author: April Wright date: 08.10.2018 autosize: true

Macroevolution and Phylogeny

Phylogenetic trees are often used to study patterns of diversification, trait evolution, and lineage dispersal

Macroevolution and Phylogeny

Phylogenetic trees are often used to study patterns of diversification, trait evolution, and lineage dispersal

But the reverse is also true

The Fossilized Birth-Death is a Macroevolutionary Model

It describes the distribution of lineages through time (Stadler 2010)

The Fossilized Birth-Death is a Macroevolutionary Model

  • Several key parameters

The Fossilized Birth-Death is a Macroevolutionary Model

  • Several key parameters

The Fossilized Birth-Death is a Macroevolutionary Model

  • Several key parameters

The Fossilized Birth-Death is a Macroevolutionary Model

  • Several key parameters

The Fossilized Birth-Death is a Macroevolutionary Model

  • Several key parameters

The Fossilized Birth-Death is a Phylogenetic Model

  • All of these parameters relate to the tree

Tree Models: Hands-On with FossilSim

vignettes/03-FossilSim

Look at section Birth and Death first.

Sampled Ancestors

  • Under the fossilized birth-death model, it is possible to sample lineages that also have descendants on the tree

When do we observe more sampled ancestors? Look at some of your tree plots to find out.

What do you think would happen to speciation rates if sampled ancestors are disallowed?

Tree Models: Hands-On with FossilSim

vignettes/03-FossilSim

Now look at the sampling section

Tree Models: Hands-On with FossilSim

plot of chunk unnamed-chunk-1

This tree implies some values: that lineages have a time since origination, a time at extinction, and a time of sampling. How do we get them?

Yesterday, we unveiled a conflation

Likelihood and Bayesian phylogenetic trees have a rate-based branch length

distance = rate x time

What if we’re interested in time?

Stigall and Wright 2017

Why are we interested in time?

  • Time might be inherently interesting.

Why are we interested in time?

  • Time might be inherently interesting
  • Time might tell us something interesting about natural history Alan Turner, 2016

Turner, et al. 2016

Why are we interested in time?

  • Time might be inherently interesting
  • Time might tell us something interesting about natural history
  • Time might be required to test hypotheses

Alfaro et al, 2018 Alfaro et al. 2018

Information in a phylogenetic matrix

  • Does this tell you when (in absolute time) Agriarctos spp diverged from Ailurarctos lufengensis?
$Agriarctos_spp
 [1] "?" "0" "?" "?" "?" "?" "?" "?" "?" "?" "?" "?" "?" "?" "?" "?" "0"
[18] "0" "0" "1" "1" "1" "1" "0" "0" "1" "?" "1" "1" "?" "0" "1" "1" "1"
[35] "1" "0" "1" "1" "0" "?" "?" "0" "1" "1" "1" "0" "?" "?" "?" "?" "?"
[52] "?" "?" "?" "?" "?" "?" "?" "?" "?" "?" "?"

$Ailurarctos_lufengensis
 [1] "?" "0" "?" "?" "?" "?" "?" "?" "?" "?" "?" "?" "?" "?" "?" "?" "?"
[18] "0" "0" "1" "1" "1" "1" "0" "1" "1" "?" "1" "1" "?" "0" "?" "?" "?"
[35] "?" "0" "1" "1" "1" "?" "0" "0" "1" "1" "1" "0" "1" "0" "1" "1" "0"
[52] "1" "1" "?" "?" "?" "?" "?" "?" "?" "?" "?"

$Ailuropoda_melanoleuca
 [1] "1" "0" "1" "1" "1" "1" "0" "1" "1" "0" "1" "0" "0" "1" "0" "0" "0"
[18] "0" "0" "1" "1" "1" "1" "0" "1" "0" "1" "1" "1" "0" "0" "1" "0" "1"
[35] "0" "0" "1" "1" "0" "0" "0" "0" "1" "1" "1" "0" "1" "0" "0" "1" "0"
[52] "1" "1" "0" "0" "0" "1" "0" "0" "0" "1" "0"

$Ballusia_elmensis
 [1] "?" "0" "?" "?" "?" "?" "0" "1" "?" "0" "0" "0" "0" "0" "0" "0" "0"
[18] "0" "?" "0" "1" "0" "0" "0" "0" "0" "?" "1" "0" "0" "0" "0" "1" "0"
[35] "0" "?" "?" "0" "?" "?" "?" "0" "0" "0" "0" "0" "1" "0" "1" "1" "1"
[52] "1" "0" "?" "?" "?" "?" "?" "?" "?" "?" "?"

No.

Information in a phylogenetic matrix

  • Where do we get our information on absolute ages?
$Agriarctos_spp
 [1] "?" "0" "?" "?" "?" "?" "?" "?" "?" "?" "?" "?" "?" "?" "?" "?" "0"
[18] "0" "0" "1" "1" "1" "1" "0" "0" "1" "?" "1" "1" "?" "0" "1" "1" "1"
[35] "1" "0" "1" "1" "0" "?" "?" "0" "1" "1" "1" "0" "?" "?" "?" "?" "?"
[52] "?" "?" "?" "?" "?" "?" "?" "?" "?" "?" "?"

