Intro to Using Galaxy

–
For Bioinformatics

Tom Doak
Carrie Ganote
National Center for Genome Analysis Support

September 17, 2013

 Summary

•  Who is NCGAS?
•  Galaxy – what is it?
•  Galaxy 101 – a guided tour
•  Short intro to transcriptome
assembly, as an example

National Center for Genome Analysis Support: http://ncgas.org

 Who is NCGAS?

The National Center for Genome

Analysis Support is based at IU in
Bloomington, but caters to a national
audience with support from the NSF.
We provide computational resources and
support for genomics, transcriptomics,
and meta projects.

National Center for Genome Analysis Support: http://ncgas.org

 Our Services
NCGAS provides support in the form of long- and short-term consultation
for genomics, proteomics, transcriptomics, and meta projects. We are
happy to answer questions about software, methods, and pipelines; basic
Linux use; experimental setup; and interpretation of results.

We administer bioinformatics software installation and upgrades on the

Mason cluster at IU, as well as provide access to Mason to users of
XSEDE’s national infrastructure. We provide support letters for NSF
proposals pledging our compute resources.

Last, but not least, we install and maintain the local Galaxy instances for
Indiana University: IU, NCGAS, and Rockhopper.

National Center for Genome Analysis Support: http://ncgas.org

 What is Galaxy?
Galaxy is a web-based framework for running command-line utilities from
a snazzy graphical user interface.
The Galaxy web server that we will be using today is hosted at Indiana
University on the XSEDE virtual machines. This is a different “instance”
than Galaxy Main, which is hosted at Penn State.

Our instance at IU Why choose us: Galaxy Main

•  IU only – less
Galaxy@IU Data busy!
Virtual Machine •  Large RAM Penn State
jobs possible Resources
Jobs •  Custom tools
Data
on request
Mason •  On-site support

Data Capacitor
National Center for Genome Analysis Support: http://ncgas.org
 Galaxy Anatomy and Physiology

Tool bar – History –

contains shows
the steps
available previously
steps to taken to
apply to manipulate
data input data
sets

Focus pane – shows options,

parameters, and output for current item.
National Center for Genome Analysis Support: http://ncgas.org
 Galaxy 101 – Quick Start

We will depart this slideshow for a short time as we go through the

basics of Galaxy using the Galaxy 101 tutorial. You can find a link
to it on the home page for galaxy.indiana.edu.

You can choose to follow along either on IU Galaxy or on Galaxy

Main – the tool layout is slightly different between the two
instances.

National Center for Genome Analysis Support: http://ncgas.org

 Today’s Menu Item

We will be assembling the DNA

Polymerase protein units from the H37Rv
strain of Mycobacterium tuberculosis, the
causative agent of TB, also known as the
consumption.
The raw reads originated from the Short
Read Archive on NCBI. The accession
number for the set is SRX212035.

This dataset consists of paired-end,

Cristobal Rojas, La miseria (1886) from Wikipedia. ~75bp RNA-Seq reads.

National Center for Genome Analysis Support: http://ncgas.org

 Let’s get some sequence data

Galaxy allows
users to
publish their
data to the
entire user
base.

Let’s start with “Shared Data”

at the top.
Then select Data Libraries
from the menu.
National Center for Genome Analysis Support: http://ncgas.org
 Let’s get some sequence data

Choose Workshop Data.

National Center for Genome Analysis Support: http://ncgas.org

 Let’s get some sequence data

Expand folder
Check both boxes

Import the Data sets to current history.

National Center for Genome Analysis Support: http://ncgas.org

 Let’s get some sequence data

Data set is imported – Click on Analyze Data to return.

National Center for Genome Analysis Support: http://ncgas.org

 Step 1: Assess the Quality of Inputs

We will first get an idea

of the quality of our
input data sets.

The FastQC tool will

produce graphical
output that makes it
easy to gauge the
characteristics of the
data – quality, patterns,
biases, gc content etc.

