Package {krt}

Title: Author, Validate, and Export Key Resources Tables
Version: 0.1.0
Description: A toolkit for creating, importing, validating, enriching, rendering, and depositing Key Resources Tables (KRTs). A KRT lists the resources used and generated in a study (antibodies, cell lines, organisms, chemicals, software, datasets, protocols, and more), each paired with a persistent identifier such as a Research Resource Identifier (RRID), a Digital Object Identifier (DOI), a repository accession, or a catalog number, so that resources are unambiguously identifiable and machine-actionable. The package models resources as typed, validated records around a neutral core schema and maps them to journal or funder output profiles, following the FAIR (Findable, Accessible, Interoperable, Reusable) principles of Wilkinson et al. (2016) <doi:10.1038/sdata.2016.18>. It normalizes and optionally resolves identifiers against public registries, extracts resources from manuscripts, and renders tables both in the STAR (Structured, Transparent, Accessible Reporting) Methods style used by Cell Press journals and in the style required by ASAP (Aligning Science Across Parkinson's), with an emphasis on transparency, reproducibility, and correct per-component licensing.
License: GPL-3
URL: https://github.com/choxos/krt, https://choxos.github.io/krt/
BugReports: https://github.com/choxos/krt/issues
Depends: R (≥ 4.1)
Imports: digest, httr2, jsonlite, methods, stats, tools, utils, xml2, yaml
Suggests: bslib, covr, DT, knitr, officer, openxlsx, pdftools, rdflib, rmarkdown, S4Vectors, shiny, spelling, testthat (≥ 3.0.0), tibble
VignetteBuilder: knitr
Config/testthat/edition: 3
Encoding: UTF-8
Language: en-US
Config/roxygen2/version: 8.0.0
LazyData: true
NeedsCompilation: no
Packaged: 2026-09-03 11:34:08 UTC; choxos
Author: Ahmad Sofi-Mahmudi ORCID iD [aut, cre], Aligning Science Across Parkinson's [cph] (Copyright holder of the ASAP Key Resources Table schema (CC BY 4.0; Zenodo doi:10.5281/zenodo.17917979) from which the bundled ASAP profile is derived. ASAP does not endorse this package.)
Maintainer: Ahmad Sofi-Mahmudi <a.sofimahmudi@gmail.com>
Repository: CRAN
Date/Publication: 2026-09-12 14:20:08 UTC

krt: Author, Validate, and Export Key Resources Tables

Description

krt is a toolkit for working with Key Resources Tables (KRTs). A KRT lists the resources used and generated in a study (antibodies, cell lines, organisms, chemicals, software, datasets, protocols, and more), each paired with a persistent identifier so that resources are unambiguously identifiable and machine-actionable.

Details

The package is built around a neutral, typed core object krt_tbl and a registry of output profiles (generic, ASAP, STAR Methods, and custom) that the core maps to. The author-facing table is a view; the underlying record set stays structured, typed, and losslessly round-trippable through JSON and YAML.

Start with new_krt() and add_resource(), validate with validate_krt(), normalize identifiers with normalize_ids(), and write output with export_krt() or render_krt(). See the package vignettes and https://choxos.github.io/krt/ for details.

Author(s)

Maintainer: Ahmad Sofi-Mahmudi a.sofimahmudi@gmail.com (ORCID)

Authors:

Other contributors:

See Also

Useful links:

The KRT S4 class

Description

An S4 wrapper around a krt_tbl, for interoperability with Bioconductor. Convert with as(x, "KRT") and as(y, "krt_tbl"), or as_krt().

Slots

schema_version,profile,table_id,title

Character metadata.

resources,approvals,contributors,provenance

Record lists.

metadata

A list of remaining metadata (study type, locale, timestamps).

Examples

k4 <- methods::as(krt_example, "KRT")
methods::is(k4, "KRT")

Add, update, remove, or get a resource in a KRT

Description

Add, update, remove, or get a resource in a KRT

Usage

add_resource(x, ...)

update_resource(x, resource_id, ...)

remove_resource(x, resource_id)

get_resource(x, resource_id)

Arguments

x

A krt_tbl.

...

For add_resource(), either a single krt_resource (from new_resource()) or the arguments of new_resource() (resource_type, display_name, and named fields).

resource_id

The id of the resource to update, remove, or get.

Value

add_resource(), update_resource(), and remove_resource() return the modified krt_tbl; get_resource() returns a krt_resource or NULL.

Examples

k <- new_krt("Demo")
k <- add_resource(k, "Software/code", "R", version = "4.4.0",
                  new_or_reuse = "reuse", rrid = "RRID:SCR_001905")
get_resource(k, k$resources[[1]]$resource_id)

Append a provenance entry to a KRT

Description

Records that an activity (for example "normalize_ids", "import", "validate", "resolve", "export") was applied to the table. Called automatically by the mutating functions; exported so custom pipelines can record their own steps.

Usage

append_provenance(x, activity, inputs = NULL, outputs = NULL, params = NULL)

Arguments

x

A krt_tbl.

activity

A short activity label.

inputs, outputs

Optional character vectors of input/output identifiers.

params

Optional named list of parameters for the activity.

Value

The krt_tbl with the entry appended.

Examples

k <- append_provenance(new_krt("Demo"), "import", params = list(format = "csv"))
length(krt_provenance(k))

Map one resource to a profile's columns

Description

Map one resource to a profile's columns

Usage

apply_mapping(resource, profile)

Arguments

resource

A krt_resource.

profile

A krt_profile or profile name.

Value

A named list of column values.

Examples

r <- new_resource("Antibody", "Anti-TH", vendor = "Millipore",
                  catalog_number = "AB152", rrid = "RRID:AB_390204",
                  new_or_reuse = "reuse")
apply_mapping(r, "asap")[["IDENTIFIER"]]

Coerce a KRT to a data frame (a rectangular view)

Description

Produces a rectangular view of the table by filling the union of present fields with NA. Many-valued fields are collapsed with "; " in this view; the lossless representation is the JSON/YAML export.

Usage

## S3 method for class 'krt_tbl'
as.data.frame(x, row.names = NULL, optional = FALSE, ..., view = "wide")

Arguments

x

A krt_tbl.

row.names, optional

Present for consistency with the as.data.frame() generic; not used.

...

Ignored.

view

The view to produce: "wide" (the full union of core fields) or a profile name such as "asap" (a profile projection). Must be named.

Value

A data frame with one row per resource.

Examples

k <- add_resource(new_krt("Demo"), "Dataset", "RNA-seq",
                  new_or_reuse = "new", doi = "10.5281/zenodo.123")
as.data.frame(k)
as.data.frame(krt_example, view = "asap")

Convert a validation report to a data frame

Description

Convert a validation report to a data frame

Usage

## S3 method for class 'krt_validation_report'
as.data.frame(x, ...)

Arguments

x

A krt_validation_report.

...

Ignored.

Value

A data frame with one row per finding.

Coerce to and from the S4 KRT class

Description

These are inverse coercions named after their target class. as_krt() returns the S3 krt_tbl (the package's primary object): pass it a krt_tbl and it is returned unchanged, or an S4 KRT and it is converted down. It is the helper to call when a function should accept either representation. as_KRT() is the opposite direction, a thin idempotent wrapper over methods::as(x, "KRT") for Bioconductor workflows.

