generate_reports: Report generation

In ilangurudev/approxmapR: ApproxMAP (APPROXimate Multiple Alignment Pattern)

Report generation

Description

A function that generates reports and exports the files to the default or specified location; the default placement is in the user's document folder. This creates a main folder with two subfolders, public and private, which exports report documents to the corresponding folder based on data sensitivity.

**Public Folder**

* Pattern file: Contains the clusters, number of sequences within the cluster (n), the percentage of sequences within that cluster (n_percent), the pattern type (ex. consensus), and the resulting sequence. This also contains a truncated weighted sequence, and the unique events within the sequence.

**Private Folder**

* Alignments file: contains the sequences and their alignment within each cluster.

* All Sequences file: Contains the clusters, number of sequences within the cluster (n), the percentage of sequences within that cluster (n_percent), the pattern type (ex. consensus), and the resulting sequence. This also contains a truncated weighted sequence, the full weighted sequence, and the unique events within the sequence.

* Weighted Sequences file: Contains the clusters, number of sequences within the cluster (n), the percentage of sequences within that cluster (n_percent), and the full weighted sequence of the cluster.

Usage

  generate_reports(w_sequence_dataframe,  sil_table = NULL, html_format = TRUE,
                  output_directory = "~", end_filename_with = "",
                  sequence_analysis_details = NULL,
                  sequence_analysis_details_definitions == NULL,
                  algorithm_comparison = FALSE)

Arguments

w_sequence_dataframe	A dataframe with class "W_Sequence_Dataframe". This will be the dataframe that resulted from extracting the patterns after clustering.
sil_table	The silhouette_object which is produced by the -find_optimal_k- function that contains the silhouette Information for the K value that was selected.
html_format	A boolean value to indicate if the exports should have HTML formatting.
output_directory	The path to where the exports should be placed. This creates a folder with the name of "approxmap_results".
end_filename_with	The option of appending to the end of the default file names. This is useful if running multiple algorithms that will be exported to the same output_directory.
sequence_analysis_details	This will generate a report that includes details pertaining to the sequence analysis. This must be a list with the following structure list("algorithm" = "a string", "k_value" = a number, "time_period" = "a string", "consensus_threshold" = a number, "notes" = "Any special notes as a string.")
sequence_analysis_details_definitions	This needs to be an data frame object with column 1 being labelled "event" which contains the events in the data, while column 2 can be any label which contains the definitions or descriptions of the event.
algorithm_comparison	The option to indicate if the report being generated is one that is comparing multiple algorithms, for example the outcome of using the K-NN and K-Medoids algorithm. This function separates the *id* column using id %>% str_split("_", simplify = TRUE). If using this option the criteria is specific for the id column. The *id* column must represent the algorithm used, cluster, and number of sequences within the cluster. For example, an id should look like "kmed_cluster1_n288" where "kmed" represents the clustering algorithm used, "_cluster1" indicates the pattern came from the first cluster, and "_n228" indicates that 228 sequences were apart of cluster 1. An example of how this can be created is: formatted_kmed %>% mutate(id = paste0("kmed", "_cluster", cluster, "_n", n))

Value

Nothing is returned, only exports results.

Examples

  data("mvad")

  clustered_kmed <- mvad %>%
                          aggregate_sequences(format = "%Y-%m-%d",
                                              unit = "month",
                                              n_units = 1,
                                              summary_stats=FALSE) %>%
                          cluster_kmedoids(k = 5)

  patterns_kmed <- clustered_kmed %>%  filter_pattern(threshold = 0.5,
                                                        pattern_name = 'consensus')

  patterns_kmed %>% generate_reports(end_filename_with = "_kmed")

ilangurudev/approxmapR documentation built on March 22, 2022, 1:15 p.m.