A repository for the lessons and tutorials for the Amplicons TOPIC channel of the BVCN
- Experience with R
- Experience with the command line
- Experience with Cyverse | Setting Up Qiime2 in Cyverse: Video tutorial
This BVCN topic will cover:
- the different approaches for analyzing amplicon data
- how to construct count tables, alignment files, and extract taxonomy from sequencing files
- statistical and data visualization techniques for analyzing amplicon data
Goals
- What are amplicons?
- What are the different pipelines for processing amplicon data?
- What are the available methods for producing count tables?
Access the presentation here and check out some of the links from presentation in this table.
This lesson is an informal discussion among instructors about topics such as
- What are your preferred workflows for analyzing amplicons?
- What are your opinions on certain caveats associated with amplicon analyses?
Watch the entire session here
Or just view the discussion on particular questions:
- How important is it to use ASVs over OTUs?
- Callahan et al. 2017- values of AVS over OTUs
- Summary of that paper’s main points at Happy Belly Bioinformatics here
- Is it still acceptable to use a 454 dataset?
- Are -omics approaches replacing amplicon approaches?
- Should we delete singletons?
- Edgar 2017: Evidence that 97% clustering falsely inflates singletons
- Have there been any recent advances for getting enough DNA from low biomass samples?
- Some kit options for 50ng input and 1ng input
- It's important to account for contamination when working with low biomass! A blog post from Rose Kantor's website.
- What kind of server do you use for running your amplicon pipeline?
- What does your typical amplicon workflow look like?
- Is there a difference between calling DADA2 in Qiime2 vs calling it on its own?
- Which interface do you use for running your code?
- How do you manage and store your data?
- How do you treat your count tables to account for the way that high throughput sequencing ‘collects’ data (ie. as relative abundances rather than absolute counts)?
These are a set of tutorials that replicate the amplicon analysis from Happy Belly using different pipeline/ software combinations. The tutorial from Happy Belly implements DADA2 in R and uses a naive Bayes classifer for taxonomy assignment.
This tutorial implements DADA2 in QIIME2 uses a naive Bayes classifer for taxonomy assignment.
- Watch the lesson.
- Follow the tutorial.
- Dataset and the notebook are available in this repo.
This tutorial implements DADA2 in R and uses DECIPHER for taxonomy assignment.
- Watch the lesson.
- Follow the tutorial.
- Dataset and DECIPHER training set are available in this repo.
Goals
- Importing files into R
- Removing contamination
- Checking sequencing depth with rarefaction curves
- Removing singletons
- Normalization
- Exporting formatted files from R
Links
- Watch the lesson
- Follow the tutorial
- Input data and lesson material are available in this repo
R crossover tutorial, see topic-R lesson 8 for more. Using the output from qiime2 analysis in Lesson 4
Goals
- Import ASV table into phyloseq
- Explore functionality of phyloseq: making tree, re-rooting tree, bar plot of taxa
- Hellinger transformations
- Ordinations with phyloseq:
- PCoA with Bray-Curtis distance matrices
- PCoA with weighted UniFrac
- NMDS with Bray-Curtis distance matrices
Links
- Watch the lesson
- Follow the tutorial
- Input data and lesson material are available in this repo
Primary tools/programs used:
R crossover tutorial, see topic-R lesson 8 for more.
Goals
- Import and manipulate ASV tables using tidyverse
- Log-ratio transformations
- Ordinations with vegan:
- Screeplots
- PCA plots
- PCoA with Jaccard and Euclidean distance matrices
- NMDS with Jaccard and Euclidean distance matrices
Links
- Watch the lesson
- Follow the tutorial
- Input data and lesson material are available in this repo
Primary tools/programs used:
Continuation of lesson 5b