Missing value imputation

Introduction

We can use masscleaner for missing value (MV) imputation.

First, we need to prepare samples for masscleaner.

library(masscleaner)
library(massdataset)
library(tidyverse)

Data preparation

Load the data in previous step.

load("peak_tables/POS/object")

get_mv_number(object)
#> [1] 0
head(massdataset::get_mv_number(object, by = "sample"))
#> sample_QC_01    sample_01    sample_02    sample_06    sample_07    sample_11 
#>            0            0            0            0            0            0
head(massdataset::get_mv_number(object, by = "variable"))
#>  M70T53_POS M70T527_POS M71T775_POS M71T669_POS M71T715_POS  M71T54_POS 
#>           0           0           0           0           0           0

head(massdataset::get_mv_number(object, by = "sample", show_by = "percentage"))
#> sample_QC_01    sample_01    sample_02    sample_06    sample_07    sample_11 
#>            0            0            0            0            0            0
head(massdataset::get_mv_number(object, by = "variable"), show_by = "percentage")
#>  M70T53_POS M70T527_POS M71T775_POS M71T669_POS M71T715_POS  M71T54_POS 
#>           0           0           0           0           0           0

Impute missing values

zero

object_zero = 
  impute_mv(object = object, method = "zero")
get_mv_number(object_zero)
#> [1] 0

KNN

object = 
  impute_mv(object = object, method = "knn")
get_mv_number(object)
#> [1] 0

More methods can be found ?impute_mv().

Note

If there are blank samples in dataset, we use different method to impute missing values.

For Blank samples, just use the zero.

For non-Blank samples, just use the knn or other method

Save data for next analysis.

save(object, file = "peak_tables/POS/object")

Session information

sessionInfo()
#> R Under development (unstable) (2022-01-11 r81473)
#> Platform: x86_64-apple-darwin17.0 (64-bit)
#> Running under: macOS Big Sur/Monterey 10.16
#> 
#> Matrix products: default
#> BLAS:   /Library/Frameworks/R.framework/Versions/4.2/Resources/lib/libRblas.0.dylib
#> LAPACK: /Library/Frameworks/R.framework/Versions/4.2/Resources/lib/libRlapack.dylib
#> 
#> locale:
#> [1] en_US.UTF-8/en_US.UTF-8/en_US.UTF-8/C/en_US.UTF-8/en_US.UTF-8
#> 
#> attached base packages:
#> [1] stats     graphics  grDevices utils     datasets  methods   base     
#> 
#> other attached packages:
#>  [1] forcats_0.5.1      stringr_1.4.0      dplyr_1.0.7        purrr_0.3.4       
#>  [5] readr_2.1.1        tidyr_1.1.4        tibble_3.1.6       ggplot2_3.3.5     
#>  [9] tidyverse_1.3.1    magrittr_2.0.1     tinytools_0.9.1    massdataset_0.99.1
#> [13] masscleaner_0.9.2 
#> 
#> loaded via a namespace (and not attached):
#>   [1] colorspace_2.0-2      rjson_0.2.21          ellipsis_0.3.2       
#>   [4] class_7.3-20          leaflet_2.0.4.1       rprojroot_2.0.2      
#>   [7] circlize_0.4.14       GlobalOptions_0.1.2   fs_1.5.2             
#>  [10] clue_0.3-60           rstudioapi_0.13       proxy_0.4-26         
#>  [13] ggrepel_0.9.1         lubridate_1.8.0       fansi_1.0.0          
#>  [16] mvtnorm_1.1-2         xml2_1.3.3            codetools_0.2-18     
#>  [19] doParallel_1.0.16     cachem_1.0.6          impute_1.69.0        
#>  [22] robustbase_0.93-8     knitr_1.37            itertools_0.1-3      
#>  [25] jsonlite_1.7.2        broom_0.7.11          dbplyr_2.1.1         
#>  [28] cluster_2.1.2         png_0.1-7             missForest_1.4       
#>  [31] rrcov_1.6-0           compiler_4.2.0        httr_1.4.2           
#>  [34] backports_1.4.1       assertthat_0.2.1      fastmap_1.1.0        
#>  [37] lazyeval_0.2.2        cli_3.1.0             htmltools_0.5.2      
#>  [40] tools_4.2.0           gtable_0.3.0          glue_1.6.0           
#>  [43] Rcpp_1.0.7            Biobase_2.55.0        cellranger_1.1.0     
#>  [46] jquerylib_0.1.4       pkgdown_2.0.1         vctrs_0.3.8          
#>  [49] iterators_1.0.13      crosstalk_1.2.0       xfun_0.29            
#>  [52] rvest_1.0.2           openxlsx_4.2.5        lifecycle_1.0.1      
#>  [55] DEoptimR_1.0-10       MASS_7.3-55           scales_1.1.1         
#>  [58] ragg_1.2.1            pcaMethods_1.87.0     clisymbols_1.2.0     
#>  [61] hms_1.1.1             parallel_4.2.0        RColorBrewer_1.1-2   
#>  [64] ComplexHeatmap_2.11.0 yaml_2.2.1            memoise_2.0.1        
#>  [67] pbapply_1.5-0         yulab.utils_0.0.4     sass_0.4.0           
#>  [70] stringi_1.7.6         S4Vectors_0.33.10     desc_1.4.0           
#>  [73] pcaPP_1.9-74          foreach_1.5.1         randomForest_4.6-14  
#>  [76] e1071_1.7-9           BiocGenerics_0.41.2   zip_2.2.0            
#>  [79] BiocParallel_1.29.10  shape_1.4.6           rlang_0.4.12         
#>  [82] pkgconfig_2.0.3       systemfonts_1.0.3     matrixStats_0.61.0   
#>  [85] evaluate_0.14         lattice_0.20-45       patchwork_1.1.1      
#>  [88] htmlwidgets_1.5.4     tidyselect_1.1.1      robust_0.6-1         
#>  [91] ggsci_2.9             plyr_1.8.6            R6_2.5.1             
#>  [94] IRanges_2.29.1        snow_0.4-4            generics_0.1.1       
#>  [97] fit.models_0.64       DBI_1.1.2             withr_2.4.3          
#> [100] haven_2.4.3           pillar_1.6.4          modelr_0.1.8         
#> [103] crayon_1.4.2          utf8_1.2.2            plotly_4.10.0        
#> [106] tzdb_0.2.0            rmarkdown_2.11        GetoptLong_1.0.5     
#> [109] grid_4.2.0            readxl_1.3.1          data.table_1.14.2    
#> [112] reprex_2.0.1          digest_0.6.29         gridGraphics_0.5-1   
#> [115] textshaping_0.3.6     stats4_4.2.0          munsell_0.5.0        
#> [118] viridisLite_0.4.0     ggplotify_0.1.0       bslib_0.3.1

Xiaotao Shen (https://www.shenxt.info/)