data-raw/clue_moa_list.R

In girke-lab/signatureSearch: Environment for Gene Expression Searching Combined with Functional Enrichment Analysis

## code to prepare `clue_moa_list` dataset goes here

## GO to QUERY tool in CLUE website (https://clue.io/query), use example query 
## signature or custom one (e.g. Peters) to query the Touchstone database (Version: 1.0.1.1).
## Go to Query History to download the result txt file to the local under data 
## direcotry as "conn.txt"
library(readr)
source("getPertInfo.R")
conn_res <- read_tsv("data/conn.txt")
lincs_drugs <- unique(as.character(conn_res[,"Name"]))
#lincs_drugs <- readLines("~/insync/project/longevityTools_eDRUG/data/lincs_drugs.txt") # 2512

## Access to the LINCS/CLUE api and clue.io requires a user key, which can be obtained 
## from the LINCS by registering at https://clue.io (free for non-commercial use). 
lincs_pert_info <- getPertInfo(lincs_drugs, user_key="yourUserKey") # Take some time
saveRDS(lincs_pert_info, "data/lincs_pert_info_api.rds")
#lincs_pert_info <- readRDS("~/insync/project/lincs_gse92742_dataset_analysis/data/lincs_pert_info_api.rds")

## add target number column to lincs_pert_info
rep_num <- sapply(lincs_pert_info$moa, length)
pert_moa <- na.omit(unique(data.frame(pert_iname=rep(lincs_pert_info$pert_iname, rep_num),  
                                      moa=unlist(lincs_pert_info$moa), stringsAsFactors = FALSE)))
moa_list <- split(pert_moa$pert_iname, pert_moa$moa)
idx <- sapply(moa_list, length)>1
moa_list <- moa_list[idx] # 346
clue_moa_list <- moa_list[names(moa_list)!="-666"] #345 
usethis::use_data("clue_moa_list")

#####################################
## Update CLUE Touchstone MOA List ##
#####################################

# Get MOA and targets info of touchstone compounds
## Export the touchstone compounds annotation table from https://clue.io/touchstone
## after selecting `Name`, `MoA`, `Target` columns and perturbagen type as 
## `Compounds`. At the time of downloading on Dec 21, 2020, the database version 
## is 1.1.1.2. There are 2429 total compounds.
ts <- read_tsv("data/touchstone_cmp_moa_targets.txt") # 2429 x 3
ts <- unique(ts[,c("Name", "MoA")]) # 2414 x 2
moa_sep <- lapply(ts$MoA, function(x) unlist(strsplit(x, split=", ")))
drug_moa_df <- do.call(rbind, lapply(seq_along(moa_sep), function(i){
    data.frame(drug=ts$Name[i], moa=moa_sep[[i]])
}))
drug_moa_df <- unique(drug_moa_df[drug_moa_df$moa != "-666",])
moa_list <- split(drug_moa_df$drug, drug_moa_df$moa) # 701
drug2moa <- split(drug_moa_df$moa, drug_moa_df$drug)
saveRDS(moa_list, "data/clue_ts_moa_list_1.2.rds")
# moa_list <- readRDS("~/insync/project/discover_paper_analysis/data/clue_ts_moa_list_1.2.rds")

clue_moa_list <- moa_list
usethis::use_data(clue_moa_list, overwrite=TRUE)