Skip to main content

Amplicon Analysis Workshop

Monday 23rd September – Friday 27 September

CSIRO Hobart, Castray Esplanade, Battery Point, Tasmania

Overview

The use of amplicons generated by high throughput sequencing methods is now common place, and many avenues exist to learn specific workflows using explicit platforms. There are, however, fewer opportunities to learn the concepts behind amplicon sequence analysis in a platform agnostic way. We are offering an introductory course to amplicon sequence analysis. The course is aimed primarily at newcomers to sequence (amplicon) analysis and will cover the basics from experimental design to basic multivariate analysis of ecological data.

The course will comprise short lectures and hands on exercises. Course notes will be made available. There are no pre-requisites. Students will be expected to install a course virtual machine (supplied prior) to their computers before the beginning of the course.

Instructors:

Contact Andrew Bissett for registration and more information.

Course Outline:

Monday Compute set up/introductions
Introduction to jupyter notebook
NGS/multiplexed Amplicon sequencing
Experimental Design
What does sequence data look like and how do I check it out?
•  file formats
Review/questions
Tuesday Initial sequence processing
•  QC and paired end merging
Error correction / denoising
•  When, why, how
Wednesday Clustering
•  When, why, how
Classification
Dealing with large data tables
Python
•  Summary statistics and plots
•  Review and Questions
Thursday Data curation – trying our best to avoid catastrophic mistakes
Alpha diversity
•  Coverage
•  Diversity indices
•  Rarefaction
•  Linear models – fitting, diagnostics, predictions, multiple comparisons
Ordination
•  Choice of approach (PCA/CA/PCoA/NMDS)
•  Data standardisation (binary/counts/relabund/Hellinger/CoDa)
•  Interpreting output (loadings, associations)
•  Plotting
Friday Constrained ordination
•  Choice of approach (RDA/CCA/CAP)
•  Data standardisation (scaling/transformation/categorical)
•  Choosing predictors (vif/envfit/ordistep)
•  Interpreting output
• Plotting
Variation partitioning
•  Estimates and hypothesis tests
•  Spatial predictors
Experimental frameworks
•  PerMANOVA
•  Dispersion
•  Plotting

For further information or to register, please contact Andrew Bisset.