In KopfLab/turnoveR: Processing tools for protein degradation experiments.

KEGG examples

library(tidyverse)
library(turnoveR)
knitr::opts_knit$set(root.dir = "..") 

Species

# available species
species <- tor_fetch_kegg_species()
species
filter(species, str_detect(species, "K-12"))

Pathways

# pathways for species
pathways <- tor_fetch_kegg_pathways_for_species(org_id = "eco")
pathways

Pathway Details

# additional information for pathways
details <- 
  pathways %>% 
  # focus on those that concern biosynthesis
  filter(str_detect(pathway, "biosynthesis")) %>% 
  # download pathway images and kgml files into kegg folder
  tor_fetch_kegg_pathway_maps() %>% 
  # fetch pathway details
  tor_fetch_kegg_pathway_details()
details

# loadsd of information in this data frame, e.g. descriptions
details %>% select(description)

# or compounds involved
details %>% unnest(compound) %>% 
   select(kegg_id, compound, pathway_map)

Experimental: KEGGgrpah

Download is clusmy but now that we have tor_fetch_kegg_pathway_maps() this looks promising for manipulating KEGG graphs... although maybe better to just work with the KEGG image API itself? If we can figure out how to POST parameters to the highlight genes and plot parameters over the 3D graph.

# how to modify this easily? to highlight the releavant genes online
# to make something like this http://www.genome.jp/tmp/search_pathway151821866943091/hsa05200_color_pathway_css3d.html
# from http://www.genome.jp/kegg/tool/map_pathway3.html (numerical values option)
# that would be ideal

url <- mark.pathway.by.objects("path:eco00260",
                       c("eco:b0002", "eco:c00263"))
        # if(interactive()){
        #     browseURL(url)
        # }

# this doesn't actually work properly
url <- color.pathway.by.objects("path:eco00260",
                       c("eco:b0002", "eco:c00263"),
                       c("#ff0000", "#00ff00"), c("#ffff00", "yellow"))

# the following is unnecssary with our new tor_fetch_kegg_pathway_maps() function
#getKGMLurl("00020", organism = "eco")
#retrieveKGML("00020", organism = "eco", destfile = file.path(maps_folder, "eco00020.xml"), method = "internal")
#retrieveKGML("04110", organism = "hsa", destfile = file.path(maps_folder, "hsa04110.xml"), method = "internal")
#retrieveKGML("00020", organism = "hsa", destfile = file.path(maps_folder, "hsa00020.xml"), method = "internal")

library(KEGGgraph)
maps_folder <- file.path("kegg")
kgml_map <- parseKGML(file.path(maps_folder, "eco00020.xml"))
kgml_map_graph <- parseKGML2Graph(tmp, expandGenes=TRUE)
kgml_map
kgml_map_graph

Experiment: KEGGprofile

Note: it seems this package just draws over the PNGs of the pathways...not really what I was hoping for.

# download every single relevant pathway map (this takes minutes)
library(KEGGprofile)
download_KEGGfile(pathway_id=pathways$kegg_pathway_id %>% str_replace("path:eco", "") %>% unique(),
                  species='eco', target_dir = maps_folder)
download_KEGGfile(pathway_id = "00020", species = "eco", target_dir = maps_folder)

# xml file
parse_XMLfile("00020", species = "eco", database_dir = maps_folder)
parse_XMLfile("04110", species = "hsa", database_dir = system.file("extdata", package = "KEGGprofile"))

XML::xmlTreeParse(file.path("test_scripts", "kegg", "eco00051.xml"))

data(pro_pho_expr)
data(pho_sites_count)
#type='lines'
col <-
  col_by_value(pho_sites_count,
               col = colorRampPalette(c('white', 'khaki2'))(4),
               breaks = c(0, 1, 4, 10, Inf))
temp <-
  plot_pathway(
    pro_pho_expr,
    bg_col = col,
    line_col = c("brown1", "seagreen3"),
    groups = c(rep("Proteome ", 6),
               rep("Phosphoproteome ", 6)),
    magnify = 1.2,
    species = 'hsa',
    database_dir = maps_folder,
    pathway_id = "04110",
    max_dist = 5
  )
#type='bg'
pho_expr <- pro_pho_expr[, 7:12]
temp <- apply(pho_expr, 1, function(x)
  length(which(is.na(x))))
pho_expr <- pho_expr[which(temp == 0), ]
col <-
  col_by_value(pho_expr, col = colorRampPalette(c('green', 'black', 'red'))(1024), range =
                 c(-6, 6))
temp <-
  plot_pathway(
    pho_expr,
    type = "bg",
    bg_col = col,
    text_col = "white",
    magnify = 1.2,
    species = 'hsa',
    database_dir =  maps_folder,
    pathway_id = "04110"
  )
#Compound and gene data
set.seed(124)
testData1 <-
  rbind(rnorm(6),
        rnorm(6),
        rnorm(6),
        rnorm(6),
        rnorm(6),
        rnorm(6),
        rnorm(6),
        rnorm(6))
row.names(testData1) <-
  c("4967",
    "55753",
    "1743",
    "8802",
    "47",
    "50",
    "cpd:C15972",
    "cpd:C16255")
colnames(testData1) <-
  c("Control0",
    "Control2",
    "Control5",
    "Sample0",
    "Sample2",
    "Sample5")
temp <-
  plot_pathway(
    testData1,
    type = "lines",
    line_col = c("brown1", "seagreen3"),
    groups = c(rep("Control", 3), rep("Sample", 3)),
    magnify = 1.2,
    species = 'hsa',
    database_dir =  maps_folder,
    pathway_id = "00020",
    max_dist = 2
  )
testData2 <- testData1[, 4:6] - testData1[, 1:3]
col <-
  col_by_value(testData2, col = colorRampPalette(c('green', 'black', 'red'))(1024), range =
                 c(-2, 2))
temp <-
  plot_pathway(
    testData2,
    type = "bg",
    bg_col = col,
    text_col = "white",
    magnify = 1.2,
    species = 'hsa',
    database_dir = system.file("extdata", package = "KEGGprofile"),
    pathway_id = "00020"
  )

Experiment: using iPath3

Potentially use this interface: https://pathways.embl.de/tools.cgi

-- could use the width of the pathway lines to highlight the turnover rates / maybe the alpha as well? should work with uniprot numbers and kegg id numbers

KopfLab/turnoveR documentation built on May 28, 2019, 3:33 p.m.