Skip to contents

Wild-type proteins and engineered variants linked to plastic degradation, with the evidence behind each link.

Usage

proteins_list(..., limit = Inf, client = plasticdb_client())

proteins_get(
  accession,
  view = c("summary", "full"),
  include_sequenceless = FALSE,
  follow_merges = FALSE,
  client = plasticdb_client()
)

proteins_interactions(
  accession,
  evidence_level = c("all", "direct", "indirect"),
  level = c("interaction", "experiment", "observation"),
  include_sequenceless = FALSE,
  limit = Inf,
  kind = NULL,
  client = plasticdb_client()
)

proteins_kinetics(accession, limit = Inf, client = plasticdb_client())

proteins_properties(accession, limit = Inf, client = plasticdb_client())

proteins_variants(
  accession,
  include_sequenceless = FALSE,
  limit = Inf,
  client = plasticdb_client()
)

proteins_sequences(..., limit = Inf, client = plasticdb_client())

proteins_fasta(ref, client = plasticdb_client())

proteins_structure(ref, path = NULL, client = plasticdb_client())

proteins_lineage(accession, client = plasticdb_client())

proteins_engineered_parents(
  updated_since = NULL,
  limit = Inf,
  client = plasticdb_client()
)

Arguments

...

Filters (see Details), named as the API names them.

limit

The most records to return (Inf for all of them).

client

A plasticdb_client().

accession, ref

A protein accession or numeric id.

view

"summary" (the default, as on the server since 2.89.0) or "full". The summary carries the record's own fields and counts (how many interactions, variants, kinetics and properties it has) but not those lists, and is much smaller for well-studied enzymes; "full" embeds interactions, variants, kinetics and properties. Use proteins_interactions(), proteins_variants(), proteins_kinetics() and proteins_properties() to read the lists as tibbles.

include_sequenceless

Include records without a sequence.

follow_merges

Return the survivor of a merged accession.

evidence_level

"all" (the default), "direct" (assayed on the protein itself) or "indirect" (credited through its engineered variants) interactions.

level

One row per "interaction", "experiment" or "observation". Deeper levels repeat the interaction's columns and add experiment.* (and observation.*) columns.

kind

Deprecated: the old name of evidence_level. It is sent as evidence_level, never alongside it; different values in both signal plasticdb_argument_error.

path

For proteins_structure(), a file or directory to write to.

updated_since

Only records changed since this time.

Value

Lists return a tibble::tibble() with the provenance (release, source, URL, total) in attr(x, "plasticdb"), see provenance(). Single records return a plasticdb_record list with the API's nesting. proteins_sequences() returns an Biostrings::AAStringSet, proteins_fasta() a string.

Details

  • proteins_list() lists proteins matching the filters in ... (wild types and variants; kind = narrows). Like every entity list, only records with a positive result are listed unless polarity = or include_negative = TRUE says otherwise.

  • proteins_get() returns one protein by accession (PB..., any case) or numeric id, by default as the "summary" view (with counts); view = "full" embeds its interactions, variants, kinetics and properties. An engineered variant's accession returns the variant, as variants_get() does. A merged accession signals plasticdb_merged naming the survivor (merged_into); with follow_merges = TRUE the survivor is returned instead, with a plasticdb_merged_warning. A withdrawn record signals plasticdb_gone (reason "withdrawn"). The per-protein tables of a merged accession signal plasticdb_merged too (its location is the survivor's table); given a variant's accession they signal plasticdb_engineered_variant, since variants have none (its record carries them: variants_get()).

  • proteins_interactions() is the protein's evidence spine: "direct" (assayed on it) or "indirect" (via its organism) interactions, as one row per interaction, experiment or observation (level).

  • proteins_kinetics() and proteins_properties() are the kinetic parameters and stability/activity properties of a protein and its variants; proteins_variants() its engineered variants, with sequences.

  • proteins_sequences() lists proteins and returns their sequences as an Biostrings::AAStringSet (names are accessions, the list columns are in mcols()).

  • proteins_fasta() and proteins_structure() are one protein's FASTA record and primary structure (PDB text, or a file with path); proteins_lineage() a wild type with every variant engineered from it; proteins_engineered_parents() the wild types that have variants. A variant's accession gives the variant's FASTA record or structure. proteins_fasta() answers a pinned release too (a merged accession's redirect is live only; a snapshot answers 404 for a ref it does not hold). The structure, lineage and engineered-parents calls are live only: a pinned client gets plasticdb_release_live_only (for a pinned structure use downloads_structure()).

proteins_list(include_sequence = TRUE) adds each record's sequence column (a variant's own); proteins_sequences() uses it, so it needs no second download.

Filters are the API's own query parameters, with the same meaning on every function: plastic, include_subtypes, microorganism, polarity, include_negative, evidence_method, min_year, max_year, updated_since, sort_by, sort_order and so on. An unknown name signals plasticdb_argument_error naming the valid ones; an unknown value is a server 422 (plasticdb_validation).

Tables have one flat row per record: nested objects become parent.child columns (source_microorganism.name, paper.doi), lists of plain values become list columns, and per-plastic evidence becomes plastics_positive, plastics_negative and plastics_inconclusive.

Examples

pet <- proteins_list(plastic = "PET", kind = "wild_type", limit = 5)
pet[, c("accession", "name", "source_microorganism.name")]
provenance(pet)$total
# \donttest{
petase <- proteins_get("PBFCTW")
petase$name
petase$counts
proteins_interactions("PBFCTW", level = "observation", limit = 3)
proteins_sequences(plastic = "PHB", kind = "wild_type", min_year = 2025)
# }