test.R

library(flexdashboard)
library(ggplot2)
library(reshape2)
library(scales)
library(tidyverse)
library(plotly)
library(FDRestimation)
library(ggrepel)
library(hrbrthemes)
library(dplyr)
library(shiny)
library(jsonlite)
library(DT)
library(patchwork)
library(network)
library(sna)
library(igraph)
library(ggraph)
library(tidygraph)
library(visNetwork)
library(stringr)
library(purrr)

ksi_gene <- read.csv('data/ksi_gene.csv')

ksi_gene_t <- data.frame(ksi_gene)
# print("test")
colnames(ksi_gene_t) <- c('From', 'To', 'Group')

k_to_group <- data.frame(ksi_gene[,c('Kinase', 'Group')])
k_to_group <- k_to_group %>% distinct(Kinase, Group, .keep_all = TRUE)



kinase_set <- unique(ksi_gene$Kinase)
substrate_set <- unique(ksi_gene$Substrate)

ks_set <- union(kinase_set, substrate_set)

ks_df <- data.frame('KS'=ks_set)

colnames(k_to_group)[which(names(k_to_group) == 'Kinase')] <- 'KS'
ks_group <- merge(x = ks_df, y = k_to_group, by = "KS",
                  all = TRUE)

ks_group$Group[is.na(ks_group$Group)] <- 'Non-kinase'
ks_id <- 1:length(ks_group$KS)
ks_group$id <- ks_id




ks_edge_list <- data.frame(ksi_gene_t[,c('From', 'To')])
ks_node_id_map <- c(as.character(ks_group$id))
names(ks_node_id_map) <- paste0("^", ks_group$KS, "$")
ks_edge_list$from <- as.integer(str_replace_all(string = ks_edge_list$From,
                                                pattern= ks_node_id_map))

ks_edge_list$to <- as.integer(str_replace_all(string = ks_edge_list$To,
                                              pattern= ks_node_id_map))

ks_ig <- graph_from_edgelist(as.matrix(ks_edge_list[,c("From", "To")]), directed = TRUE)
subg <- make_ego_graph(ks_ig, order=1, nodes=c('STK11'))[[1]]
#visIgraph(subg[[1]])


group_colors <- c("#1f77b4ff","#ff7f0eff","#2ca02cff","#d62728ff","#9467bdff","#c49c94ff","#f7b6d2ff","#c7c7c7ff","#dbdb8dff","#9edae5ff","#808080ff")
group_names <- c("TK","OTHER","CAMK","CMGC","AGC","STE","TKL","ATYPICAL","CK1","RGC","Non-kinase")

get_color <- function(name){
  gp = ks_group$Group[which(ks_group$KS == name)]
  group_colors[which(group_names == gp)]
}
# 
# V(subg)$color = sapply(V(subg)$name,get_color)
unip <- read_delim('data/uniprot.txt',delim="\t")

lapply(unip['Gene Names'], function(s){strsplit(s," ")}) -> x
unip$Primary <- unlist(map(x[[1]],1))