Supplemental_Code_2 – R code used to calculate enrichment scores (ES) 



#This will return probability of it being below the first component number. To get p-value use 
#       1-phyper(<number of white balls drawn out of an urn with both black and white urns>-1,
#                <number of white balls in the urn>
#                <number of black balls in the urn>
#                <number of balls drawn from the urn>)
#The -1 in the first component is to make the phyper be the likelihood of the white balls being less than the number, with the 1- it gives the probablility of it being the number or higher 

#function to perform the phyper test
#dataset and enriched_in should be in the *same order* and *restricted to SNPs that are shared between both*
hyperGeo_RASQUAL<-function(dataset, enriched_in){
  
  #Count the number of white balls drawn out of the urn. These are the number of successes that are in the dataset AND STARR-seq sig
  white_drawn<-as.numeric(table(dataset$Q.value<0.05 & enriched_in$adj.P.Val<0.05)['TRUE'])
  #Count the number of white balls in the urn. This is the number of successes in dataset that were in the STARR-seq library
  white<-as.numeric(table(dataset$Q.value<0.05)['TRUE'])
  #Count the number of black balls. These are the number of snps in the library that are not successes in the dataset
  black<-as.numeric(table(dataset$Q.value<0.05)['FALSE'])
  #How many samples are being taken. AKA the number of significant STARR-seq SNPs (since this is what we want to see enrichment in)
  n_pick<-as.numeric(table(enriched_in$adj.P.Val<0.05)['TRUE'])
  
  #Calculating the probability of picking the number of white drawn or higher
  return(c((1-phyper(white_drawn-1, white, black, n_pick)), #Gives hypergeometric pvalue
           (white_drawn/((white/(white+black))*n_pick)), #Gives enrichment Score (ES)
           white)) #Gives the n of the dataset
}