Added batch Correction as an option to processing the data

[PaulJonasJost]

• Batch Correction based on a singular column
• First batch correction -> preprocessing, though basic filtering is done before as well.
• Keep DESeq from using batch in combination with any factor already chosen
• added option to switch between processed or batch corrected processed data
• improved error catching in those two cases
• to keep in mind: deseq's vet transformation is independent of model matrix, thus batch correction here has for now no effect. This was our current approach, should we change it now?

[LeaSeep]

Still needed:

• Note of batch correction to html report
• Bacth correction as part of preprcoessing in downloadable snippets
• sva as new package to renv (I updated it locally but there where also other update - would suggest to open a separate issues + PR to first check whether everything works fine with the new updated packakes)
• When testing - sva puts out interesting info messages : e.g.: Found4batches Adjusting for0covariate(s) or covariate level(s) Standardizing Data across genes Fitting L/S model and finding priors Finding parametric adjustments Adjusting the Data => maybe add to info box of general statistics
• using batch corrected data on sample correlation does not work (test dada, filtering none, batch experiment)
• PCA using batch data also did not work

Stopped further checking here

[LeaSeep]

• added an enhancement issue Report SVA batch output to GUI  #220
• forgot to add sva also new issue add sva to lock file #221

Still to do:

• Now on PCAs I get the error [1] "Do LoadingsPlot an issue?" Warning: Error in match.fun: 'ggfun("scales_transform_df")' is not a function, character or symbol 108: stop 107: match.fun 106: lapply 105: ggplotly_build 104: gg2list 103: ggplotly.ggplot 101: renderPlotly [R/pca/server.R#472] 100: func 97: shinyRenderWidget 96: func 83: renderFunc 82: output$PCA-PCA_plot 1: runApp (using 'cell' as batch and 'filter only'

• Single Gene Vis . shows error on top of sidebar missing value where TRUE/FALSE needed (can't choose batch corrected)

[PaulJonasJost]

• Now on PCAs I get the error [1] "Do LoadingsPlot an issue?" Warning: Error in match.fun: 'ggfun("scales_transform_df")' is not a function, character or symbol 108: stop 107: match.fun 106: lapply 105: ggplotly_build 104: gg2list 103: ggplotly.ggplot 101: renderPlotly [R/pca/server.R#472] 100: func 97: shinyRenderWidget 96: func 83: renderFunc 82: output$PCA-PCA_plot 1: runApp (using 'cell' as batch and 'filter only'

Could not reproduce this error with test data using cell and filter only... Is it perhaps also a version thing?

• Single Gene Vis . shows error on top of sidebar missing value where TRUE/FALSE needed (can't choose batch corrected)

Solved this. Was a weird thing.