Usage

as_krt(x)

as_KRT(x)

Arguments

x

A krt_tbl or KRT object.

Value

as_krt() returns a krt_tbl; as_KRT() returns an S4 KRT.

Examples

k4 <- as_KRT(krt_example)
identical(length(as_krt(k4)$resources), length(krt_example$resources))

Export the provenance graph as PROV-JSON

Description

Serializes the table's recorded activities as a W3C PROV-JSON document: the table is an entity, each recorded step is an activity associated with the krt software agent and generating the table.

Usage

as_prov_json(x, path = NULL, audience = c("author", "public"))

Arguments

x

A krt_tbl.

path

Output path, or NULL to return the JSON string.

audience

"author" (default) keeps provenance parameters; "public" redacts the table first, so recorded parameters that could echo sensitive input values are dropped.

Value

The JSON string, or the path (invisibly).

Examples

cat(substr(as_prov_json(normalize_ids(krt_example)), 1, 40))

Export a KRT as RDF (JSON-LD or Turtle)

Description

Export a KRT as RDF (JSON-LD or Turtle)

Usage

as_rdf(
  x,
  format = c("jsonld", "turtle"),
  path = NULL,
  audience = c("author", "public")
)

Arguments

x

A krt_tbl.

format

"jsonld" (default) or "turtle" (requires the rdflib package).

path

Output path, or NULL to return the text.

audience

"author" (default) or "public" (redacts the table first).

Value

The serialized RDF text, or the path (invisibly).

Examples

invisible(as_rdf(krt_example, format = "jsonld"))

Export a KRT as an RO-Crate 1.1 metadata document

Description

Describes the table as an RO-Crate Dataset whose parts are its resources, each carrying its name, type, and identifiers.

Usage

as_rocrate(x, path = NULL, audience = c("author", "public"))

Arguments

x

A krt_tbl.

path

Output path, or NULL to return the JSON-LD string.

audience

"author" (default) or "public" (redacts the table first).

Value

The JSON-LD string, or the path (invisibly).

Examples

cat(substr(as_rocrate(krt_example), 1, 40))

Coerce a field value to its declared type.

Description

Light, non-destructive coercion used when building resource records. Enum fields are lower/normalized where the vocabulary is lowercase; unknown enum values are passed through unchanged (validation reports on them separately).

Usage

coerce_field(name, value)

Arguments

name

Field name.

value

The value to coerce.

Value

The coerced value.

Examples

coerce_field("new_or_reuse", "NEW")
coerce_field("pmid", 12345)

Compose a compound identifier string from a resource's typed fields

Description

Compose a compound identifier string from a resource's typed fields

Usage

compose_identifier(resource, order = NULL)

Arguments

resource

A krt_resource (or a named list of fields).

order

Optional character vector giving the field order; defaults to a sensible canonical order.

Value

A single "; "-joined identifier string.

Examples

r <- new_resource("Antibody", "Anti-TH", catalog_number = "AB152",
                  rrid = "RRID:AB_390204", new_or_reuse = "reuse")
compose_identifier(r)

Detect and parse a Key Resources Table embedded in a document

Description

Detect and parse a Key Resources Table embedded in a document

Usage

detect_existing_krt(doc)

Arguments

doc

The result of read_input_text().

Value

A krt_tbl parsed from the embedded table, or NULL if none is found.

Examples

detect_existing_krt(read_input_text("No table here."))

Export a KRT in the ASAP six-column format

Description

Projects the table to the ASAP columns and writes it as CSV or into an xlsx template. If template is supplied (or the bundled ASAP template is used), the data is written into its KRT sheet, preserving its dropdowns and attribution worksheet.

Usage

export_asap(
  x,
  path = NULL,
  template = NULL,
  format = NULL,
  audience = c("author", "public"),
  redact = NULL,
  attribution = TRUE
)

Arguments

x

A krt_tbl.

path

Output path, or NULL to return CSV text.

template

Optional xlsx template path. When supplied, output is xlsx.

format

"csv" or "xlsx"; inferred from path/template.

audience

"author" (full) or "public" (redacted).

redact

Redaction strength for public output, or FALSE to disable.

attribution

If TRUE (default) and a path is given, write the ASAP CC BY 4.0 attribution block as a sidecar next to it.

Value

The path (invisibly) when written, or CSV text.

Examples

cat(substr(export_asap(krt_example), 1, 60))

Export citable resources as RIS or BibTeX

Description

Only resources that are citable scholarly objects (datasets, software, protocols, or anything with a DOI) are emitted. This format cannot encode the full biological or ethics semantics of a KRT; use it for reference managers, not as an archival copy.

Usage

export_citation(
  x,
  path = NULL,
  format = c("ris", "bibtex"),
  audience = c("author", "public"),
  redact = NULL
)

Arguments

x

A krt_tbl.

path

Output path, or NULL to return the text.

format

"ris" or "bibtex".

audience

"author" (full) or "public" (redacted).

redact

Redaction strength ("basic"/"strict") for public output, or FALSE to disable; NULL uses the profile default.

Value

The path (invisibly) when written, or the citation text.

Examples

cat(export_citation(krt_example, format = "bibtex"))

Export a Key Resources Table

Description

Writes a table in a chosen format. "json" and "yaml" are lossless; "csv", "tsv", "xlsx", "asap", "ris", and "bibtex" are lossy views and raise a lossy-export warning listing fields that are not preserved as columns. When audience = "public", sensitive ethics fields are redacted by default.

Usage

export_krt(
  x,
  path = NULL,
  format = c("json", "yaml", "csv", "tsv", "xlsx", "asap", "ris", "bibtex"),
  profile = NULL,
  audience = c("author", "public"),
  redact = NULL,
  attribution = TRUE,
  template = NULL,
  view = NULL
)

krt_write(
  x,
  path = NULL,
  format = c("json", "yaml", "csv", "tsv", "xlsx", "asap", "ris", "bibtex"),
  profile = NULL,
  audience = c("author", "public"),
  redact = NULL,
  attribution = TRUE,
  template = NULL,
  view = NULL
)

Arguments

x

A krt_tbl.

path

Output file path, or NULL to return the content as a string.

format

Output format; inferred from path when possible.

profile

Profile whose columns tabular exports use (default the table's profile).

audience

"author" (full) or "public" (redacted).

redact

Redaction strength ("basic"/"strict") for public exports, or FALSE to disable (with a warning).

attribution

If TRUE (default) and the profile carries an attribution requirement, write an attribution sidecar next to path.

template

Optional template path for ASAP/xlsx export.

view

Optional explicit view for tabular formats.

Value

The path (invisibly) when written, otherwise the content string.

