After running registration function
scale_and_register_data() as shown in the registering
data article, users can summarise and visualise the results as
illustrated in the figure below.
The total number of registered and non-registered genes can be
obtained by running function
registration_results$model_comparison_df as an
summary_model_comparison() returns a list which
contains three different objects:
df_summarycontains result summaries of the registration results,
registered_genesis list of gene accessions which were successfully registered, and
non_registered_genesis a list of non-registered gene accessions.
# Get all of summary <- summary_model_comparison(registration_results$model_comparison_df) all_summary $df_summary %>% all_summary::kable() knitr
|Stretch||1.5, 2, 2.5, 3|
The list of gene accessions which were registered can be viewed by calling:
$registered_genes all_summary#>  "BRAA04G005470.3C" "BRAA09G045310.3C" "BRAA03G051930.3C" "BRAA06G025360.3C" #>  "BRAA02G043220.3C" "BRAA03G023790.3C" "BRAA05G005370.3C" "BRAA02G018970.3C" #>  "BRAA07G030470.3C" "BRAA07G034100.3C"
plot_registration_results() allows users to
plot registration results of the genes of interest.
# Plot registration result plot_registration_results( $imputed_mean_df, registration_resultsncol = 3 )
Users also have an option to include information or label on the plot whether particular genes are registered or not, as well as the registration parameters by include model comparison data frame as shown below.
# Plot registration result plot_registration_results( $imputed_mean_df, registration_results$model_comparison_df, registration_resultsncol = 3, sync_timepoints = TRUE )
Notice that to only include same time points between samples, users
sync_timepoints = TRUE.
After registering the sample data, users can compare the overall
similarity before and after registering using the function
<- calculate_between_sample_distance( sample_distance registration_results,accession_data_ref = "Ro18" )
calculate_between_sample_distance() returns a
list of seven data frames:
distance_mean_dfis distance of mean expression values.
distance_scaled_mean_dfis distance of scaled mean expression (all genes).
distance_scaled_mean_df_only_nonregis distance of scaled mean expression (only not-registered genes).
distance_scaled_mean_df_only_regis distance of scaled mean expression (only registered genes).
distance_registered_dfis distance of registered & scaled mean expression (all genes).
distance_registered_df_only_regis distance of registered & scaled mean expression (only registered genes).
Each of these data frames above can be visualised using function
# Plot heatmap of mean expression profiles distance before scaling plot_heatmap(sample_distance$distance_mean_df)
# Plot heatmap of mean expression profiles distance after scaling plot_heatmap(sample_distance$distance_scaled_mean_df)
# Plot heatmap of mean expression profiles distance after registration process plot_heatmap( $distance_registered_df_only_reg, sample_distancesame_max_timepoint = TRUE, same_min_timepoint = TRUE )