khturner

Provision EC2 instance

• Use AMI Ubuntu Server 18.04 LTS (HVM), SSD Volume Type - ami-005bdb005fb00e791 (64-bit x86)
• At least a t2.medium
• At least 30 GB Root volume
• Set up TCP port 22 and port 8787 (RStudio server) access for my IP
• Tag with Name = turbo-rstudio
• Use the turbo-rstudio keypair
  • You already have this .pem file in your ~/Downloads/ dir, already did chmod 0400

khturner

library(tidyverse)

shared_file <- read_tsv("data/mothur/kws_final.an.shared") %>%
  select(-label, -numOtus) %>%
  # melt, onvert to relative abundance
  gather("OTU", "abundance", starts_with("Otu")) %>%
  group_by(Group) %>%
  mutate(abundance = abundance / sum(abundance)) %>% ungroup

tax_file <- read_tsv("data/mothur/kws_final.an.cons.taxonomy") %>% select(-Size) %>%

khturner

Soulless Sam Simon - A Simpsons script bot

First we're gonna get an EC2 instance up and running with the training data

# Run on an AWS c4.xlarge EC2
$ sudo apt-get update
$ sudo apt-get install r-base-core docker.io