Choose either the left or right reads. Compare the results with your neighbor.

National Center for Genome Analysis Support: http://ncgas.org

 Step 1: Assess the Quality of Inputs
The input data usually
declines in quality as the
reads progress.

The quality score is

assigned by the
sequencing machine as
it reads each base. It is a
rough estimate of how
ambiguous the signal is.

National Center for Genome Analysis Support: http://ncgas.org

 Step 2: Trim Input Sequences
We’ve determined
that the input data
sets need some work
before they are used
in downstream
processes. We’ll use
the FASTQ quality
trimmer by sliding
window to trim reads
based on quality
score.

Run this tool for both input data sets.

National Center for Genome Analysis Support: http://ncgas.org

 Step 3: Rinse, Repeat

Now that the files

are trimmed, we will
re-assess their
quality. If necessary,
keep trimming away
until you are
satisfied with the
input files.

I renamed my trimmed files to help me keep them straight.

National Center for Genome Analysis Support: http://ncgas.org
 Step 3: Rinse, Repeat
Pictured are the left and right reads after trimming is complete.
These will do!

National Center for Genome Analysis Support: http://ncgas.org

 Step 4: Assembly

Next we will put the reads

together to create a complete
picture of the actively
transcribed genes of the
sample organism.

Trinity is a de novo assembler

that has been optimized for
use on Mason. We will use it
to assemble our reads.

National Center for Genome Analysis Support: http://ncgas.org

 It finished! We’re done, right?
An assembler solves a computer problem of putting
together a puzzle from tiny pieces. The output of the
assembler is a guess – but we don’t know how accurate it
is. We could look at:
•  Basic stats of the assembly – “Contigs”
•  Number of “Contigs” vs. Expected Number
•  N50 – a weighted average
•  Average Length
•  Max Length
•  Check contigs against known genes with Blast (large or
rare transcripts)
National Center for Genome Analysis Support: http://ncgas.org
 Step 5: Assessing Quality of Assembly
Important statistics for assembly quality:
Contig Length Distribution
Assemblies will typically produce a
number of complete contigs representing
whole transcripts, and a large number of
partial transcripts. This biases the
average contig length toward the low
end. The N50 is a measure weighted by
total sequence length in the assembly.

National Center for Genome Analysis Support: http://ncgas.org

 Step 5: Assessing Quality of Assembly
Getting these
stats in Galaxy:

Run assemblystats to
get a summary and
histograms of your
contig length
distribution.

National Center for Genome Analysis Support: http://ncgas.org

 Step 6: Check Against Database
For this last step, we’ll
check to see how well
our assembled
transcripts compare to
what we already know.

Use this step to give a

rough annotation of
genes, to make sure
that your transcripts are
from nuclear genes, or
to gauge how complete
your sequence is.

For sake of time, we’ll just Blast one gene. Filter out to get the smallest.

National Center for Genome Analysis Support: http://ncgas.org

 Step 6: Check Against Database

We will use Blastx to

search the NR
database for our
gene.

Use default search

settings for this test
set.

Make sure to choose Pairwise HTML output for readability.

National Center for Genome Analysis Support: http://ncgas.org
 Step 6: Check Against Database

We see the expected

genes as the top
hits!

We could limit the number of hits depending on output desired.

National Center for Genome Analysis Support: http://ncgas.org
 Step 7..?
RNA-Seq is a very versatile technology. You can use the data for:

•  Gene discovery based on transcripts

•  Genome evidence – introns, exons, junction
•  Gene expression patterns
•  SNP calling/other variants
•  Protein divergence between samples

We have gotten to the assembly step, but there is a lot to learn about
the data now that it is put together. A foundation in the use of Galaxy
coupled with Indiana University resources will enable you to reach
these goals.

National Center for Genome Analysis Support: http://ncgas.org

 Fin

Thanks for watching!

Questions and comments:
Email help@ncgas.org

National Center for Genome Analysis Support: http://ncgas.org