README.md

In RemiMahmoud/kclique: Find maximal k-cliques in a k-partite graph

kclique

The goal of kclique is to enumerate all maximal kcliques within a k-partite graph.

Installation

You can install the development version of kclique from GitHub with:

# install.packages("devtools")
devtools::install_github("RemiMahmoud/kclique")

Example

This is a basic example which shows you how to solve a common problem:

library(kclique)

Load toy data

data("data_example", package ="kclique")
knitr::kable(head(data_example))

| factor_A | factor_B | factor_C | |:---------|:---------|:---------| | A02 | B1 | C01 | | A15 | B1 | C01 | | A04 | B1 | C01 | | A01 | B1 | C01 | | A11 | B1 | C01 | | A13 | B1 | C01 |

As an example, imagine a set of three factors, with various modalities and combinations of modalities. You can represent the co-occurence of the modalities by a graph.

Set graph

# get graph from the data
graph_example <- get_graph_from_dat(data_example)

Plot data

We can plot the graph, using one color for each factor

plot_graph_example <- plot_graph(G = graph_example, dat = data_example)
plot_graph_example

List maximal kcliques

If we want to look for all the maximal complete factorial design, we need to enumerate all the maximal kcliques present in the graph.

kcliques_example <- maximal_kclique_enumeration(graph_example, data_example)
knitr::kable(kcliques_example,col.names = "List of kcliques")

| List of kcliques | |:-------------------------------------------| | B1;C01;C03;C11;C07;C10;A14 | | B1;C01;C03;C11;C07;A09;A14 | | B1;C01;A05;C06;C02;C08 | | B1;C01;A02;A09;A14;A05;A13;A11;A01;A04;A15 | | B1;A10;A03;C11;C05;C09 | | B1;A10;A03;C11;A14;A09 | | B1;A12;C12;C04 | | B1;A07;A09;A14;C03 | | A08;A06;B2;C01 |

Format kclique in a nice output format

tibble_kclique <- function_list_kcliques_to_tibble(kcliques_example, data_example)
knitr::kable(tibble_kclique)

| factor_A | factor_B | factor_C | |:------------------------------------|:---------|:--------------------| | A15-A14-A13-A11-A09-A05-A04-A02-A01 | B1 | C01 | | A14 | B1 | C11-C10-C07-C03-C01 | | A14-A09 | B1 | C11-C07-C03-C01 | | A05 | B1 | C08-C06-C02-C01 | | A10-A03 | B1 | C11-C09-C05 | | A14-A10-A09-A03 | B1 | C11 | | A14-A09-A07 | B1 | C03 | | A12 | B1 | C12-C04 | | A08-A06 | B2 | C01 |

We identified 9 maximal kcliques, involving various combinations of factors’ modalities.

Highlight a kclique

plot_kclique_1 <- plot_kclique(G = graph_example, dataset = data_example, tibble_kcliques = tibble_kclique, vector_kcliques = 1)
plot_kclique_1

plot_kclique_3 <- plot_kclique(G = graph_example, dataset = data_example, tibble_kcliques = tibble_kclique, vector_kcliques = 3)
plot_kclique_3

# plot_graph_example

RemiMahmoud/kclique documentation built on March 18, 2023, 1:41 p.m.