get percent deviance explained for LDA model, from n bootstraps

Source: R/scgps_prediction_summary.R

the training results from training were written to the object LSOLDA_dat, the summary_prediction summarises prediction explained for n bootstrap runs and also returns the best deviance matrix for plotting, as well as the best matrix with Lasso genes and coefficients

summary_prediction_lda(LSOLDA_dat = NULL, nPredSubpop = NULL)

Arguments

LSOLDA_dat

is a list containing the training results from training
nPredSubpop

is the number of subpopulations in the target mixed population

Value

a dataframe containg information for the LDA prediction results, each column contains prediction results for all subpopulations from each bootstrap run

Examples

c_selectID<-1
day2 <- day_2_cardio_cell_sample
mixedpop1 <-new_scGPS_object(ExpressionMatrix = day2$dat2_counts, 
GeneMetadata = day2$dat2geneInfo, CellMetadata = day2$dat2_clusters)
day5 <- day_5_cardio_cell_sample
mixedpop2 <-new_scGPS_object(ExpressionMatrix = day5$dat5_counts, 
    GeneMetadata = day5$dat5geneInfo, CellMetadata = day5$dat5_clusters)
genes <-training_gene_sample
genes <-genes$Merged_unique
LSOLDA_dat <- bootstrap_prediction(nboots = 1,mixedpop1 = mixedpop1, 
mixedpop2 = mixedpop2, genes=genes, c_selectID, listData =list(),
    cluster_mixedpop1 = colData(mixedpop1)[,1],
    cluster_mixedpop2=colData(mixedpop2)[,1])

#> Warning: variables are collinear
#> Warning: variables are collinear
#> Warning: variables are collinear
#> Warning: variables are collinear
#> Warning: variables are collinear
#> Warning: variables are collinear
#> Warning: variables are collinear
#> Warning: variables are collinear
#> Warning: variables are collinear
#> Warning: variables are collinear
#> Warning: variables are collinear
#> Warning: variables are collinear
#> Warning: variables are collinear
#> Warning: variables are collinear
#> Warning: variables are collinear
#> Warning: variables are collinear
#> Warning: variables are collinear
#> Warning: variables are collinear
#> Warning: variables are collinear
#> Warning: variables are collinear
#> Warning: variables are collinear
#> Warning: variables are collinear
#> Warning: variables are collinear
#> Warning: variables are collinear
#> Warning: variables are collinear
#> Warning: variables are collinear
#> Warning: variables are collinear
#> Warning: variables are collinear
#> Warning: variables are collinear
#> Warning: variables are collinear
#> Warning: variables are collinear
#> done training for bootstrap 1, moving to prediction...
summary_prediction_lda(LSOLDA_dat=LSOLDA_dat, nPredSubpop=4)

#>                 V1                                names
#> 1 0.53475935828877 LDA for subpop 1 in target mixedpop2
#> 2 23.5714285714286 LDA for subpop 2 in target mixedpop2
#> 3 1.50375939849624 LDA for subpop 3 in target mixedpop2
#> 4              2.5 LDA for subpop 4 in target mixedpop2