Examples

export_krt(krt_example, format = "json") |> substr(1, 30)
suppressWarnings(export_krt(krt_example, format = "asap")) |> substr(1, 40)

Export a KRT as a delimited table or spreadsheet

Description

Export a KRT as a delimited table or spreadsheet

Usage

export_tabular(
  x,
  path = NULL,
  format = c("csv", "tsv", "xlsx"),
  profile = NULL,
  view = NULL,
  audience = c("author", "public"),
  redact = NULL,
  attribution = TRUE
)

Arguments

x

A krt_tbl.

path

Output path, or NULL to return the content (csv/tsv only).

format

"csv", "tsv", or "xlsx".

profile

Profile whose columns to use (default the wide core view).

view

Optional explicit view name.

audience

"author" (full) or "public" (redacted).

redact

Redaction strength for public output, or FALSE to disable.

attribution

If TRUE (default), a path is given, and profile carries an attribution requirement, write it as a sidecar next to path.

Value

The path (invisibly) when written, or the delimited text.

Examples

cat(substr(export_tabular(krt_example, format = "csv"), 1, 60))

Extract candidate resources from text (regex engine)

Description

Extract candidate resources from text (regex engine)

Usage

extract_candidates(text)

Arguments

text

A character vector of manuscript text.

Value

A list of krt_resource candidates.

Examples

cand <- extract_candidates("Anti-TH (RRID:AB_390204); FIJI (RRID:SCR_002285)")
length(cand)

Extract a Key Resources Table from a manuscript

Description

Reads a manuscript (PDF, JATS/NISO XML, DOCX, or text), extracts candidate resources with the deterministic regex engine (default) or an LLM, and returns them as a validated krt_tbl. Every result is provenance-stamped with the engine used.

Usage

extract_krt(
  input,
  engine = c("regex", "llm"),
  format = NULL,
  profile = "generic",
  llm = NULL,
  resolve = FALSE,
  existing = c("detect", "ignore"),
  title = NULL
)

Arguments

input

A file path or text string.

engine

"regex" (deterministic, offline) or "llm" (requires a configured provider and API key).

format

Optional explicit input format.

profile

Profile to assign to the extracted table.

llm

A krt_llm() configuration (for the LLM engine).

resolve

Whether to resolve identifiers during validation.

existing

"detect" to also parse an embedded KRT, or "ignore".

title

Optional title for the extracted table.

Value

A list with krt (the extracted krt_tbl), candidates (the raw resource candidates), existing_krt (a parsed embedded table or NULL), and report (a validation report).

Examples

res <- extract_krt("Anti-TH (RRID:AB_390204); FIJI (RRID:SCR_002285).")
nrow(as.data.frame(res$krt))

Extract resources from text with an LLM

Description

Sends the text to the configured provider and parses the returned JSON into candidate resources. Requires an API key (from the environment); returns an empty list if none is configured or the call fails.

Usage

extract_llm(text, llm = krt_llm(), ...)

Arguments

text

Manuscript text.

llm

A krt_llm() configuration.

...

Reserved.

Value

A list of krt_resource candidates.

Examples

# Uses a mock provider so no network or key is needed:
register_llm_provider("mock",
  function(prompt, llm) '[{"resource_type":"Software/code","display_name":"R"}]')
extract_llm("text", krt_llm("openai"))  # empty without a key
extract_llm("text", structure(list(provider = "mock"), class = "krt_llm"))

Fetch manuscript full text

Description

Retrieves full text (JATS XML where available) from Europe PMC or bioRxiv, for feeding into extract_krt(). Returns NULL offline or when the text is not openly available.

Usage

fetch_fulltext(id, source = c("auto", "europepmc", "biorxiv"), timeout = 30)

Arguments

id

A PMCID, DOI, or bioRxiv DOI.

source

"auto", "europepmc", or "biorxiv".

timeout

Request timeout in seconds.

Value

A character string of the full text or abstract, or NULL.

Examples


# Contacts Europe PMC. Offline, or when the text is not openly available,
# this returns NULL rather than failing.
txt <- fetch_fulltext("PMC5334499")
is.null(txt) || nchar(txt) > 0

Key Resources Table field registry

Description

The field registry declares every field a resource record may carry, together with its group, type, cardinality, and the resource types it applies to.

Usage

all_fields()

field_spec(name)

fields_for_type(resource_type)

field_label(name, locale = NULL)

Arguments

name

A field name.

resource_type

A resource type (one of krt_resource_types()).

locale

Optional locale (currently advisory; labels fall back to a title-cased field name).

Value

all_fields() returns a named list of field specifications; field_spec() returns one specification (or NULL); fields_for_type() returns the names of fields that apply to a resource type; field_label() returns a human-readable label.

Examples

str(field_spec("rrid"))
fields_for_type("Antibody")

Find duplicate resources in a KRT

Description

Find duplicate resources in a KRT

Usage

find_duplicates(x, fuzzy = FALSE)

Arguments

x

A krt_tbl.

fuzzy

If TRUE, also report near-duplicates whose display name and vendor are close by edit distance.

Value

A list of duplicate groups; each element is a character vector of resource_ids that share a signature (or are near-duplicates).

Examples

k <- new_krt("Demo")
k <- add_resource(k, "Antibody", "Anti-TH", vendor = "Millipore",
                  catalog_number = "AB152", new_or_reuse = "reuse")
k <- add_resource(k, "Antibody", "Anti-TH (dup)", vendor = "Millipore",
                  catalog_number = "AB152", new_or_reuse = "reuse")
find_duplicates(k)

Retrieve a registered profile

Description

Retrieve a registered profile

Usage

get_profile(name)

Arguments

name

Profile name.

Value

A krt_profile object.

Examples

get_profile("asap")$columns

Parse and classify an identifier string

Description

Detects the scheme of a single identifier token (DOI, ORCID, ROR, RRID, PMID, PMCID, URL, or a database accession) using the bundled syntax patterns.

Usage

id_parse(idstring)

Arguments

idstring

A single identifier token.

Value

A list with scheme (character, or NA if unrecognized), value (the cleaned identifier), and field (the resource field the value belongs in, e.g. "doi", "rrid", "accession").

Examples

id_parse("https://doi.org/10.5281/zenodo.123")
id_parse("RRID:AB_390204")
id_parse("GSE12345")

Import an ASAP or Cell Press Key Resources Table

Description

Import an ASAP or Cell Press Key Resources Table

Usage

import_asap(path, sheet = 1, title = NULL)

Arguments

path

A file path (csv/tsv/xlsx) or a data frame already read in.

sheet

Worksheet (for xlsx).

title

Optional title for the resulting table.

Value

A krt_tbl; profile "asap" for a six-column ASAP sheet, or "star-methods" for a Cell Press three-column table.

Examples

f <- tempfile(fileext = ".csv")
writeLines(export_asap(krt_example), f)
k <- import_asap(f)
length(k$resources)

Import a Key Resources Table from a file or string

Description

Detects the format (JSON, YAML, ASAP/Cell Press table, or generic tabular) and reads it into a krt_tbl.

Usage

import_krt(input, format = NULL, profile = NULL, mapping = NULL, sheet = 1)

Arguments

input

A file path, or a JSON/YAML string.

format

One of "json", "yaml", "asap", "tabular"; auto-detected when NULL.

profile

Optional profile to assign to the imported table.

mapping

Optional column-to-field mapping for tabular import.

sheet

Worksheet (for xlsx).

Value

A krt_tbl.

