Fetching metadata (-m)

Download only the sample metadata for an accession and skip the sequence data.

adaptiseq -i PRJNA211801 -m
adaptiseq -i CRR343031 -m

Metadata is always fetched (with or without -m) because it drives resolution; if metadata cannot be retrieved, adaptiSeq stops before downloading. The metadata bytes are pulled with wget, so the files are exactly what the archive serves.

SRA / ENA / DDBJ / GEO

adaptiSeq first queries ENA. If sample information is available it downloads metadata in TSV format via the ENA Portal API (typically ~191 columns), saved as ${accession}.metadata.tsv.

If ENA has no record yet (common for very recently released SRA data), adaptiSeq falls back to the SRA Database Backend, downloading CSV (~30 columns) and converting it to TSV for consistency.

GSA

For GSA accessions, adaptiSeq uses GSA's getRunInfo interface to download CSV metadata (~25 columns), saved as ${accession}.metadata.csv, and the exportExcelFile interface for the parent Project's XLSX (sheets Sample, Experiment, Run), saved as CRA*.metadata.xlsx.

In the Python API

get_metadata returns the parsed rows as a list of dicts (and writes the same files):

from adaptiseq import get_metadata

rows = get_metadata("SRR7706354")          # ENA/SRA -> list[dict] of TSV columns
rows = get_metadata("CRR311377")           # GSA     -> list[dict] of CSV columns

See the Python API page.

Output

Database	Files
SRA / ENA / DDBJ / GEO	${accession}.metadata.tsv
GSA	${accession}.metadata.csv, CRA*.metadata.xlsx

Note NCBI E-utilities is rate-limited to 3 req/s without a key and 10 req/s with one. Set NCBI_API_KEY (and optionally NCBI_EMAIL) in the environment to use the higher limit when resolving large batches.