Metadata associated with the published article: Prediction of the functional impact of missense variants in BRCA1 and BRCA2 with BRCA-ML

SummaryThe data described in this metadata record underlie the figures and supplementary tables in the related manuscript. The data are in silico predictions of missense variants of uncertain significance (VUS) in the BRCA1 and BRCA2 genes (BReast CAncer genes), generated by training and evaluating hundreds of machine learning algorithms based on results from validated functional assays to better predict missense variants in these genes as damaging or neutral.Data descriptionThe primary data output of the related publication is Supplemental Data Set 1 (as it appears in the article), which contains all possible SNVs (single nucleotide variants) in both BRCA1 and BRCA2 genes. In the GitHub repository and the zip file that appears as part of this metadata record, this data output is named as TableS2.tsv, in the following location: BRCA-ML\results\TableS2.tsv.The AutoML approach was employed, with the R (version 3.4.2) CRAN - Package h2o.ai (version: 3.16.0.2). This is available at https://CRAN.R-project.org/package=h2o (Accessed: 29th January 2018).Data accessThe data and code are available at GitHub here: https://github.com/Steven-N-Hart/BRCA-MLAdditionally, a static version of the data and code as it was at the time the related publication was published is available in the zipped folder attached to this metadata record.The data is shared under an MIT Licence.