tseda

Tree Sequence Exploratory Data Analysis. Currently builds on code base in tsbrowse and retains some of the functionality. tseda adds functionality to view the spatial distribution of population data and connect that information to the genetic structure of the populations.

The input trees file must be tszipped and preprocessed into a tsbrowse file. Given a trees file data.trees the following commands perform compression and preprocessing of the file

tszip data.trees python -m tseda preprocess data.trees.tsz

to produce the output data.trees.tseda. To launch the web app run

python -m tseda serve data.trees.tseda

Screenshot of the prototype GNN viewer

Installation

tseda is currently in development. To install the latest dev version from github, run

python -m pip install git+https://github.com/tforest/tseda

Features (WIP)

• sample set editor to customize sample set definitions, names and colors
• summary statistics (Fst, genealogical nearest neighbours (GNN)) compared and averaged over sample sets
• genealogical nearest neighbour (GNN) analyses of samples
• plots of common summary statistics over sequence length (diversity, Tajimas_D, divergence, Fst)
• marginal tree plots
• pairwise coalescence plots

Tests

The package comes with test data files corresponding to an uncompressed trees file (test.trees), a tszipped trees file (test.trees.tsz), and a tseda file (test.trees.tseda) residing in tests/data. To try out the web app, load the file tests/data/test.trees.tseda as described above.