$Ailurarctos_lufengensis
 [1] "?" "0" "?" "?" "?" "?" "?" "?" "?" "?" "?" "?" "?" "?" "?" "?" "?"
[18] "0" "0" "1" "1" "1" "1" "0" "1" "1" "?" "1" "1" "?" "0" "?" "?" "?"
[35] "?" "0" "1" "1" "1" "?" "0" "0" "1" "1" "1" "0" "1" "0" "1" "1" "0"
[52] "1" "1" "?" "?" "?" "?" "?" "?" "?" "?" "?"

$Ailuropoda_melanoleuca
 [1] "1" "0" "1" "1" "1" "1" "0" "1" "1" "0" "1" "0" "0" "1" "0" "0" "0"
[18] "0" "0" "1" "1" "1" "1" "0" "1" "0" "1" "1" "1" "0" "0" "1" "0" "1"
[35] "0" "0" "1" "1" "0" "0" "0" "0" "1" "1" "1" "0" "1" "0" "0" "1" "0"
[52] "1" "1" "0" "0" "0" "1" "0" "0" "0" "1" "0"

$Ballusia_elmensis
 [1] "?" "0" "?" "?" "?" "?" "0" "1" "?" "0" "0" "0" "0" "0" "0" "0" "0"
[18] "0" "?" "0" "1" "0" "0" "0" "0" "0" "?" "1" "0" "0" "0" "0" "1" "0"
[35] "0" "?" "?" "0" "?" "?" "?" "0" "0" "0" "0" "0" "1" "0" "1" "1" "1"
[52] "1" "0" "?" "?" "?" "?" "?" "?" "?" "?" "?"

Information in a phylogenetic matrix

So, where do we get the information on absolute ages?

Information in a phylogenetic matrix

So, where do we get the information on absolute ages?

Divergence time estimation is the process by which we model the distribution of branch rates, speciation (& extinction events) and fossil presence to obtain node ages on a tree

Divergence time estimation

Image via Rachel Warnock

Divergence time estimation

  • Strict clock
library(phangorn)
library(ggtree)
library(ggplot2)
aln_path <- 'data/bears_fasta.fa'
sample_df <- generate_sliding(bears, start_char = 1, stop_char = 2, steps = 10)
output_vector <- generate_tree_vis(sample_df = sample_df, alignment =                                         aln_path, tree = tree, phy_mat = bears)
output_vector[[1]]

plot of chunk unnamed-chunk-4

If we know the rate of sequence divergence is 1% per million years, we can date the tree by extrapolating from the amount of divergence.

Branch Rates

  • Strict clock

    • Reality: the rate at which mutations arise varies based on generation time, lifestyle, DNA repair. Evolutionary forces cause differences in if a mutation is fixed, or removed from the population.

Branch Rates

  • Strict clock

Reality: the rate at which mutations arise varies based on generation time, lifestyle, DNA repair. Evolutionary forces cause differences in if a mutation is fixed, or removed from the population.

Branch Rates

  • Relaxed clock

Branch Rates

Many ways to describe how branch rates are distributed

  • Global clock (Zuckerkandl & Pauling, 1962)
  • Uncorrelated/independent rates models (Drummond et al. 2006; Rannala & Yang 2007; Lepage et al. 2007)
  • Log-normally distributed autocorrelated rates (Thorne, Kishino & Painter 1998; Kishino, Thorne & Bruno 2001; Thorne & Kishino 2002)
  • Local clocks (Hasegawa, Kishino & Yano 1989; Kishino & Hasegawa 1990; Yoder & Yang 2000; Yang & Yoder 2003, Drummond and Suchard 2010)
  • Mixture models on branch rates (Heath, Holder, Huelsenbeck 2012)
  • Punctuated rate change model (Huelsenbeck, Larget and Swoford 2000)

Slide from Rachel Warnock

Branch Rates

tree <- sim.fbd.taxa(n=10, numbsim=1, lambda=3, mu=2, psi=2)
plot(tree[[1]])

plot of chunk unnamed-chunk-6

  • Fundamental: In order to know something about the time since divergence, we need to know something about the rate at which differences accumulate

Divergence Time Estimation

So we’re done! We got it, right?

Divergence time estimation

Image via Rachel Warnock

Including Fossils

  • Including fossils substantially improves estimations of divergence times
    • Fossils allow us to establish minimum ages for divergences

Including Fossils: Calibrations

  • Fossils have typically been incorporated into analyses as what are called node calibrations

  • Use distributions to describe the relationship between the fossil sampling event and the parent node (which generated the fossil)

Including Fossils: Calibrations

Heath Sys Bio 2012

Heath 2012

Including Fossils: Calibrations

Heath Sys Bio 2012

  • Phylogenetic position not estimated from data
  • We can only use the age of the oldest fossil

Including Fossils: Calibrations

Warnock et al. 2012 Biol Letters

Warnock et al. 2012

Including Fossils: Calibrations

Warnock et al. 2015 Biol Letters

Warnock et al. 2015

Including Fossils: Fossilized Birth-Death

Include fossils as tips in our analysis

Heath, 2014

Including Fossils: Fossilized Birth-Death

Including Fossils: Fossilized Birth-Death

  • Fossils can be placed simply as we did yesterday in the discrete morphology tutorial
  • Can also be placed via reversible jump MCMC to be sampled ancestors

Including Fossils: Fossilized Birth-Death

With a partner: describe this graphical model in a couple sentences. What does each piece do?

image from Rachel Warnock