Examples

k <- import_krt(write_krt_json(krt_example))
length(k$resources)

Import a generic tabular Key Resources Table

Description

Import a generic tabular Key Resources Table

Usage

import_tabular(
  path,
  mapping = NULL,
  sheet = 1,
  profile = "generic",
  title = NULL
)

Arguments

path

A file path (csv/tsv/xlsx) or a data frame.

mapping

Optional named list mapping column names to core field names; guessed from the headers when NULL.

sheet

Worksheet (for xlsx).

profile

Profile to assign to the imported table.

title

Optional title.

Value

A krt_tbl.

Examples

df <- data.frame(type = "Antibody", name = "Anti-TH", rrid = "RRID:AB_1",
                 "new/reuse" = "reuse", check.names = FALSE)
import_tabular(df)$resources[[1]]$resource_type

Is a value the ASAP "identifier pending" placeholder?

Description

Is a value the ASAP "identifier pending" placeholder?

Usage

is_pending_identifier(str)

Arguments

str

A candidate identifier string.

Value

TRUE if the string matches the "Identifier from ... pending" convention.

Examples

is_pending_identifier("Identifier from Cellosaurus pending")

Coerce a KRT's resources to a DataFrame (Bioconductor)

Description

Coerce a KRT's resources to a DataFrame (Bioconductor)

Usage

krt_as_dataframe(x, view = "wide")

Arguments

x

A krt_tbl.

view

The view to project (default "wide").

Value

An S4Vectors::DataFrame of the resources (requires the S4Vectors package).

Examples

if (requireNamespace("S4Vectors", quietly = TRUE)) krt_as_dataframe(krt_example)

Attribution text for a table's profile

Description

Returns the attribution block that should accompany outputs produced with the table's profile. For the ASAP profile (CC BY 4.0) this is the required attribution; for profiles with no redistribution obligation it is a short note.

Usage

krt_attribution(x)

Arguments

x

A krt_tbl or a profile name.

Value

A character string.

Examples

cat(krt_attribution(krt_example))

Audit the licenses of the package and its bundled assets

Description

Audit the licenses of the package and its bundled assets

Usage

krt_audit_licenses()

Value

A data frame with one row per licensable component (the package code, each profile, the bundled reference data and license text), giving its source, license, DOI, whether it is redistributable, and notes.

Examples

krt_audit_licenses()

Run the krt command-line interface

Description

Dispatches a subcommand. Used by the krt shell script (system.file("scripts", "krt", package = "krt")); can also be called directly with a character vector of arguments.

Usage

krt_cli(args = commandArgs(trailingOnly = TRUE))

Arguments

args

Command-line arguments (defaults to those passed to Rscript).

Value

Invisibly an integer status code (0 success, 1 validation failure or runtime error, 2 usage error such as an unknown command or a missing required --out).

Examples

krt_cli("audit-licenses")
f <- tempfile(fileext = ".json")
writeLines(write_krt_json(krt_example), f)
krt_cli(c("validate", f))

Deposit a KRT to Figshare

Description

Creates a Figshare article, reserves a DOI, and uploads the table as JSON. The DOI is recorded in the table's provenance. Deposits default to the redacted public audience.

Usage

krt_deposit_figshare(
  x,
  token = Sys.getenv("FIGSHARE_TOKEN"),
  sandbox = TRUE,
  metadata = NULL,
  audience = c("public", "author"),
  timeout = 60
)

Arguments

x

A krt_tbl.

token

Figshare token (defaults to FIGSHARE_TOKEN).

sandbox

Use the Figshare sandbox (default TRUE).

metadata

Optional article metadata.

audience

"public" (redacted, the default) or "author" (full).

timeout

Request timeout in seconds.

Value

A list with article (the API response), doi, uploaded (logical), and x.

Examples

## Not run: 
# Not executable without a Figshare (sandbox) account and access token, so
# this example cannot be run during a check.
krt_deposit_figshare(krt_example)

## End(Not run)

Deposit a KRT to Zenodo

Description

Creates a Zenodo deposition, uploads the table as JSON, sets metadata, and optionally publishes. The minted (pre-reserved) DOI is written back into the table's provenance and table_id. Deposits default to the redacted public audience, since a deposit is a shared artifact.

Usage

krt_deposit_zenodo(
  x,
  token = Sys.getenv("ZENODO_TOKEN"),
  sandbox = TRUE,
  metadata = NULL,
  publish = FALSE,
  audience = c("public", "author"),
  timeout = 60
)

Arguments

x

A krt_tbl.

token

Zenodo access token (defaults to ZENODO_TOKEN).

sandbox

Use the Zenodo sandbox (default TRUE).

metadata

Optional Zenodo metadata list (a sensible default otherwise).

publish

Whether to publish immediately (default FALSE).

audience

"public" (redacted, the default) or "author" (full).

timeout

Request timeout in seconds.

Value

A list with deposit (the API response), doi, uploaded and published (per-stage status), and x (the table, with the DOI recorded only when every stage succeeded). On failure, deposit is NULL.

Examples

## Not run: 
# Not executable without a Zenodo (sandbox) account and access token, so
# this example cannot be run during a check.
res <- krt_deposit_zenodo(krt_example, sandbox = TRUE)
res$doi

## End(Not run)

Diff two Key Resources Tables

Description

Compares resources by identity signature and reports which were added, removed, or changed (with field-level deltas).

Usage

krt_diff(x, y)

Arguments

x, y

Two krt_tbl objects (x is the baseline).

Value

A krt_diff object with print() and as.data.frame() methods.

Examples

a <- add_resource(new_krt("A"), "Dataset", "D", doi = "10.5281/zenodo.1",
                  new_or_reuse = "new")
b <- update_resource(a, a$resources[[1]]$resource_id, notes = "added")
as.data.frame(krt_diff(a, b))

An example Key Resources Table

Description

A small krt_tbl spanning several common resource types (antibody, cell line, software, dataset, organism/strain, and protocol). Used throughout the documentation and tests so that examples run offline.

Usage

krt_example

Format

An object of class krt_tbl with six resources.

Source

Constructed by data-raw/06-build-example-krt.R. Identifiers are real, publicly resolvable examples; the study is fictional.

Examples

krt_example
summary(krt_example)

Import resources from an eLabFTW experiment

Description

Fetches an experiment and extracts candidate resources from its body text.

Usage

krt_import_elabftw(
  base_url,
  experiment_id,
  token = Sys.getenv("ELABFTW_TOKEN"),
  timeout = 30
)

Arguments

base_url

The eLabFTW instance base URL (e.g. ⁠https://elab.example.org⁠).

experiment_id

The experiment id.

token

API token (defaults to ELABFTW_TOKEN).

timeout

Request timeout in seconds.

Value

A krt_tbl of extracted resources.

Examples

## Not run: 
# Not executable without an eLabFTW instance and an API token, so this
# example cannot be run during a check.
krt_import_elabftw("https://elab.example.org", experiment_id = 42)

## End(Not run)

Import a protocol from protocols.io

Description

Fetches a protocol and records it as a Protocol resource.

