pdc2qpx¶

One-shot download + conversion of PDC/CPTAC studies into QPX datasets.

pdc2qpx orchestrates these steps in a single command:

Download each study's CDAP .psm files (and, with --include-spectra, the mzML spectra) from PDC via pridepy.
Convert the .psm files to QPX psm/feature/pg/dataset/ontology/provenance views (CdapConverter).
Build sample/run views from PDC GraphQL metadata (on by default) — this recovers the TMT/iTRAQ channel → biological-sample mapping that .psm files lack. Disable with --no-metadata.
With --include-spectra, convert the mzML files to a full-spectra mz.parquet, linked to PSM/feature by run_file_name + scan.

The download layer requires the optional pdc extra:

pip install qpx[pdc]

Usage¶

qpxc pdc2qpx -a PDC000109 --download-dir ./downloads --output-folder ./qpx/PDC000109 [OPTIONS]

-a/--accession accepts a single ID, comma-separated IDs, or a CSV with a pdc_study_id/pdc_id column (the same input forms as pridepy download-pdc-files -a). With more than one study, each is written to <output-folder>/<study>/ and a failed study (e.g. one with no CDAP .psm) is reported without aborting the rest — unless --stop-on-error is set.

Parameters¶

Option	Description	Default
`-a`, `--accession`	PDC study accession(s): single ID, comma-separated, or a CSV with `pdc_study_id`/`pdc_id`	required
`--download-dir`	Directory for downloaded files (written under `<download-dir>/<study>/`)	required
`--output-folder`	QPX output dir. One study writes here; multiple studies write to `<output-folder>/<study>/`	required
`--include-spectra`	Also download mzML and produce a full-spectra `<study>.mz.parquet`	off
`--no-metadata`	Skip building `sample`/`run` views from PDC metadata	off (views built)
`--ms-levels`	MS levels for the mz view (e.g. `2` or `1,2`)	all
`--max-cpus`	Threads for the CDAP conversion	24
`--max-memory`	Memory limit for the CDAP conversion	16GB
`--download-threads`	Parallel HTTP Range threads per file	24
`--skip-download`	Use files already present under `<download-dir>/<study>/`	off
`--stop-on-error`	With multiple studies, abort on the first failure	off (continue)

Examples¶

One study, entire QPX including full spectra¶

For quantms-style reanalysis you need MS1 (precursor LFQ quantification) as well as MS2 (identification), so include all MS levels (the default when --ms-levels is omitted):

qpxc pdc2qpx -a PDC000109 \
    --download-dir ./downloads \
    --output-folder ./qpx/PDC000109 \
    --include-spectra --max-cpus 24

Many studies from a CSV¶

Each study is written to ./qpx/<study>/; studies without CDAP .psm are skipped and reported at the end.

qpxc pdc2qpx -a cptac_lfq.csv \
    --download-dir ./downloads \
    --output-folder ./qpx --max-cpus 24

Re-run on already-downloaded files¶

qpxc pdc2qpx -a PDC000109 \
    --download-dir ./downloads \
    --output-folder ./qpx/PDC000109 \
    --include-spectra --skip-download

Output¶

Per study, prefixed with the study accession (under --output-folder for a single study, or <output-folder>/<study>/ for many):

<study>.psm.parquet, <study>.feature.parquet, <study>.pg.parquet
<study>.sample.parquet, <study>.run.parquet (unless --no-metadata) — biological sample metadata and the channel → sample mapping from PDC
<study>.dataset.parquet, <study>.ontology.parquet, <study>.provenance.parquet
<study>.mz.parquet (only with --include-spectra) — full spectra, joinable to PSM/feature by run_file_name + scan

Notes¶

Only PDC studies that publish CDAP .psm files can produce the base QPX views; --include-spectra additionally requires mzML to be available on PDC for that study.
sample/run views come from PDC GraphQL metadata (studyExperimentalDesign + biospecimenPerStudy + fileMetadata), not from the .psm files. A metadata fetch failure is logged and skipped without breaking the rest of the conversion.
Downloads go through pridepy's PDC layer (signed CloudFront URLs, md5 validation, automatic 403 URL re-signing).
See Converters for the underlying qpxc convert cdap and qpxc convert mz commands.