CHANGES IN VERSION 1.7.1 ------------------------ BUG FIXED o Null eigenvectors for TBA with null variance or NA values in the training set are returned correctly CHANGES IN VERSION 1.3.7 ------------------------ UPDATED VIGNETTE o Package vignette has been updated to address the new features ALSO MODELS' P-VALUES AND RHO ARE RETURNED IN CROSS-VALIDATION o For each fold, for each gene for which a GReX could be imputed, the following values are now returned o Squared correlation of the model's predictions with training data o P-value of the model o Corrected p-value of the model with benjamini-hochberg procedure o Correlation of the model's predictions with validation data o Squared correlation of the model's predictions with validation data o P-value of the correlation test of the model's predictions with validation data CHANGES IN VERSION 1.3.5 ------------------------ P-VALUE OF THE CORRELATION TEST o P-value of the cor.test() between predicted GReX and real expression values of genes is returned in when performing cross-validation o When using affiXcanTrain in cross-validation mode, three main values for each gene are therefore returned: rho and rho squared (see changes in v 1.3.1), and the p-value of cor.test() CHANGES IN VERSION 1.3.3 ------------------------ UPDATED DOCUMENTATION o Formatting of functions documentation has been improved o Important: vignette is still outdated (AffiXcan 1.2.0) CHANGES IN VERSION 1.3.2 ------------------------ POPULATION STRUCTURE COVARIATES ARE OPTIONAL o Providing population structure covariates is now not mandatory to perform models training CHANGES IN VERSION 1.3.1 ------------------------ K-FOLD CROSS-VALIDATION IS SUPPORTED o ANOVA p-value < 0.05 to assess prediction significance is not used anymore; instead: o A k-fold cross-validation on the training dataset may be performed; k can be defined by the user o Pearson correlation coefficients (R) and determination coefficients (R^2) between predicted GReX and real expression values of genes are computed o In literature, GReX of genes for which the mean of the R^2 is above 0.01 are considered non-randomly predicted, according to the new benchmarking standards [ref] CHANGES IN VERSION 1.1.3 ------------------------ BUG FIXES o Fix a bug in generating the vignette for 'devel' version of Bioconductor