Skip to contents

Changelog

tidyGWAS 1.0.0

This version comes with major changes to tidyGWAS.

  1. File-parsing is now done with data.table. No need to use the delim argument
  2. tidyGWAS now accepts four possible inputs to tbl
  1. local filepath,
  2. in-memory data.frame
  3. a GWAS catalogs study id (will download and process)
  4. a URL to a downloadable file (will download and process)
  1. Column name guessing is now done automatically, and in most cases you should not need to provide the column_names argument
  2. INDELS are now supported, but requires you to update the reference data, see the zenodo repository. Set indel_strategy =“qc” to apply the same QC strategy to indels
  3. tidyGWAS can now estimate ancestry proportions of the cleaned summary statistics by the function decribed in https://academic.oup.com/bioinformatics/article/38/13/3477/6590645

tidyGWAS 0.9.9.2

Updates: 1. Added experimental support for automatic column name guessing 2. Added experimental support for downloading and cleaning files directly from GWAS catalog, using only study_id

tidyGWAS 0.9.9

Added reference allele frequencies to the dbSNP155 reference data from zenodo. Added the options to impute allele frequenecy from this reference data.

tidyGWAS 0.9.7

added functionality for QTL summary statistics by providing the option to not remove duplicated variants.

in addition, added the option to provide a custom reference file for allele frequency, that will impute the allele frequency from the reference file if allele frequency is missing.

Added the option to impute N using SE and allele frequency.

tidyGWAS 0.9.6

Added values that tidyGWAS detects as correct columns: 24 -> “Y” 25 -> “XY” 26 -> “MT”

tidyGWAS 0.9.0

Major revision from earlier versions. The reference data format has been updated:

  1. Compression switched from gzip to snappy (better speed, and less installation issues)
  2. GRCh37 and GRCh38 have been merged into a single file, requiring only one merge to get CHR, POS and RSID for both builds.

This has introduced breaking changes in 0.9 vs 0.8, and in speed up of ~100% ( 2.5 minutes instead of ~5minutes). The new reference data can be found at