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

Retrieve input data using biomaRt

Browse source
This commit is contained in:
Elias Projahn 2021-09-16 00:06:54 +02:00
parent 040aabc610
commit 1cea6c3631
205 changed files with 187 additions and 3296961 deletions
Download patch file Download diff file Expand all files Collapse all files

19
process.R
View file

@ -6,28 +6,27 @@ library(rlog)
#' The return value will be a table containing genes and data to take in
#' account when regarding them as TPE-OLD candidates.
#'
#' @param input Data from [`load_input()`].
#' @param distances Gene distance data to use.
#' @param species_ids IDs of species to include in the analysis.
process_input <- function(input, species_ids) {
results <- data.table(gene = input$genes$id)
gene_ids <- input$genes[, id]
#' @param gene_ids Genes to include in the computation.
process_input <- function(distances, species_ids, gene_ids) {
results <- data.table(gene = gene_ids)
gene_count <- length(gene_ids)
for (i in seq_along(gene_ids)) {
gene_id <- gene_ids[i]
log_info(sprintf("Processing gene %i/%i (%i)", i, gene_count, gene_id))
log_info(sprintf("Processing gene %i/%i (%s)", i, gene_count, gene_id))
distances <- input$distances[
data <- distances[
species %chin% species_ids & gene == gene_id,
.(species, distance)
]
if (distances[, .N] < 12) {
if (data[, .N] < 12) {
next
}
clusters <- hclust(dist(distances[, distance]))
clusters <- hclust(dist(data[, distance]))
clusters_cut <- cutree(clusters, h = 1000000)
# Find the largest cluster
@ -36,7 +35,7 @@ process_input <- function(input, species_ids) {
which.max(tabulate(match(clusters_cut, cluster_indices)))
]
cluster <- distances[which(clusters_cut == cluster_index)]
cluster <- data[which(clusters_cut == cluster_index)]
results[
gene == gene_id,