About us

• Mark
  • Cancer Research Uk, Cambridge
  • @DrMarkDunning
  • mark.dunning@cruk.cam.ac.uk
• Tom
  • Medical Research Council, Clinical Sciences Centre, London
  • thomas.carroll@csc.mrc.ac.uk
• Nuno
  • Instituto de Medicina Molecular, Lisboa
  • nmorais@medicina.ulisboa.pt

Admin

About the Course

• We will tell about ‘best practice’ tools that we use in daily work as Bioinformaticians
• You will (probably) not come away being an expert
• We cannot teach you everything about NGS data
  • plus, it is a fast-moving field
• RNA and ChIP only
  • much of the initial processing is the same for other assays
• However, we hope that you will
  • Understand how your data are processed
  • Be able to explore your data - no programming required
  • Increase confidence with R and Bioconductor
  • Be able to explore new technologies, methods, tools as they come out

Further disclaimer

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

If you haven’t designed your experiment properly, then all the Bioinformatics we teach you won’t help: Consult with your local statistician - preferably not the day before your grant is due!!!!

• We have some materials you can look at

Day 1

• Recap of R
• Introduce the Bioconductor project
• Hands-on experience with NGS data
  • IGV
  • FastQC

Day 2

• Data structures for NGS analysis in R
• Statistical theory behind RNA-seq analysis
• RNA-seq intro
• Aligning and counting for RNA-seq

Day 3

• Differential expression analysis for RNA-seq
• Annotating RNA-seq results
• Downstream analysis of RNA-seq

Day 4

• Downstream analysis of RNA-seq
• Intro to ChIP-seq
• QA and analysis of ChIP data
• Wrap-up

Cast your minds back a few years..

Plenty of success stories with microarrays

What did we learn from arrays?

• Experimental Design; despite this fancy new technolgy, if we don’t design the experiments properly we won’t get meaningful conclusions
• Quality assessment; Yes, NGS experiments can still go wrong!
• Normalisation; NGS data come with their own set of biases and error that need to be accounted for
• Stats; testing for RNA-seq is built-upon the knowledge from microarrays
• Plenty of tools and workflows were established.
• Don’t forget about arrays; the data are all out there somewhere waiting to be discovered and explored

Reproducibility is key

Two Biostatiscians (later termed ‘Forensic Bioinformaticians’) from M.D. Anderson used R extensively during their re-analysis and investigation of a Clinical Prognostication paper from Duke. The subsequent scandal put Reproducible Research on the map.

Keith Baggerly’s talk from Cambridge in 2010 is highy-recommended.

Advantages of R

The R programming language is now recognised beyond the academic community as an effect solution for data analysis and visualisation. Notable users of R include Facebook, google, Microsoft (who recently invested in a commerical provider of R), and the New York Times.

Key features

• Open-source
• Cross-platform
• Access to existing visualisation / statistical tools
• Flexibility
• Visualisation and interactivity
• Add-ons for many fields of research
• Facilitating Reproducible Research

Support for R

• Online forums such as Stack Overflow regularly feature R
• Blogs
• Local user groups
• Documentation via ? or help.start()

RStudio

• Rstudio is a free environment for R
• Convenient menus to access scripts, display plots
• Still need to use command-line to get things done
• Developed by some of the leading R programmers

R recap

R can do simple numerical calculations

2  + 2

## [1] 4

sqrt(25)

## [1] 5

Here, sqrt is a function and the number 25 was used as an argument to the function. Functions can have multiple arguments

Variables

We can save the result of a computation as a variable using the assignment operator <-

x <- sqrt(25)
x + 5

## [1] 10

y <- x +5
y

## [1] 10

Vectors

A vector can be used to combine multiple values. The resulting object is indexed and particular values can be queried using the [] operator

vec <- c(1,2,3,6)
vec[1]

## [1] 1

Vectors

Calculations can be performed on vectors

vec*2

## [1]  2  4  6 12

mean(vec)

## [1] 3

sum(vec)

## [1] 12

Data frames

These can be used to represent familiar tabular (row and column) data

df <- data.frame(A = c(1,2,3,6), B = c(7,8,10,12))
df

##   A  B
## 1 1  7
## 2 2  8
## 3 3 10
## 4 6 12

Data frames

Don’t need the same data type in each column

df <- data.frame(A = c(1,2,3,6), 
                 B = month.name[c(7,8,10,12)])
df

##   A        B
## 1 1     July
## 2 2   August
## 3 3  October
## 4 6 December

Data frames

We can subset data frames using the [], but can specify row and column indices

df[1,2]

## [1] July
## Levels: August December July October

df[2,1]

## [1] 2

Data frames

df[1,]

##   A    B
## 1 1 July

df[,2]

## [1] July     August   October  December
## Levels: August December July October

Or leave the row or column index blank to get all rows and columns respectively

Plotting

All your favourite types of plot can be created in R

Plotting

• Simple plots are supported in the base distribution of R (what you get automatically when you download R).
  • boxplot, hist, barplot,… all of which are extensions of the basic plot function
• Many different customisations are possible
  • colour, overlay points / text, legends, multi-panel figures
• We will show how some of these plots can be used to inform us about the quality of NGS data, and to visualise our results.
• References..
  • Introductory R course
  • Quick-R

The Bioconductor project

• Packages analyse all kinds of Genomic data (>800)
• Compulsory documentation (vignettes) for each package
• 6-month release cycle
• Course Materials
• Example data and workflows
• Common, re-usable framework and functionality
• Available Support
  • Often you will be able to interact with the package maintainers / developers and other power-users of the project software
• Annual conferences in U.S and Europe
  • The last European conference was in Cambridge

The Bioconductor project

Many of the packages are by well-respected authors and get lots of citations.

Downloading a package

Each package has its own landing page. e.g. http://bioconductor.org/packages/release/bioc/html/beadarray.html. Here you’ll find;

• Installation script (will install all dependancies)
• Vignettes and manuals
• Details of package maintainer
• After downloading, you can load using the library function. e.g. library(beadarray)
• Only need to download once for each version of R
• CRAN packages installed by install.packages
• What packages to install?
  • METACRAN can help

Introducing the practical

• Refresh your memory of R skills
  • reading data
  • subsetting data
  • plotting