Data analysis for single-cell genomics

This is not an advanced R-programming course

What do you need to analyse [single-cell omics] data?

You need

• Domain specific knowledege (biological field) +
• Data analysis mindset (abstraction, working with numbers) +
• Knowledge of tools (process data files, visualise, conclude)

You need to understand the concepts to ask the right questions.

Focus of the course

1. Learn concepts of high-throughput data analysis, instead of coding syntax.
2. Develop abstract thinking to find the common core in various problems.
3. Analyse publicly available single-cell transcriptomic datasets to practice some of the skills.

Concepts of high-throughput data analysis

• Learn concepts of high-throughput data analysis, instead of coding syntax
  • While syntax of base-R is stable, the syntax of R-packages or, BioInfo software is not
    • They expire in 2-3 years
  • Consequence:
    • We gonna learn a lot of machine learning concepts.
    • There is gonna be limited amount of coding.
    • Exceptions:
      • Basic coding in bash, using the linux terminal.

Develop abstract thinking to solve various problems

• Throughout the entire course we transform biological problems into numeric problems.
  • With this, we can find concepts how to generally solve them
  • Although we essentially do math, we can explain these in plain words, so you won’t be burdened with a lot of equations.
• This will allow you take one step back, and think on a higher level in whatever domain you will be.

Single-cell transcriptomics is booming, and its going to reach you

What is so special about single-cell mRNA-sequencing, that it worth your time?

• This is the most widespread method to date
• Consequence:
  • You have access to a lot of data, that you can use it for yourself.
  • The method will become available to you soon.
• We will have an outlook to other x-omics fields

Field is booming and it reaches the parasitology field

You can now mearure host-pathogen interaction in the same cell!

• Parasite - host-cell / virus - host-cell
• How do transcriptional changes in the parasite affect the host cell?

Heterogeneity of individual parasites may be a key feature

How does heterogeneity:

• Help to overcome the immune system (arms race)?
• How does it arise, and

• How is it controlled?

Access to publicly available single-cell transcriptomics dataset

They are typically required to share the data. If lucky, they provide: - a gene expression matrix (.xlsx), or you have to download the - raw data (.fastq files )

We are standing on the shoulders of anonym giants

A Disclaimer

This course material is based on hundreds of sources published on the web. Referencing each half sentence and graph was a task beyond my capacities, hereby I thank all authors making their work publicly available, and apologise to unnamed authors. Consequently, I would not like take credit for this course is “my achievement” or “my original work”. I only hope that it will be useful for learning.

Some notable sources:

• Kolodziejczyk, Aleksandra A., Jong Kyoung Kim, Valentine Svensson, John C. Marioni, and Sarah A. Teichmann. 2015. “The Technology and Biology of Single-Cell RNA Sequencing.” *Molecular Cell- 58 (4). Elsevier BV: 610–20. doi:[10.1016/j.molcel.2015.04.005](https://doi.org/10.1016/j.molcel.2015.04.005).
• https://github.com/hemberg-lab/scRNA.seq.course/
• Korflab UNIX course

Overview - Approximate schedule

1. Next generation sequencing
2. History & Concepts in sc-seq
3. scRNA-seq Experiments
4. Analysis of scRNA-seq data
   1. Data
      1. Unix intro (P)
   2. Analysis pipeline
   3. File formats
   4. Tools & software
5. Example dataset analysis with Seurat (P)
6. (Novel & experimental data analysis)

Data analysis is part of the experiment - it’s not something for later

To consult the statistician after an experiment is finished is often merely to ask him to conduct a post mortem examination. He can perhaps say what the experiment died of.”

R.A. Fisher, 1938

Your first take home message.