Usage

krt_import_protocolsio(
  id,
  token = Sys.getenv("PROTOCOLSIO_TOKEN"),
  timeout = 30
)

Arguments

id

A protocols.io protocol id or DOI.

token

API token (defaults to PROTOCOLSIO_TOKEN).

timeout

Request timeout in seconds.

Value

A krt_tbl with a single Protocol resource.

Examples


# Contacts the public protocols.io API. Offline, this warns and returns an
# empty table rather than failing.
k <- suppressWarnings(krt_import_protocolsio("kxygx3w"))
length(k$resources)

Configure an LLM for extraction

Description

Configure an LLM for extraction

Usage

krt_llm(
  provider = c("openai", "anthropic", "gemini", "openai_compat"),
  model = NULL,
  base_url = NULL,
  api_key = NULL,
  temperature = 0,
  max_tokens = 4096
)

## S3 method for class 'krt_llm'
format(x, ...)

## S3 method for class 'krt_llm'
print(x, ...)

Arguments

provider

One of the registered providers ("openai", "anthropic", "gemini", "openai_compat" for local/OpenAI-compatible servers).

model

Model id (a sensible default per provider when NULL).

base_url

Base URL for "openai_compat" (e.g. a local server).

api_key

API key; defaults to the provider's environment variable.

temperature, max_tokens

Generation parameters.

x

A krt_llm object.

...

Ignored.

Value

A krt_llm configuration object.

Examples

krt_llm("openai", model = "gpt-4o-mini")$provider

Merge Key Resources Tables

Description

Combines resources from two or more tables, matching by normalized identity signature. Matching resources are merged field by field; the strategy resolves field-level conflicts. Conflicts are attached to the result as the "merge_conflicts" attribute.

Usage

krt_merge(
  x,
  y,
  ...,
  strategy = c("union", "prefer_x", "prefer_y", "manual"),
  by = resource_signature
)

Arguments

x, y, ...

Tables to merge (x supplies the result's metadata).

strategy

"union"/"prefer_x" (x wins conflicts), "prefer_y" (later table wins), or "manual" (x wins but every conflict is recorded).

by

A function computing a resource's match key (default resource_signature()).

Value

The merged krt_tbl.

Examples

a <- add_resource(new_krt("A"), "Antibody", "Anti-TH", vendor = "Millipore",
                  catalog_number = "AB152", new_or_reuse = "reuse")
b <- add_resource(new_krt("B"), "Dataset", "D", doi = "10.5281/zenodo.1",
                  new_or_reuse = "new")
krt_merge(a, b)

Get or set KRT metadata

Description

Get or set KRT metadata

Usage

krt_meta(x)

krt_meta(x) <- value

Arguments

x

A krt_tbl.

value

A named list of metadata fields to set (any of title, profile, study_type, locale).

Value

krt_meta() returns a named list of the table's metadata.

Examples

k <- new_krt("Demo")
krt_meta(k)$title
krt_meta(k) <- list(title = "Renamed")

The krt plugin API

Description

The krt plugin API

Usage

krt_plugin_api()

Value

A data frame describing each plugin kind, its registration function, and its contract.

Examples

krt_plugin_api()

Extend krt with plugins

Description

krt is extensible through five registries. Register custom components with the corresponding function; krt_plugin_api() lists the entry points and their contracts, and validate_plugin_contract() checks an object before you register it.

See Also

register_profile(), register_validator(), register_resolver(), register_llm_provider(), register_suggest_source().

Pre-flight release readiness check for a Key Resources Table

Description

Runs the checks that together decide whether a table is ready to share or deposit, and returns a single machine-readable verdict plus a human-readable checklist: profile validation (no errors), lossless JSON round-trip, a working public (redacted) export that leaks none of the policy's dropped fields, attribution availability, and whether the profile projection would drop fields.

Usage

krt_preflight(x, profile = NULL)

Arguments

x

A krt_tbl.

profile

Profile to check against (defaults to the table's profile).

Value

A krt_preflight object: a list with profile, ok (TRUE when no check fails), and checks (a data frame of check, status, detail), with print(), format(), and as.data.frame() methods.

Examples

pf <- krt_preflight(krt_example)
pf$ok

Describe a profile

Description

Describe a profile

Usage

krt_profile_info(name)

Arguments

name

A profile name or a krt_profile.

Value

The krt_profile (invisibly); prints a human-readable summary including its license and redistribution status.

Examples

krt_profile_info("asap")

Licensing of a profile

Description

Licensing of a profile

Usage

krt_profile_license(name)

krt_profile_sources(name)

krt_profile_attribution(name)

Arguments

name

A profile name or krt_profile.

Value

krt_profile_license() returns the SPDX license id; krt_profile_sources() returns the source metadata list; krt_profile_attribution() returns the attribution text (or NULL).

Examples

krt_profile_license("asap")
krt_profile_sources("asap")$doi

List available profiles

Description

List available profiles

Usage

krt_profiles()

Value

A data frame with one row per registered profile (name, title, license, whether it is officially endorsed).

Examples

krt_profiles()

Provenance of a KRT

Description

Returns the ordered provenance entries as a krt_provenance object (a list of krt_prov_entry, so length() gives the number of steps). It has print() and as.data.frame() methods; serialize the provenance graph with as_prov_json() and as_rocrate().

Usage

krt_provenance(x)

Arguments

x

A krt_tbl.

Value

A krt_provenance object.

Examples

krt_provenance(normalize_ids(krt_example))

Suggest canonical names and identifiers from public authorities

Description

Queries ontology and registry search endpoints to autocomplete a resource's canonical name or identifier. Requires network access; returns an empty result offline.

Usage

krt_suggest(query, type = NULL, authority = "auto", n = 10, resolve = TRUE)

Arguments

query

The text to search for.

type

Optional resource type hint.

authority

Which authority to query: "auto", or the name of any registered source (built-ins: "taxonomy", "cellosaurus", "chebi", "ror"; plus any added with register_suggest_source()).

n

Maximum number of suggestions.

resolve

Whether to contact the network (default TRUE).

Value

A data frame with columns label, id, authority, score, uri.

Examples

krt_suggest("dopamine", authority = "chebi", resolve = FALSE)

The validation report object

Description

validate_krt() returns a krt_validation_report. It has print(), summary(), and as.data.frame() methods. valid is TRUE when there are no error findings.

See Also

validate_krt()

Access a KRT controlled vocabulary

Description

Access a KRT controlled vocabulary

Usage

krt_vocab(key)

Arguments

key

Vocabulary name (e.g. "resource_type", "new_or_reuse", "status").

Value

A character vector of allowed values.

Examples

krt_vocab("new_or_reuse")

KRT controlled vocabularies

Description

Convenience accessors for the controlled vocabularies used across the package.

Usage

krt_resource_types()

krt_new_or_reuse()

krt_statuses()

krt_approval_types()

krt_roles()

krt_redaction_levels()

Value

A character vector of allowed values.

Examples

krt_resource_types()
krt_new_or_reuse()

Write attribution to a file

Description

Write attribution to a file

Usage

krt_write_attribution(x, path)

Arguments

x

A krt_tbl or profile name.

path

Output file path.

Value

The path, invisibly.

