CB2040 Lab Prep

Lab course Introduction

For these slides: https://almaan.github.io/genetech/

Teaching assistants

There are 4 of us:

• Alma Andersson (almaan@kth.se)
• Ludvig Larsson (ludvig.larsson@scilifelab.se)
• Sami Saarenpää (sami.saarenpaa@scilifelab.se)
• Trang Le (trangle@kth.se)

A word on emails

Emails are welcome, but hands-on help with the exercises will mainly be limited to the lab sessions. You can also reach us on Canvas, but slower response times.

NOTE: Tag emails with [CB2040] in the subject line.
For example: "[CB2040] Question about lab 2"

Today

• Course structure, formalities, access to information, grading and deadlines
• Overview of tools/software
• Installation of tools and software
• Question time!

Course Structure

In this course 4 labs in total, one "main TA" for each:

1. Introduction to R [Alma]
2. Bulk RNA-seq analysis [Sami]
3. Single Cell RNA-seq analysis [Ludvig]
4. Spatial Transcriptomics analysis [Alma]

All labs are in R, we assume that you have:

• access to a computer
• (zoom installed)
• no prior knowledge of R
• some familiarity with the terminal

Formalities

Workflow:

• Labs are written in R markdown (more on this later)
• Download exercises –> modify files in-place –> hand in modified file
• You hand in the modified file via Canvas
• The labs are P/F

Deadlines:

• Hand-in: 1 week after scheduled lab session
• Grading : 1 week after hand-in deadline
• Revision : 1 week after grading deadline

NOTE: Late reports will be corrected, but you will get a set of additional exercises to solve. An incomplete report is also considered late.

Working in pairs?:

• Strongly Encouraged to work alone
  • Helping each other out is also encouraged!
• Pairs are ok, but remember pair = 2 people
• Hand in identical copies
• State both names in : file and comments

Finding Information

All information you need can be found at:

https://github.com/almaan/genetech

• I will refer to this as "the course repo".
• Bookmark or save the link, we will be using it today
• All information is there

A word about zoom

• Make sure you have the latest version : download
• We will have a zoom room open in case you are sick or can't make it

R

What?

• Programming language for "statistical computing"
• Released in 1993
• Open Source –> we like free stuff!
• More than 15,000 additional packages! (add-ons)
• Interpreted language –> very easy to work with!

• Example C++ : Compiled language

      std::vector<double> vec1 = {1,2,3,4};
      std::vector<double> vec2 = {5,6,7,8};
      double prod = 0;
  
      for (int i = 0; i<= vec1.size(); i++){
        prod += vec1[i]*vec2[i];
      }
  

  Example R : Interpreted language

    vec1 <- c(1,2,3,4)
    vec2 <- c(5,6,7,8)
    prod <- vec1 %*%vec2
  

Why?

• One the most popular languages within the bioinformatic community
• Extremely well designed for "data wrangling"
• Has a huge ecosystem of packages for bioinformatic analysis
• Superior graphics system (ggplot) - don't underestimate this…
• Often requested that PhD and job applicants know one of Python or R.

Rstudio

What is it?

• When editing code it's common to use an IDE (Integrated Development Environment)
  • This is like a "Microsoft Word" for code
• IDEs are designed to facilitate programming; there are plenty of them
  • Personal favorite is Emacs (first release 1976)
• We will be using Rstudio
  • an IDE specifically designed for R
  • also very user friendly
• In short, Rstudio is the tool you will be using to edit your code.
  • Remember, just as "Miscrosoft Word ≠ English", Rstudio ≠ R

Features of Rstudio

• Create a new file by: File –> New File –> R Script
• You should see something like this:

• There are some key features I want to point out: Source, Console, Terminal, and Environment
• One more feature (knitting) that I will come back to

GitHub

What?

How::Clone labs from GitHub

• Clone repo : [Course Material] Online –> Local
• Go to a folder where you want to save your files
  • We call the path to this folder GENETECH_PATH
• Make a new folder called genetech in GENETECH_PATH
• cd into the genetech folder and do:
  git clone https://github.com/almaan/genetech.git code
• cd into the (newly cloned) code folder and do:
  git checkout cb2040 <– this puts you on the right branch

Docker

What?

• Open source containerization platform
• Allows you to bundle up OS, programs, packages, and whatever into an image that you can share
  • Like a portable OS, more robust and lightweight than Virtual Machines
• Enables:
  • Reproducibility (IMPORTANT)
  • Sharing of analysis workflows

Why?

• Previous years we've spent a lot of time on installation and software issues
• You don't think it's fun, we don't think it's fun
• Using Docker, we minimize the risk of these issues
  • Focus on learning material rather than install
• I have created a Docker environment with all R packages and rstudio
• Commonly used in research, good to be familiar with

How::Download Docker?

• Go to the GitHub course repo : link
  • Pick the master branch

• Download and install Docker
• To test if successful, in terminal do: docker ps

How::Download Docker Image?

• Download the docker image (we will create a container from this)
  • docker pull almaand/genetech:TAG
• Check if successful with the command:
  • docker images

How::Build container and run Docker

• Build your container and bind your genetech to a volume:
  docker run -d -p 1337:8787 --name genetech -e USER='genetech' -e PASSWORD='genetech' --memory=6g --mount type=bind,source=GENETECH_PATH,target=/home/genetech/code -e ROOT=TRUE almaand/genetech:TAG
• Check if successful with: docker container ls -a
  • You should see a container called genetech

How::Access rstudio

• Having started a genetech container
• Go to your favorite browser and enter:
  localhost:1337
• You will be prompted for credentials:
  • Username: genetech
  • Password: genetech
• You can only open files that are in mounted folders

How::Start and stop a container

• To stop a running contianer do: docker stop CONTAINER_ID
• To start a running contianer do: docker start CONTAINER_ID

The Labs

Working with the code

To get an idea of how you work with the code, we will have a look at Lab 1

• Go to Rstudio (in browser –> localhost:1337)
• Do: File–>Open File
• Choose the file /home/code/labs/ex1/main.Rmd

Changing author name

At the top of the file you will see:

  ---
  title: "Lab 1 - Introduction to R"
  author: "Alma Andersson"
  date: "17-09-2021"
  output:
    tufte::tufte_html: default 
  ---

Change the author field to your name(s)

Code chunks

{r,eval = TRUE, echo=TRUE}
# bind the values 1337 to the variable fancy_name_1
fancy_name_1 <- 1337
# print fancy_name_1
print(sprintf("fancy_name_1 is : %d",fancy_name_1))

• The "boxes" with code are called chunks.
• If you "run" a chunk (press green arrow) the code inside will be executed
• This is where you will put in code (when prompted)
• Don't change the settings (in curly brackets)

Knitting

• R markdown –> html and pdf documents.
• This process is called knitting

• To knit your document, press the yarn symbol that says "Knit"

  
• Your rendered document should show up in the "Viewer" pane
• You should always hand in a knitted report (html)
• If your document does not knit successfully, something is wrong.

Wrap up

• First lab session: 08:00-13:00, 2021/09/17
• Questions are welcome!
• We will hang around a bit after this session.

Thanks for the attention!