Applied Statistics and Bioinformatics with R and Bioconductor

Dates

22-26 January 2018

Where

Berlin

Overview

This course will provide biologists and bioinformaticians with practical statistical and data analysis skills to perform rigorous analysis of high-throughput biological data. The course assumes some familiarity with genomics and with R programming, but does not assume prior statistical training. It covers the statistical concepts necessary to design experiments and analyze high-dimensional data generated by high-throughput sequencing, including: exploratory data analysis, principal components analysis, unsupervised clustering, batch effects, linear modeling for differential expression, gene set analysis

Preparation

Come to the first class with the following installed:

• R and Bioconductor: www.bioconductor.org/install
• R Studio: https://www.rstudio.com/products/rstudio/download3/
• Github desktop client (or any other Github client): https://desktop.github.com/

Additionally, please create an account at www.github.com, and use it to introduce yourself at https://github.com/waldronlab/AppStatTrento/issues.

Labs

Each day will include a hands-on lab session, that students should attempt and hand in before the following day by committing to the course Github repository. A selection of labs will be reviewed the following day.

Program

Monday 22nd - Classes from 09:30 to 17:30

Session 1 - Introduction

Lecture 1: Data distributions

• random variables

• distributions

  Lecture 2: Statistical inference and sampling

• populations and samples

• Central Limit Theorem

• t-distribution

  Lab 1: Introduction to R and Bioconductor

Lab 2: Creating graphics

Tuesday 23nd - Classes from 09:30 to 17:30

Session 2- Hypothesis testing

Lecture 1: hypothesis testing concepts

• type I and II error and power
• confidence intervals

• multiple hypothesis testing: false discovery rate, familywise error rate

  Lecture 2: hypothesis testing in practice

• hypothesis tests for categorical variables (chi-square, Fisher's exact)

• Monte Carlo simulation
• permutation tests
• bootstrap simulation

• exploratory data analysis

  Lab: bootstrap simulation and permutation tests

Wednesday 24th - Classes from 09:30 to 17:30

Session 3 - Linear modeling

Lecture 1: linear modeling

• linear regression and multiple regression
• model matrix and model formulae

Lecture 2: generalized linear models for count data

• intro to generalized linear models
• logistic regression and log-linear models
• Poisson and Negative Binomial error models
• Zero-inflated models

Lab: RNA-seq differential expression workflow

Thursday 25th - Classes from 09:30 to 17:30

Session 4 - Unsupervised methods

Lecture 1: distances and PCA

• distance in high dimensions
• singular value decomposition
• principal components analysis and multidimensional scaling

Lecture 2: unsupervised clustering

• unsupervised clustering
• batch effects

Lab 1: applications of unsupervised methods to shotgun metagenomics microbiome data analysis

Lab 2: option to work on students' own data.

Friday 26th - Classes from 09:30 to 17:30

Session 5 - Gene set and multi-omics data analysis

Lecture 1 - gene set enrichment analysis

• background on gene set testing
• types and interpretations of gene set tests
• advantages and pitfalls of gene set testing

Lab 1 - gene set analysis with applications to gene expression and multi-omics experiments

Lab 2 - multi-omics data analysis

Lab 3 - option to work on students' own data.

For more information about the course, please visit our WEBSITE