JakeConway

purity.check <- function(maf) {
  maf[which(is.na(maf$purity)), ]$purity <- 0
  samples <- length(unique(maf[which(maf$purity == 0), ]$Tumor_Sample_Barcode))
  rows <- nrow(maf[which(maf$purity == 0), ])
  maf <- maf[which(maf$purity > 0), ]
  cat("Removed", rows, "rows from", samples, "samples with purity values of 0 or NA\n")
  return(maf)
}

tcn.check <- function(maf) {

JakeConway

library(data.table)
# set directory to where file is located
setwd("/Users/jakeconway/Downloads")

# name of file
file.name <- "PR_VanAllen_Haq_Melanoma_Capture_All_Pairs.aggregated_case_sample.seg"

# let the following do the rest
data <- fread(file.name)
data <- data[order(data$Chromosome, data$Start), ]

JakeConway

pairwiseCorrelationOfGenes <- function(maf, sig.genes) {
  genes <- sig.genes[which(sig.genes$p < 0.05 & sig.genes$q < 0.3), ]$gene
  df <- as.data.frame(t(combn(genes[1:3], 2)))
  names(df) <- c("gene1", "gene2")
  apply(df, 1, function(x) {
    gene1 <- unname(x["gene1"])
    gene2 <- unname(x["gene2"])
    samples.w.gene1 <- unique(maf[which(maf$Hugo_Symbol == gene1), ]$Tumor_Sample_Barcode)
    samples.w.gene2 <- unique(maf[which(maf$Hugo_Symbol == gene2), ]$Tumor_Sample_Barcode)
    samples.w.both.genes <- intersect(samples.w.gene1, samples.w.gene2)

JakeConway

                var styleJSON = JSON.stringify(success.data).split(",");
                for(var i = 0; i < styleJSON.length; i++) {
                    styleJSON[i] = styleJSON[i].split(":");
                    var innerLength = styleJSON[i].length;
                    for(var j = 0; j < innerLength; j++) {
                        if(styleJSON[i][j].indexOf("[{") > -1) {
                            styleJSON[i][j] = styleJSON[i][j].split("{");
                            styleJSON[i][j][0] = "<span style='color: #660000;'>" + "[" + "</span>" + "<br>" +
                                "<span style='color: #660000;'>" + "{" + "</span>" + "<br>";
                            styleJSON[i][j][1] = "<span style='color: #006699;'>" + styleJSON[i][j][1] + "</span>" + ": ";

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))

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

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

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

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

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]}))