Numeric Matrices K-NN and PCA Imputation // Fast k-nearest neighbors (K-NN) and principal component analysis (PCA) imputation algorithms for missing values in epigenetic data or other high-dimensional numeric matrices. For PCA, a locally optimal block preconditioned conjugate gradient (LOBPCG) eigensolver with warm starts of both the eigenblock and search direction is also supported. Two complementary imputation strategies are available. Group-wise imputation (e.g., by chromosome) is recommended for Illumina DNA methylation microarrays (e.g., 450K, EPIC) and other matrices with groupable columns. A sliding window approach for K-NN or PCA imputation is recommended only for whole-genome methylation data such as whole-genome bisulfite sequencing (WGBS) or Enzymatic Methyl-seq (EM-seq). The package also supports hyperparameter tuning via repeated cross-validation. The K-NN algorithm is described in: Hastie, T., Tibshirani, R., Sherlock, G., Eisen, M., Brown, P. and Botstein, D. (1999) "Imputing Missing Data for Gene Expression Arrays". The PCA imputation is an optimized reimplementation of the imputePCA() function from the 'missMDA' package described in: Josse, J. and Husson, F. (2016) doi:10.18637/jss.v070.i01 "missMDA: A Package for Handling Missing Values in Multivariate Data Analysis".