Input Files¶

SDRF (Recommended)¶

The required input is an SDRF file describing your experiment:

nextflow run bigbio/quantmsdiann \
    -profile docker \
    --input experiment.sdrf.tsv \
    --database uniprot.fasta \
    --outdir results/

The SDRF file contains sample metadata, raw file paths, and experimental conditions in a standardized format. quantmsdiann currently supports SDRF inputs only when they use the .sdrf.tsv extension.

Supported Raw Formats¶

Format	Extension	Instrument
Thermo RAW	`.raw`	Orbitrap, Exploris
Bruker .d	`.d`	timsTOF
mzML	`.mzML`	Any (pre-converted)

Automatic conversion

Thermo .raw files are automatically converted to mzML via ThermoRawFileParser. Bruker .d directories are converted via tdf2mzml. Pre-converted .mzML files are used directly.

FASTA Database¶

Standard UniProt FASTA protein sequences. DIA-NN generates decoy sequences automatically if needed.

# Download human proteome
wget https://ftp.uniprot.org/pub/databases/uniprot/current_release/knowledgebase/reference_proteomes/Eukaryota/UP000005640/UP000005640_9606.fasta.gz
gunzip UP000005640_9606.fasta.gz

Spectral Library (Optional)¶

For library-based DIA analysis, provide a spectral library:

nextflow run bigbio/quantmsdiann \
    -profile docker \
    --input experiment.sdrf.tsv \
    --database uniprot.fasta \
    --spectral_library library.tsv \
    --outdir results/

If no library is provided, DIA-NN runs in library-free mode using in-silico predicted spectra.