Lineage through time(LTT plot)

[fateme moein]

Hello to All,

Sorry if this question is not exactly related to BEAST software, I have done ltt plot via ape and phytool, but I would like to change the x-axis to geo time classification.

I was wondering if anyone have an idea which R-script to use to add geo time.

Thanks

Fatemeh

[Santiago Sánchez]

Hi Fatemeh,

There is a package called "geoscale" which has a couple of plotting functions that add a geological scale. Personally, I find them of little use, however the package includes a data set of geological ages "data(timescales)" . You can use these to add a scale to your own plot. For instance, you could do something like:

library(geoscale)

library(phytools)

set.seed(1)

# generate LTT plot and tree data

tr <- pbtree(n=100) # simulate tree

tr$edge.length * 10 -> tr$edge.length # increase edge lengths

bt <- ltt.plot.coords(tr) # LTT plot data

bt.log <- cbind(time=bt[,"time"],N=log(bt[,"N"])) # convert to log

# prepare time scale data

data(timescales)

timescales$ICS2015 -> ts # select data set

ts[ ts$Type == "Epoch",] -> ts.epoch # get only data for epoch

which.min(abs(ts.epoch$End - abs(min(bt[,1])))) -> min.date # get the epoch with is closer to the root of the tree

# generate your vector data

start <- ts.epoch$Start[1:min.date]

end <- ts.epoch$End[1:min.date]

midpoint <- ts.epoch$Midpoint[1:min.date]

abbrev <- ts.epoch$Abbrev[1:min.date]

# remove the holocene (useless)

end <- end[-1]

start <- start[-1]

midpoint <- midpoint[-1]

abbrev <- abbrev[-1]

# make vectors negative

end <- -end

start <- -start

midpoint <- -midpoint

# generate a binary (1,0) vector to make a two-color vector

rep(c(1,0),length(start))[1:length(start)] -> col.fac

# plot with no data

plot(bt.log, xlim=c(min(start),0), ylim=c(0,max(bt.log[,"N"]+1)), type="n", ylab="lineages", xlab="time")

# add rectangles

rect(start, rep(0,length(start)), end, rep(log(max(bt[,"N"])),length(start)), col=c("grey","white")[factor(col.fac)], border=F)

# add LTT plot

lines(x=bt[,"time"], y=log(bt[,"N"]), type="b", col="red")

# add labels

text(x=midpoint, y=max(log(bt[,"N"]))+0.5, labels=abbrev, cex=0.5)

This code should generate a plot like the one attached.

Hope this helps.

Cheers,

Santiago

--
You received this message because you are subscribed to the Google Groups "beast-users" group.
To unsubscribe from this group and stop receiving emails from it, send an email to beast-users...@googlegroups.com.
To post to this group, send email to beast...@googlegroups.com.
Visit this group at https://groups.google.com/group/beast-users.
For more options, visit https://groups.google.com/d/optout.

--

==========================

Santiago Sanchez-Ramirez, PhD

Postdoctoral Associate

Ecology and Evolutionary Biology

University of Toronto

==========================

[Jinmin Chen]

Dear All,

For lineage-through-time (LTT) plot, how can we visualize the accumulation of lineages within different regions (see attached)?

Best wishes.

Chen

[Santiago Sánchez]

Hi Jinmin,

This is more a question for R users (e.g. r-sig-phylo) than for BEAST users, but here it goes..

First you need to do ancestral state reconstruction on your tree based on traits for your different regions. Then, you can extract which nodes were inferred with the highest probability for the different regions. Afterwards, simply retrieve the node ages of each of the different regions and then plot. This is rather straight forward to do in R, but will require some knowledge R coding and use of functions from the APE package.

Sorry for the short answer..

Good luck,

Santiago

--

You received this message because you are subscribed to the Google Groups "beast-users" group.
To unsubscribe from this group and stop receiving emails from it, send an email to beast-users...@googlegroups.com.
To post to this group, send email to beast...@googlegroups.com.
Visit this group at https://groups.google.com/group/beast-users.
For more options, visit https://groups.google.com/d/optout.