Examples

f <- tempfile(fileext = ".md")
krt_write_attribution(krt_example, f)

Launch the interactive KRT editor

Description

Starts a Shiny application for importing, editing (in the generic view), validating, normalizing identifiers, and exporting a Key Resources Table. Requires the shiny, bslib, and DT packages.

Usage

launch_krt(...)

Arguments

...

Passed to shiny::runApp().

Value

Called for its side effect (runs the app).

Examples

if (interactive()) launch_krt()

List registered LLM providers

Description

List registered LLM providers

Usage

list_llm_providers()

Value

A character vector of provider names.

Examples

list_llm_providers()

List registered resolver schemes

Description

List registered resolver schemes

Usage

list_resolvers()

Value

A character vector of scheme names.

Examples

list_resolvers()

List autocomplete sources

Description

List autocomplete sources

Usage

list_suggest_sources()

Value

A character vector of source names.

Examples

list_suggest_sources()

List registered validation rules

Description

List registered validation rules

Usage

list_validators()

Value

A data frame of registered rules (id, layer, default severity, standard).

Examples

head(list_validators())

Load a profile from a directory

Description

Reads schema.yml, mappings.yml, and (optionally) validation.yml, provenance.json, and ATTRIBUTION.md from a profile directory.

Usage

load_profile(path)

is_profile(x)

Arguments

path

Path to the profile directory.

x

An object to test.

Value

A krt_profile object.

Examples

p <- load_profile(system.file("extdata", "profiles", "asap", package = "krt"))
p$columns

Fields lost or folded when projecting to a profile

Description

Returns the core fields that are present in the table but are not preserved as their own column by the profile (they are either folded into a free-text catch-all column or dropped). These drive the lossy-export warning.

Usage

mapping_lossy_fields(x, profile)

Arguments

x

A krt_tbl.

profile

A krt_profile or profile name.

Value

A character vector of field names (empty for a lossless profile).

Examples

mapping_lossy_fields(krt_example, "asap")

Create and add an ethics or governance approval

Description

Create and add an ethics or governance approval

Usage

new_approval(approval_type, ..., .id = NULL)

add_approval(x, ...)

Arguments

approval_type

One of krt_approval_types() (e.g. "IACUC", "IRB", "REB", "ethics").

...

Additional named fields: board_name, protocol_number, institution_name, institution_ror, jurisdiction, approved_on, consent_obtained, consent_scope, data_use_restrictions, redaction_level.

.id

Optional explicit approval id.

x

A krt_tbl.

Value

new_approval() returns a krt_approval; add_approval() returns the updated krt_tbl.

Examples

a <- new_approval("IACUC", protocol_number = "2026-017",
                  board_name = "Example IACUC")
k <- add_approval(new_krt("Demo"), a)

Create and add a contributor

Description

Create and add a contributor

Usage

new_contributor(name, ..., .id = NULL)

add_contributor(x, ...)

Arguments

name

Contributor name.

...

Additional named fields: orcid, role (see krt_roles()), affiliation, affiliation_ror.

.id

Optional explicit contributor id.

x

A krt_tbl.

Value

new_contributor() returns a krt_contributor; add_contributor() returns the updated krt_tbl.

Examples

k <- add_contributor(new_krt("Demo"), "Ada Researcher",
                     orcid = "0000-0002-1825-0097", role = "author")

Create a Key Resources Table

Description

Constructs an empty (or pre-populated) krt_tbl: the neutral, typed core object of the package. Resources, approvals, and contributors are stored as present-only records; the rectangular table an author sees is a view produced on demand by as.data.frame().

Usage

new_krt(
  title = NULL,
  profile = "generic",
  study_type = NULL,
  locale = NULL,
  resources = list(),
  approvals = list(),
  contributors = list()
)

krt_new(
  title = NULL,
  profile = "generic",
  study_type = NULL,
  locale = NULL,
  resources = list(),
  approvals = list(),
  contributors = list()
)

is_krt(x)

Arguments

title

A short table or study title.

profile

Output profile name (default "generic"); see krt_profiles().

study_type

Optional character vector describing the study (e.g. c("wet-lab", "computational")).

locale

Optional locale string (e.g. "en-US").

resources, approvals, contributors

Optional lists of records to seed the table with.

x

An object to test.

Value

An object of class krt_tbl.

Examples

k <- new_krt("Example study", study_type = "wet-lab")
k <- add_resource(k, "Antibody", "Rabbit Anti-TH", vendor = "Millipore",
                  catalog_number = "AB152", rrid = "RRID:AB_390204",
                  new_or_reuse = "reuse")
k

Create a resource record

Description

Builds a single krt_resource: a flat, present-only record describing one research resource. Identifier fields are stored separately by type (catalog_number, rrid, doi, ...); they are only combined into a compound identifier string at export time.

Usage

new_resource(
  resource_type,
  display_name = NULL,
  ...,
  .id = NULL,
  .validate = TRUE
)

is_resource(x)

Arguments

resource_type

One of krt_resource_types().

display_name

The resource name as it appears in the manuscript.

...

Additional named fields (see all_fields() for the vocabulary), for example vendor, catalog_number, rrid, doi, new_or_reuse, notes.

.id

Optional explicit resource id; generated from content if omitted.

.validate

If TRUE (default), check resource_type against the controlled vocabulary and warn about unknown fields.

x

An object to test.

Value

An object of class krt_resource.

Examples

new_resource("Antibody", "Rabbit Anti-TH", vendor = "Millipore",
             catalog_number = "AB152", rrid = "RRID:AB_390204",
             new_or_reuse = "reuse")

Normalize identifiers to canonical forms

Description

Canonicalizes the identifier fields of a table, a resource, or a bare character vector: strips resolver prefixes from DOIs, hyphenates ORCIDs, ensures the ⁠RRID:⁠ prefix, and so on. Applied consistently, this prevents the same identifier from appearing in several syntactic forms.

Usage

normalize_ids(x, ...)

Arguments

x

A krt_tbl, a krt_resource, or a character vector of identifiers.

...

Ignored.

Value

An object of the same type as x, with identifiers normalized.

Examples

normalize_ids("https://doi.org/10.1038/SDATA.2016.18")
r <- new_resource("Software/code", "Fiji", rrid = "SCR_002285",
                  new_or_reuse = "reuse")
normalize_ids(r)$rrid

Parse a compound identifier string into typed fields

Description

Splits an ASAP-style IDENTIFIER value (parts joined by ⁠;⁠ or newlines) and classifies each part into a resource field.

Usage

parse_compound_identifier(str)

Arguments

str

A compound identifier string.

Value

A named list of fields (e.g. catalog_number, rrid, doi, accession, url), plus other for unclassified parts.

Examples

parse_compound_identifier("Cat# AB152; RRID:AB_390204")

Project a KRT onto a profile as a data frame

Description

Project a KRT onto a profile as a data frame

Usage

project_profile(x, profile)

Arguments

x

A krt_tbl.

profile

A krt_profile or profile name.

Value

A data frame with the profile's columns, one row per resource.

