FetchPGSCatalogRule

For each requested pgs.catalog.model spec the rule:

• fetches full score metadata from GET /rest/score/{pgs_id}/;
• selects the best available scoring file (GRCh38 harmonised -> GRCh37 harmonised -> original) and downloads it via pg.download();
• parses the gzipped TSV, normalising column names for both harmonised and original file layouts;
• splits REST metadata into $gwas (discovery phenotype / sample info) and $generation (PRS method / publication provenance);
• returns a PGS whose build is the canonical name (genomeBuilds$name()) of the downloaded file's build; a file written from it declares GRCh37 or GRCh38.

Reactive rule that materialises one pgs.catalog.model resource from the PGS Catalog REST API.

PolyGenius::Rule -> FetchPGSCatalogRule

Public methods

• FetchPGSCatalogRule$new()
• FetchPGSCatalogRule$matches()
• FetchPGSCatalogRule$inputs()
• FetchPGSCatalogRule$requirements()
• FetchPGSCatalogRule$run()
• FetchPGSCatalogRule$clone()

Method new()

Register this rule under the name "fetch.pgs.catalog".

Usage

FetchPGSCatalogRule$new()

Method matches()

Accept only pgs.catalog.model output specs.

Usage

FetchPGSCatalogRule$matches(output.spec)

Arguments

output.spec — Resource output spec list with a $type field.

Returns

TRUE when type is resources.type$pgs.catalog.model.

Method inputs()

This rule has no upstream dependencies -- PGS Catalog is fetched directly.

Usage

FetchPGSCatalogRule$inputs(output.spec)

Arguments

output.spec — Resource output spec (unused; kept for interface parity).

Returns

Empty list.

Method requirements()

Declare lightweight resource requirements. Scoring files can be large (>50 MB harmonised) but processing is single-threaded and memory-light.

Usage

FetchPGSCatalogRule$requirements(output.spec, inputs)

Arguments

output.spec — Resource output spec (unused).

inputs — Empty list from inputs() (unused).

Returns

Named list: cores = 1, memory = 1.

Method run()

Orchestrate metadata fetch, scoring file download, parsing, and PGS construction for one PGS Catalog accession.

Usage

FetchPGSCatalogRule$run(output.spec, inputs, logger = pg.logger())

Arguments

output.spec — Resource output spec; $params$pgs.id gives the accession.

inputs — Empty list (no upstream resources required).

logger — Rule-scoped logger for trace/info messages.

Returns

Named list: spec, data (PGS), meta, logs.

Method clone()

The objects of this class are cloneable with this method.

Usage

FetchPGSCatalogRule$clone(deep = FALSE)

Arguments

deep — Whether to make a deep clone.

Other source-rules: FetchGWASCatalogRule, FetchOpenGWASRule, LoadLocalGWASRule