Expanded trait identifier validation: Expanded
accepted GWAS Catalog trait identifier formats beyond the original
EFO-only validation.
Backward compatibility preserved: Preserved the
existing efo_id argument and --efo CLI option
for backward compatibility.
Identifier normalization added: Added identifier
normalization for underscore and colon formats.
Selected supported prefixes: Added support for
selected disease and cancer trait identifier prefixes including EFO,
MONDO, and NCIT when supported by the GWAS Catalog API.
Compatibility prefixes accepted: Added
compatibility validation for selected phenotype and rare-disease
prefixes, including HP, Orphanet, and ORPHA.
GO explicitly rejected: Explicitly rejected GO
identifiers as primary GWAS Catalog trait identifiers in gwas2crispr
0.1.5.
Retrieval cascade improved: Added direct
identifier-based association retrieval attempts before label-based
retrieval.
Coordinate recovery improved: Improved coordinate
recovery for rsID records when GWAS Catalog responses lack complete
GRCh38 coordinates.
CLI, documentation, and tests updated: Updated CLI
validation, documentation, and tests.
gwas2crispr 0.1.4
GWAS Catalog REST API v2 backend: Replaced the
former third-party GWAS retrieval workflow with direct EMBL-EBI GWAS
Catalog REST API v2 retrieval.
Removed old GWAS retrieval dependency: Removed
gwasrapidd from package dependency metadata, code, tests,
and user-facing documentation.
Updated fetch_gwas() output:fetch_gwas() now returns a package-native list with:
associations
risk_alleles
cache
Updated run_gwas2crispr() integration:run_gwas2crispr() now consumes the new package-native
fetch_gwas() result and no longer expects an S4
associations object.
Direct endpoint workflow: Added direct handling for
GWAS Catalog REST API v2 trait resolution, association retrieval, and
SNP metadata retrieval.
Optional FASTA behavior improved: CSV and BED
outputs remain available even when FASTA dependencies are unavailable.
FASTA is generated only when BSgenome.Hsapiens.UCSC.hg38
and Biostrings are installed.
Output path handling improved:run_gwas2crispr() no longer returns NA entries
inside the written vector when FASTA output is
skipped.
Summary logic improved:SNPs_w_gene is
now calculated as the number of distinct SNPs with gene annotation,
avoiding inflated counts when variants appear in multiple association
rows.
Tests updated: Updated tests for input validation,
direct REST API output structure, optional FASTA behavior, safe writing
to tempdir(), and no-default-write behavior.
Documentation updated: Updated README,
cran-comments.md, roxygen documentation, and package
metadata for the direct GWAS Catalog REST API v2 workflow.
Version bump: Bumped package version to
0.1.4.
gwas2crispr 0.1.2
CLI improvements: The command-line script
(inst/scripts/gwas2crispr.R) no longer forces file output.
The --out/-o option is now truly optional:
omit it to run the pipeline in memory without writing any files. A new
--verbose/-v flag prints a concise summary
when no outputs are written. The double comma in the option definition
has been fixed.
Console output: All side-effects
(print()/cat()) have been replaced with
message() behind a verbose flag. By default,
functions run quietly unless verbose = TRUE or the new CLI
flag is used.
Testing: Added
tests/testthat/test-run_gwas2crispr.R to ensure that
run_gwas2crispr() writes only to a provided
out_prefix and returns objects without writing by
default.
Documentation: Updated DESCRIPTION with expanded
acronyms and method references; ensured Language: en-US and
valid Bioconductor biocViews. Updated the README and
vignette to reflect the new no-default-write behaviour and optional CLI
output. Added a simple inst/CITATION entry. Bumped version
to 0.1.2.
gwas2crispr 0.1.1
Addressed CRAN review comments:
Expanded acronyms in DESCRIPTION: GWAS, EFO, SNP, BED, FASTA, and
CRISPR.