Examples

project_profile(krt_example, "asap")[, c("RESOURCE TYPE", "IDENTIFIER")]

Read manuscript text and tables from an input

Description

Read manuscript text and tables from an input

Usage

read_input_text(input, format = NULL)

Arguments

input

A file path (pdf/xml/jats/docx/txt) or a plain-text string.

format

Optional explicit format; auto-detected when NULL.

Value

A list with text (character), tables (list of data frames), and format.

Examples

read_input_text("We used FIJI (RRID:SCR_002285).")$text

Read a KRT from canonical JSON

Description

Read a KRT from canonical JSON

Usage

read_krt_json(input)

Arguments

input

A file path or a JSON string.

Value

A krt_tbl.

Examples

k <- read_krt_json(write_krt_json(krt_example))
identical(length(k$resources), length(krt_example$resources))

Read a KRT from canonical YAML

Description

Read a KRT from canonical YAML

Usage

read_krt_yaml(input)

Arguments

input

A file path or a YAML string.

Value

A krt_tbl.

Examples

k <- read_krt_yaml(write_krt_yaml(krt_example))
identical(length(k$resources), length(krt_example$resources))

Redact sensitive fields for public sharing

Description

Removes or generalizes fields flagged by the redaction policy, so a table can be shared publicly without exposing internal ethics or consent details.

Usage

redact_krt(x, level = c("basic", "strict"), policy = NULL)

Arguments

x

A krt_tbl.

level

"basic" (default) removes basic-tagged fields; "strict" removes basic- and strict-tagged fields.

policy

An optional policy data frame overriding redaction_policy().

Value

The redacted krt_tbl.

Examples

k <- add_approval(new_krt("Demo"), "IRB", protocol_number = "IRB-1",
                  consent_scope = "study-specific")
redact_krt(k)$approvals[[1]]$protocol_number

Default redaction strength for a profile's public exports

Description

Default redaction strength for a profile's public exports

Usage

redaction_default(profile = NULL)

Arguments

profile

A profile name or krt_profile. A profile may declare a redaction_default in its schema.yml; otherwise the strength is "basic".

Value

"basic" or "strict".

Examples

redaction_default("asap")

The default redaction policy

Description

The default redaction policy

Usage

redaction_policy()

Value

A data frame with columns scope ("approval" or "resource"), field, level (the strip strength at which the field is removed: "basic" fields are removed at both "basic" and "strict"; "strict" fields only at "strict"), and action ("drop" or "generalize").

Examples

redaction_policy()

Register an LLM provider

Description

Register an LLM provider

Usage

register_llm_provider(name, request_fn, parse_fn = NULL, replace = FALSE)

Arguments

name

Provider name (e.g. "openai", "anthropic", a local endpoint).

request_fn

A function ⁠function(prompt, llm)⁠ returning the model's text output, or NULL on failure.

parse_fn

Optional custom parser ⁠function(text)⁠; defaults to JSON array extraction.

replace

Overwrite an existing provider named name? Defaults to FALSE so a plugin cannot silently replace a built-in provider.

Value

Invisibly NULL.

Examples

register_llm_provider("echo", function(prompt, llm) "[]", replace = TRUE)
"echo" %in% list_llm_providers()

Register an output profile

Description

Register an output profile

Usage

register_profile(name = NULL, path = NULL, profile = NULL, replace = FALSE)

Arguments

name

Profile name. If omitted, taken from the profile's schema.yml or the supplied object.

path

Path to a profile directory containing schema.yml and mappings.yml (loaded lazily), or NULL.

profile

A pre-built krt_profile object, or NULL.

replace

Overwrite a profile already registered under name? Defaults to FALSE, so a plugin cannot silently replace a built-in profile (such as asap); pass TRUE to deliberately override one.

Value

Invisibly the profile name.

Examples

krt_profiles()

Register an identifier resolver

Description

Register an identifier resolver

Usage

register_resolver(scheme, fn, replace = FALSE)

Arguments

scheme

The identifier scheme (e.g. "rrid", "doi", "orcid").

fn

A function ⁠function(id, resolve = TRUE, ...)⁠ returning a normalized result list with at least input, normalized, and resolved.

replace

Overwrite an existing resolver for scheme? Defaults to FALSE so a plugin cannot silently replace a built-in resolver.

Value

Invisibly NULL.

Examples

"rrid" %in% list_resolvers()

Register an autocomplete source

Description

Register an autocomplete source

Usage

register_suggest_source(name, fn, replace = FALSE)

Arguments

name

Source name (e.g. "taxonomy", "ror").

fn

A function ⁠function(query, n)⁠ returning a data frame with columns label, id, authority, score, uri.

replace

Overwrite an existing source named name? Defaults to FALSE so a plugin cannot silently replace a built-in source.

Value

Invisibly NULL.

Examples

"ror" %in% list_suggest_sources()

Register a validation rule

Description

Adds a rule to the validation engine. A rule is a function fn(x, ctx) that inspects a krt_tbl and returns a list of issues (each created with the internal issue helper); the engine attaches the rule id, layer, standard, and resolved severity.

Usage

register_validator(
  rule_id,
  fn,
  layer = c("structural", "semantic"),
  severity = c("error", "warning", "note", "info"),
  applies = function(x) TRUE,
  standard = NA_character_,
  replace = FALSE
)

Arguments

rule_id

A unique rule identifier, e.g. "struct-missing-name".

fn

The rule function ⁠function(x, ctx)⁠ returning a list of issues.

layer

Either "structural" or "semantic".

severity

Default severity: one of "error", "warning", "note", "info".

applies

A predicate ⁠function(x)⁠; the rule runs only when it returns TRUE (used by conditional rule packs).

standard

Optional reporting standard the rule enforces (e.g. "cell-line-auth-minimum").

replace

Overwrite an existing rule with the same rule_id? Defaults to FALSE, so a plugin cannot silently replace a built-in rule; pass TRUE to deliberately override one.

Value

Invisibly NULL; called for its side effect.

Examples

# A rule inspects the table and returns a list of issues. This demonstration
# rule reports nothing, so registering it leaves validation results unchanged.
demo_rule <- function(x, ctx) list()
register_validator("demo-no-op", demo_rule, layer = "semantic",
                   severity = "note", replace = TRUE)
"demo-no-op" %in% list_validators()$rule_id
validate_krt(krt_example)$valid

Render a KRT as a formatted table

Description

Produces a human-readable Key Resources Table. The "star-methods" profile projects the table to the three Cell Press columns (REAGENT or RESOURCE, SOURCE, IDENTIFIER) and groups resources under the twelve standard STAR Methods category headers, in the order the template uses. The "generic" profile (the default) renders the ASAP six-column layout; any other named profile renders through its own declared columns.

Usage

render_krt(
  x,
  path = NULL,
  format = c("md", "html", "docx"),
  profile = NULL,
  template = NULL,
  audience = c("author", "public"),
  redact = NULL
)

Arguments

x

A krt_tbl.

path

Output path, or NULL to return the text (md/html).

format

"md", "html", or "docx".

profile

Profile controlling the layout (default the table's profile).

template

Unused placeholder for a future Word template.

audience

"author" (full) or "public" (redacted) for shared tables.

redact

Redaction strength for public output, or FALSE to disable.

Value

The rendered text (md/html), or the path (invisibly) for docx.

