clairemcwhite

# PROJECT_ID is the billable project code "word-word-number"

#  Having -life-sciences suffix may be important
export BUCKET=gs://PROJECT_ID-life-sciences

# Make a bucket (storage folder) for input and output data
# See intro to installing gsutil in this gist https://gist.github.com/clairemcwhite/ca33c9b7385ee4e2b64641353399f3ba
gsutil mb ${BUCKET}

# First do the example script

clairemcwhite

# This is extra instructions to access and query the hail tables from the pan ukbiobank
# https://pan.ukbb.broadinstitute.org/docs/hail-format/index.html
#
# Open a google cloud account, set up billing, get project id (word-word-number)

# Create a virtual machine
# Select Ubuntu 16.04
# Allow access to all Cloud API (unknown if needed)

# Need python 3.7 for hail

clairemcwhite

read_fasta <- function(fasta_filename, annot = FALSE){
    fasta <- seqinr::read.fasta(fasta_filename, as.string = TRUE)

    # Convert seqinr SeqFastadna object to data.frame
    fasta_df <- fasta %>%
                   sapply(function(x){x[1:length(x)]}) %>%
                   as.data.frame %>%
                   broom::fix_data_frame(newcol = "ID", newnames = "Sequence")

    if(annot == TRUE){

clairemcwhite

# Only needed if using theme_cowplot. Otherwise, use any theme. 
library(cowplot)

theme_cowplot_consistent_text <- function (font_size = 8) {
  
  theme_cowplot() %+replace%
       theme(strip.text = element_text(size = font_size),
             axis.text = element_text(colour = "black", size = font_size),
             plot.title = element_text(size = font_size),

clairemcwhite

library(tidyverse)

get_order <- function(df,  distmethod = "pearson", hclustmethod = "average", output_ordername = "order"){
  #Get the row ordering from a clustering
  hr <- hclust(as.dist(1-cor(t(df), method=distmethod)), method=hclustmethod)
  order <- data.frame(hr$labels[hr$order])
  order$ordering <- rownames(order)
  names(order) <- c("ID", output_ordername)
  return(order)
}

clairemcwhite

library(tidyverse) # For everything
library(gganimate) # For animating graphs
library(ggraph) # For plotting graphs
library(magick) # For exporting gif
library(tidygraph) # For manipulating graphs
library(colorspace) # For sequential_hcl
library(igraph) # For random walk and as_ids

clairemcwhite

library(tidygraph)
library(tidyverse)
library(ggraph)
library(dendextend)
library(igraph)

palette_OkabeIto <- c("#E69F00", "#56B4E9", "#009E73", "#F0E442", "#0072B2", "#D55E00", "#CC79A7", "#999999")

cut_df <- function(dendrogram, height, c){