johrpan/geposanui
1
0
Fork 0
mirror of https://github.com/johrpan/geposanui.git synced 2025-10-26 19:27:24 +01:00
Code Issues Projects Releases Packages Wiki Activity Actions

Base clusteriness on species count

Browse source
This commit is contained in:
Elias Projahn 2021-10-04 09:19:38 +02:00
parent 8b727a0329
commit 397b8d0ba2
2 changed files with 7 additions and 7 deletions
Download patch file Download diff file Expand all files Collapse all files

12
clustering.R
View file

@ -7,9 +7,10 @@ library(rlog)
#' This function will cluster the data using `hclust` and `cutree` (with the #' This function will cluster the data using `hclust` and `cutree` (with the
#' specified height). Every cluster with at least two members qualifies for #' specified height). Every cluster with at least two members qualifies for
#' further analysis. Clusters are then ranked based on their size in relation #' further analysis. Clusters are then ranked based on their size in relation
#' to the total number of values. The return value is a final score between #' to the total number of possible values (`n`). The return value is a final
#' zero and one. Lower ranking clusters contribute less to this score. #' score between zero and one. Lower ranking clusters contribute less to this
clusteriness <- function(data, height = 1000000) { #' score.
clusteriness <- function(data, n, height = 1000000) {
# Cluster the data and compute the cluster sizes. # Cluster the data and compute the cluster sizes.
tree <- hclust(dist(data)) tree <- hclust(dist(data))
@ -19,7 +20,6 @@ clusteriness <- function(data, height = 1000000) {
# Compute the "cluteriness" score. # Compute the "cluteriness" score.
score <- 0.0 score <- 0.0
n <- length(data)
for (i in seq_along(cluster_sizes)) { for (i in seq_along(cluster_sizes)) {
cluster_size <- cluster_sizes[i] cluster_size <- cluster_sizes[i]
@ -70,11 +70,11 @@ process_clustering <- function(distances, species_ids, gene_ids) {
.(species, distance) .(species, distance)
] ]
if (data[, .N] < 12) { if (data[, .N] < 10) {
next next
} }
score <- clusteriness(data[, distance]) score <- clusteriness(data[, distance], length(species_ids))
results[ results[
gene == gene_id, gene == gene_id,

2
correlation.R
View file

@ -41,7 +41,7 @@ process_correlation <- function(distances, species_ids, gene_ids,
gene_id <- gene_ids[i] gene_id <- gene_ids[i]
gene_distances <- distances[gene == gene_id] gene_distances <- distances[gene == gene_id]
if (nrow(gene_distances) < 12) { if (nrow(gene_distances) < 10) {
next next
} }