Changes in version 0.1.5 (2026-06-02)                  

  - 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.
  - Optional Ensembl fallback added: Added optional non-fatal Ensembl
    REST fallback for missing rsID coordinates.
  - CLI, documentation, and tests updated: Updated CLI validation,
    documentation, and tests.

                 Changes in version 0.1.4 (2026-05-09)                  

  - 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.

                 Changes in version 0.1.2 (2025-08-22)                  

  - 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.

                        Changes in version 0.1.1                        

  - Addressed CRAN review comments:
      - Expanded acronyms in DESCRIPTION: GWAS, EFO, SNP, BED, FASTA,
        and CRISPR.
      - Added method reference: Sudlow et al. (2015)
        doi:10.1093/nar/gkv1256.
      - Replaced \dontrun{} with \donttest{} where network I/O may
        occur.
      - Kept examples short.
      - Ensured run_gwas2crispr(out_prefix = NULL) returns objects only
        and does not write files by default.
      - Ensured examples, vignette, and tests write only to tempdir()
        when file output is needed.
      - Made console output suppressible via verbose and message().
  - Qualified utils::capture.output to avoid NOTES.
  - Updated vignette with object-only example by default and optional
    tempdir() example for file outputs.
  - Fixed CLI script by correcting the --out option to use type =
    "character".
  - Fixed CITATION by defining the meta object before using
    meta$Version.
  - Corrected the Language field.
  - Expanded DESCRIPTION with acronym explanations and references.

                        Changes in version 0.1.0                        

  - Initial public release.
  - Fetch GWAS associations and export CSV, BED, and FASTA outputs for
    GRCh38/hg38.
  - Added CLI script at inst/scripts/gwas2crispr.R.