pipeline, or understand how to go from SDRF to analysis.
user-invocable	true
argument-hint	[pipeline name or experiment description]

SDRF Pipeline Guidance

You are helping the user choose and configure an analysis pipeline from their SDRF.

Supported Pipelines

sdrf-pipelines can convert SDRF to these formats:

Pipeline	Command	Best For
OpenMS	convert-openms	Flexible workflows, custom pipelines
MaxQuant	convert-maxquant	DDA label-free, TMT, SILAC (desktop)
DIA-NN	convert-diann	DIA/SWATH, PlexDIA (fast, scalable)
MSstats	convert-msstats	Statistical analysis (downstream of search)
NormalyzerDE	convert-normalyzerde	Normalization and differential expression
quantms	Nextflow pipeline	Cloud/HPC, complete workflow (uses SDRF natively)

Pipeline Recommendation Logic

Is it DIA data?
├── YES → DIA-NN (fastest, best DIA performance)
│         Also consider: quantms with DIA module
│
└── NO (DDA) →
    ├── Label-free?
    │   ├── Small study (<50 samples) → MaxQuant
    │   ├── Large study (>50 samples) → quantms (scalable)
    │   └── Custom workflow needed → OpenMS
    │
    ├── TMT/iTRAQ?
    │   ├── Standard TMT → MaxQuant or quantms
    │   ├── TMT + phospho → MaxQuant (PTM scoring)
    │   └── Large TMT cohort → quantms
    │
    └── SILAC?
        └── MaxQuant (best SILAC support)

For statistical analysis (after search):
  → MSstats (gold standard for proteomics statistics)
  → NormalyzerDE (normalization comparison)

For cloud/HPC processing:
  → quantms (Nextflow, reads SDRF directly, no conversion needed)

Compatibility Checks

Before recommending, verify SDRF compatibility:

1. Label type: DIA-NN doesn't support TMT (except PlexDIA). MaxQuant supports all.
2. File format: Check if raw files are compatible with the pipeline
3. Modification support: Some pipelines have limited PTM support
4. Column completeness: Each pipeline needs specific SDRF columns

Conversion Commands

The sdrf-pipelines Python package provides the parse_sdrf CLI tool. Install: pip install sdrf-pipelines

# For MaxQuant:
parse_sdrf convert-maxquant --sdrf file.sdrf.tsv --fastafilepath proteins.fasta

# For OpenMS:
parse_sdrf convert-openms --sdrf file.sdrf.tsv --onetable

# For DIA-NN:
parse_sdrf convert-diann --sdrf file.sdrf.tsv

# For MSstats (from OpenMS output):
parse_sdrf convert-msstats --sdrf file.sdrf.tsv --openswathtomsstats

# For NormalyzerDE:
parse_sdrf convert-normalyzerde --sdrf file.sdrf.tsv

# For quantms (Nextflow — reads SDRF directly, no conversion needed):
nextflow run bigbio/quantms --input file.sdrf.tsv --fasta proteins.fasta

Note: quantms reads SDRF natively via its own Nextflow modules. The SDRF is the pipeline input — no parse_sdrf conversion step is needed.

When the User Asks About a Specific Pipeline

1. Explain what the pipeline does and when to use it
2. Check if their SDRF is compatible
3. Show the conversion command
4. Explain what output files will be generated
5. Mention any limitations or common issues