Examples

cat(substr(render_krt(krt_example, profile = "star-methods"), 1, 80))

Resolve any identifier by detecting its scheme

Description

Resolve any identifier by detecting its scheme

Usage

resolve_id(id, resolve = TRUE, ...)

Arguments

id

An identifier string.

resolve

Whether to contact the registry.

...

Passed to the scheme-specific resolver.

Value

A resolver result list, or NULL if the scheme is unsupported.

Examples

resolve_id("RRID:AB_390204", resolve = FALSE)$normalized

Resolve identifiers against public registries

Description

Each resolver normalizes an identifier and, when resolve = TRUE and the registry is reachable, retrieves a display name and type. All calls degrade gracefully: offline or on error, resolved is FALSE.

Usage

resolve_rrid(rrid, resolve = TRUE, timeout = 15)

resolve_doi(doi, resolve = TRUE, timeout = 15)

resolve_orcid(orcid, resolve = TRUE, timeout = 15)

resolve_pubmed(pmid, resolve = TRUE, timeout = 15)

resolve_ror(ror, resolve = TRUE, timeout = 15)

resolve_cellosaurus(cvcl, resolve = TRUE, timeout = 15)

Arguments

rrid, doi, orcid, pmid, ror, cvcl

The identifier to resolve.

resolve

Whether to contact the registry (default TRUE). Set FALSE for a purely offline, normalize-only result.

timeout

Request timeout in seconds.

Value

A list with input, normalized, resolved, source, name, type, and url.

Examples

# Offline: normalize only (no network).
resolve_rrid("RRID:AB_390204", resolve = FALSE)
resolve_doi("10.1038/sdata.2016.18", resolve = FALSE)$normalized

# Live lookup. This contacts Crossref; offline it simply returns the
# normalized identifier with `resolved = FALSE` instead of failing.
resolve_doi("10.1038/sdata.2016.18", resolve = TRUE)$resolved

Normalized signature of a resource

Description

Builds a lowercased, trimmed tuple from the identifying fields ⁠(resource_type, vendor, catalog_number, lot_number, rrid, doi, accession)⁠. Two resources with the same signature are considered duplicates.

Usage

resource_signature(resource)

Arguments

resource

A krt_resource or named list.

Value

A single signature string.

Examples

r <- new_resource("Antibody", "Anti-TH", vendor = "Millipore",
                  catalog_number = "AB152", new_or_reuse = "reuse")
resource_signature(r)

Resource type implied by an RRID

Description

Resource type implied by an RRID

Usage

rrid_type(rrid)

Arguments

rrid

An RRID string (with or without the ⁠RRID:⁠ prefix).

Value

The implied resource type (character) or NA if the authority is unknown.

Examples

rrid_type("RRID:AB_390204")
rrid_type("CVCL_0063")

Scan free text for research identifiers

Description

Finds RRIDs, DOIs, catalog numbers, database accessions, and PMIDs in a block of text and classifies each. Used by the regex extraction engine and available on its own.

Usage

scan_identifiers(text)

Arguments

text

A character vector of text.

Value

A data frame with columns value, field, type (an inferred resource type, or NA), and confidence ("high" for precisely anchored schemes such as RRID/DOI/accession/PMID, "medium" for looser catalog-number matches) so the results can be triaged before acceptance.

Examples

scan_identifiers("We used anti-TH (RRID:AB_390204) and FIJI (RRID:SCR_002285).")

Summarize a KRT

Description

Summarize a KRT

Usage

## S3 method for class 'krt_tbl'
summary(object, ...)

Arguments

object

A krt_tbl.

...

Ignored.

Value

A data frame with resource counts per resource type, including the number of newly generated versus reused resources.

Examples

summary(krt_example)

Summarize a validation report

Description

Summarize a validation report

Usage

## S3 method for class 'krt_validation_report'
summary(object, ...)

Arguments

object

A krt_validation_report.

...

Ignored.

Value

A data frame of finding counts by severity, layer, and standard.

Validate a Key Resources Table

Description

Runs the registered validation rules over a table and returns a krt_validation_report. Structural rules check schema conformance offline; semantic rules check cross-field consistency and, when resolve = TRUE, identifier existence. Conditional packs (cell-line authentication, organism metadata, ethics/consent) apply only when the relevant resource types are present; they are minimal checks, not full ICLAC or ARRIVE assessments.

Usage

validate_krt(
  x,
  profile = NULL,
  layers = c("structural", "semantic"),
  resolve = FALSE,
  severity = NULL,
  attach = FALSE
)

Arguments

x

A krt_tbl.

profile

Profile whose severity overrides apply (default the table's profile).

layers

Which layers to run: "structural", "semantic", or both.

resolve

If TRUE, semantic rules may perform online existence checks (off by default; never on CRAN).

severity

Optional named list mapping rule_id to a severity (or "off"), overriding rule and profile defaults.

attach

If TRUE, return the table with the findings stored in its validation slot instead of returning the report.

Details

Each finding's severity is resolved from the rule default, then any profile override, then any per-rule value in severity. A severity of "off" disables the rule.

Value

A krt_validation_report, or the krt_tbl when attach = TRUE.

Examples

validate_krt(krt_example)
validate_krt(krt_example, layers = "structural")

Check that an object satisfies a plugin contract

Description

Check that an object satisfies a plugin contract

Usage

validate_plugin_contract(kind, obj)

Arguments

kind

One of "profile", "validator", "resolver", "llm_provider", "suggest_source".

obj

The plugin object or function to check.

Value

Invisibly TRUE; errors early if the contract is not met.

Examples

validate_plugin_contract("validator", function(x, ctx) list())
validate_plugin_contract("suggest_source", function(query, n) NULL)

Match a value against a controlled vocabulary

Description

Match a value against a controlled vocabulary

Usage

vocab_match(value, vocab, fuzzy = FALSE)

Arguments

value

A character value to check.

vocab

A character vector of allowed values, or the name of a vocabulary (resolved with krt_vocab()).

fuzzy

If TRUE, when there is no exact match return the nearest vocabulary term (by edit distance) as a suggestion instead of NA.

Value

A list with ok (logical), value (the matched canonical term or NA), and suggestion (nearest term when fuzzy and not matched).

Examples

vocab_match("Antibody", "resource_type")
vocab_match("antibodies", "resource_type", fuzzy = TRUE)

Write a KRT to canonical JSON

Description

Write a KRT to canonical JSON

Usage

write_krt_json(x, path = NULL, pretty = TRUE)

Arguments

x

A krt_tbl.

path

Output file path, or NULL to return the JSON as a string.

pretty

Whether to pretty-print (default TRUE).

Value

The JSON string (invisibly, the path when written to a file).

Examples

json <- write_krt_json(krt_example)
substr(json, 1, 40)

Write a KRT to canonical YAML

Description

Write a KRT to canonical YAML

Usage

write_krt_yaml(x, path = NULL)

Arguments

x

A krt_tbl.

path

Output file path, or NULL to return the YAML as a string.

Value

The YAML string (invisibly, the path when written to a file).

Examples

cat(substr(write_krt_yaml(krt_example), 1, 40))