Research and Teaching With RevBayes

Overview

• Introduction to Coding to Learn
• Introduction to graphical modeling
• How graphical modeling makes phylogenetic estimation more flexible
• Using RevBayes in the classroom

Learning to Code

We’re all familiar with the idea of learning to code

Learning to Code

We’re all familiar with the idea of learning to code

• Maybe students in your degree program take an upper-division course where codin is the aim

Learning to Code

We’re all familiar with the idea of learning to code

• Maybe students in your degree program take an upper-division course where codin is the aim
• This is the “learn to code” framework. Being able to produce code is the end goal

Learning to Code

There are some unsatisfying things about this model

Learning to Code

There are some unsatisfying things about this model

• If we confine coding to niche classes, we’re likely to miss much of the population
• Large proportion of students who seek out computational training have seen computation in high school

Learning to Code

There are some unsatisfying things about this model

• If we confine coding to niche classes, we’re likely to miss much of the population
• Large proportion of students who seek out computational training have seen computation in high school

== Confining learning to people who are already convinced of the utility

Learning to Code

There are some unsatisfying things about this model

• If we confine coding to niche classes, we’re likely to miss much of the population
• Large proportion of students who seek out computational training have seen computation in high school

== Students get coding either late in undergrad, or after they’ve been mucking around unproductively with data

Code to Learn

Coding doesn’t have to be the main focus of what you’re doing.

Code to Learn

Coding doesn’t have to be the main focus of what you’re doing

• Code to learn: work with data computationally to develop biological insight

Code to Learn

Coding doesn’t have to be the main focus of what you’re doing

• Code to learn: work with data computationally to develop biological insight

• Example: Rather than talking about Bergmann’s Rule, have students download occurrence data from GBIF and key it to a trait database to observe body size along longitudinal gradients

Code to Learn

Citations for many of the things in the previous couple slides can be found in this manuscript

Code to Learn

Coding doesn’t have to be the main focus of what you’re doing

• Systematics is a field ripe with opportunities
• To understand systematics, you need to know some

• Biology

• Statistics

• Math

• Geology & Earth history

Code to Learn

Coding doesn’t have to be the main focus of what you’re doing

• Systematics is a field ripe with opportunities
• In performing systematics research, we learn about

• Biology

• Statistics

• Math

• Geology & Earth history

RevBayes

RevBayes is a Bayesian modeling software underlain by a graphical model framework

RevBayes

RevBayes is a Bayesian modeling software underlain by a graphical model framework

• RevBayes implements an R-like statistical modeling language called Rev

RevBayes

RevBayes is a Bayesian modeling software underlain by a graphical model framework

• RevBayes implements an R-like statistical modeling language called Rev
• The design of the language is meant to give users control over the details of their analysis

The Graphical Modeling Framework

• What is a model?

The Graphical Modeling Framework

• What is a model?
  • A model uses mathematics to describe a process, or the behavior of a system
  • Important facets of the generating process of the data are written down as parameters

The Graphical Modeling Framework

• What is a model?
  • A model uses mathematics to describe a process, or the behavior of a system
  • Important facets of the generating process of the data are written down as parameters
• A graphical model represents the relationships and dependencies between model paramters

A simple case

• An archer: Let’s say the archer is a pretty good archer

A simple case

good_shots <- rnorm(n= 100, mean= 0, sd = 1)
hist(good_shots, main="Archer arrow shots",  xlab = "Inches from bullseye")

A simple case

• An archer: Let’s say the archer is a pretty bad archer

A simple case

bad_shots <- rnorm(n= 100, mean= 0, sd = 5)
hist(bad_shots, main="Archer arrow shots",  xlab = "Inches from bullseye")

A not so simple case

• It’s easy to describe a model where we are manipulating one parameter

• But we rarely have such a simple case when we’re analysing biological data

• We’re describing the mechanisms that generated our observed data

• In the case of a tree, this is the process of molecular or morphological evolution

A not so simple case

• This is where graphical models come in

A not so simple case

• This is where graphical models come in
• A graphical model represents the joint probability distribution as a graph

A not so simple case

• This is where graphical models come in
• A graphical model represents the joint probability distribution as a graph
• Nodes are variables. Edges represent dependencies between nodes.

A not so simple case

• This is where graphical models come in
• A graphical model represents the joint probability distribution as a graph
• Nodes are variables. Edges represent dependencies between nodes.

A not so simple case

• This is where graphical models come in
• A graphical model represents the joint probability distribution as a graph
• Nodes are variables. Edges represent dependencies between nodes.

A not so simple case

• This is where graphical models come in
• A graphical model represents the joint probability distribution as a graph
• Nodes are variables. Edges represent dependencies between nodes.

Pr( data | model )

A not so simple case

Phylogenetical Complexity

• This is where graphical models come in
• A graphical model represents the joint probability distribution as a graph
• Nodes are variables. Edges represent dependencies between nodes.

Phylogenetical complexity

Phylogenetical complexity

Phylogenetical complexity

Phylogenetical complexity

Phylogenetical complexity

Who cares about graphical modeling?

• In previous iterations of software, you might have to email a developer and ask them to implement something

• RevBayes’ graphical model framework emphasizes flexibility

Who cares about graphical modeling?

Over the next two hours, we’re going to make the case that this framework enables flexible, code-to-learn classrooms