JakeConway

fromExpression <- function(vd){
vd <- list(vd)
intersections <- lapply(vd, function(x) strsplit(names(unlist(x)), "&"))
intersections <- lapply(intersections[[1]], function(x) unlist(as.list(x)))
sets <- unique(unlist(intersections))
data <- na.omit(data.frame(matrix(NA, ncol = length(sets))))
names(data) <- sets
counts <- lapply(vd, function(x) unlist(x))
names(counts[[1]]) <- NULL
counts[[1]] <- as.numeric(counts[[1]])

JakeConway

fromList <- function(input){
elements <- unique(unlist(input))
data <- unlist(lapply(input, function(x){x <- as.vector(match(elements, x))}))
data[is.na(data)] <- as.integer(0); data[data != 0] <- as.integer(1)
data <- data.frame(matrix(data, ncol = length(input), byrow = F))
data <- data[which(rowSums(data) !=0), ]
names(data) <- names(input)
return(data)
}

JakeConway

setwd("/Users/jakeconway/Desktop/icgcSampleData/Set 1")
fileList <- list.files()
set1 <- data.frame(); set2 <- data.frame(); set3 <- data.frame();
for(file in fileList){
  temp_data <- read.table(file, header=TRUE, sep="\t")
  temp_data$Mutation.ID <- as.character(temp_data$Mutation.ID)
  set1 <- rbind(set1, temp_data)
  rm(temp_data)
}
setwd("/Users/jakeconway/Desktop/icgcSampleData/Set 2")

JakeConway

require(jsonlite)
require(curl)

#An example of the function input. 
#Specify the project name, fields, and number of entries(size) to pull in each list

data <- list(list(project = "THCA-US", fields = c("id", "mutation", "chromosome", "start", "end"), size = 6659),
             list(project = "THCA-SA", fields = c("id", "mutation", "chromosome", "start", "end"), size = 45126),
             list(project = "LUSC-US", fields = c("id", "mutation", "chromosome", "start", "end"), size = 65063),
             list(project = "LUSC-KR", fields = c("id", "mutation", "chromosome", "start", "end"), size = 64671),

JakeConway

myplot <- function(data, colour){
  data <- data[which(data$color == colour), ]
  plot_title <- as.character(unique(data$project))
  data <- count(data["mutation"])
  data$freq <- as.numeric(data$freq)
  data$mutation <- as.character(data$mutation)
  data <- data[which(nchar(data$mutation) == 3), ]
  data <- data[order(data$mutation), ]
  bases <- strsplit(data$mutation, ">")
  original <- unlist(lapply(bases, function(x){x <- x[1]}))

JakeConway

input <- c(
  Musa_acuminata = 759,
  Phoenix_dactylifera = 769,
  Arabidopsis_thaliana = 1187,
  Oryza_sativa = 1246,
  Sorghum_bicolor = 827,
  Brachypodium_distachyon = 387,
  "Phoenix_dactylifera&Musa_acuminata" = 467,
  "Oryza_sativa&Musa_acuminata" = 29,
  "Arabidopsis_thaliana&Oryza_sativa" = 6,

JakeConway

date_to_day <- function(file_name, first_date){
  data <- read.table(file_name, header=TRUE, sep='\t', quote = NULL, row.names = NULL, check.names = FALSE)
  data$Date <- as.character(data$Date)
  data <- data[order(as.Date(data$Date, "%m/%d/%y")), ]
  data$Date <- as.Date(data$Date, "%m/%d/%y")
  if(is.null(first_date) == TRUE){
    first_date <- data[1, ]$Date
  }
  else{
    first_date <- as.Date(first_date, "%m/%d/%y")

JakeConway

import MySQLdb, itertools, pprint

#establish connection to UCSC DB
connection = MySQLdb.connect(host = 'genome-mysql.cse.ucsc.edu', user = 'genome',
                                 passwd = '', db = 'hg19')

#generate cursor so we can make queries to the DB
cursor = connection.cursor()

#grab all of the tables in the DB

JakeConway

#create myocardial infarction data frame based on Figure 1A in repo
#repo link: https://github.com/JakeConway/ClinPlots
MI_df <- data.frame(
  gene=c(NA, 'LPA', 'THBS2', 'LDLR', 'LIPC', 'ESR2', 'ESR2', 'FXN'),
  SNP_loc=c(NA, 'rs3798220', 'rs8089', 'rs14158', 'rs11630220', 'rs1271572', 'rs35410698', 'rs3793456'),
  genotype=c(NA, 'CT', 'AC', 'GG', 'AG', 'CC', 'GG', 'AA'),
  LR=c(NA, 1.86, 1.09, 2.88, 1.15, 0.73, 1.03, 0.94),
  studies=c(NA, 2, 1, 1, 1, 1, 1, 1),
  samples=c(NA, 17031, 4868, 3542, 3542, 3089, 1094, 1094),
  pt_probability=c(2.0, 3.7, 4.0, 10.6, 12.0, 9.1, 9.4, 8.9)

JakeConway

binData <- function(nBins, data, cohort, binSize) {
  counts <- c()
  for(i in seq(nBins)) {
    start <- (i-1)*binSize
    end <- i*binSize
    count <- c(which(data$start > start & data$start < end))
    count <- c(count, which(data$end > start & data$end < end))
    count <- c(count, which(data$start < start & data$end > end))
    count <- c(count, which(data$start > start & data$end < end))
    count <- length(unique(count))