bpp

Introduction

The aim of this project is to implement a versatile high-performance version of the BPP software. It should have the following properties:

• open-source code with an appropriate open-source license.
• 64-bit multi-threaded design that handles very large datasets.
• easy to use and well-documented.
• SIMD implementations of time-consuming parts.
• Linux, Mac and Microsoft Windows compatibility.

BPP is a Bayesian Markov chain Monte Carlo (MCMC) program for analyzing sequence alignments from multiple loci and multiple closely-related species under the multispecies coalescent (MSC) model. BPP takes into account episodic and continuous gene flow through its implementation of the multispecies coalescent with introgression (MSC-I) model for episodic events, and the multispecies coalescent with migration (MSC-M) model for continuous interactions.

BPP currently implements four methods:

• Estimation of the parameters of species divergence times, population sizes, and potentially introgression probabilities or migration rates under the MSC (Rannala and Yang, 2003), MSC-I (Flouri et al, 2020) or MSC-M (Flouri et al, 2023) models when the species phylogeny is given.

• Inference of the species tree when the assignments are given by the user (Rannala and Yang, 2017)

• Species delimitation using a user-specified guide tree (Yang and Rannala, 2010; Rannala and Yang, 2013)

• Joint species delimitation and species tree estimation (Yang and Rannala 2014)

Further functionalities:

• Accommodates variable mutation rates among loci (Burgess and Yang,2008) and heredity multipliers (Hey and Nielsen, 2004).

• Handles diploid data, where phasing is done analytically as described by Gronau et al, 2011.

• Implements relaxed-clock models to estimate species divergence times and ancestral population sizes using genomic data sampled from contemporary species when the strict clock assumption is violated (Flouri et al, 2022).

• Nucleotide substitution models JC69, K80, F81, HKY, T92, TN93, F84 and GTR.

• 19 amino acid substition models (Dayhoff, LG, DCMUT, JTT, etc).

• Site rate variation (+Gamma).

• Partitioned analyses.

• Easy to use tool for generating MSC-I networks by specifying a list of introgressions/hybridizations events on a species tree.

For more information on usage please see the BPP manual, the BPP Documentation Project and/or the wiki documentation.

Download and install

Binary distribution Starting with version 4.1.3, binary distribution files containing pre-compiled binaries will be available as part of each release. The included executables are statically compiled whenever possible such that no library dependencies are necessary.

Binary distributions are provided for x86-64 systems running GNU/Linux, macOS (version 10.13 or higher) and Windows (64-bit, version 7 or higher).

Download the appropriate executable for your system using the following commands if you are using a Linux x86_64 system:

wget https://github.com/bpp/bpp/releases/download/v4.8.7/bpp-4.8.7-linux-x86_64.tar.gz
tar zxvf bpp-4.8.7-linux-x86_64.tar.gz

Or these commands if you are using a Linux ppc64le system:

wget https://github.com/bpp/bpp/releases/download/v4.8.7/bpp-4.8.7-linux-ppc64le.tar.gz
tar zxvf bpp-4.8.7-linux-ppc64le.tar.gz

Or these commands if you are using a Linux aarch64 (arm64) system:

wget https://github.com/bpp/bpp/releases/download/v4.8.7/bpp-4.8.7-linux-aarch64.tar.gz
tar zxvf bpp-4.8.7-linux-aarch64.tar.gz

Or these commands if you are using a Mac with an Apple Silicon CPU (e.g. M1/M2/M3):

wget https://github.com/bpp/bpp/releases/download/v4.8.7/bpp-4.8.7-macos-aarch64.tar.gz
tar zxvf bpp-4.8.7-macos-aarch64.tar.gz

Or these commands if you are using a Mac with an Intel CPU:

wget https://github.com/bpp/bpp/releases/download/v4.8.4/bpp-4.8.4-macos-x86_64.tar.gz
tar zxvf bpp-4.8.4-macos-x86_64.tar.gz

Or if you are using Windows, download and extract (unzip) the contents of this file:

https://github.com/bpp/bpp/releases/download/v4.8.7/bpp-4.8.7-win-x86_64.zip

Linux and Mac: You will now have the binary distribution in a folder called bpp-4.8.7-linux-x86_64 or bpp-4.8.7-macos-x86_64. The binary file is located in the bin subfolder, i.e. bin/bpp. We recommend making a copy or a symbolic link to the binary in a folder included in your $PATH.

Windows: You will now have the binary distribution in a folder called bpp-4.8.7-win-x86_64. The bpp executable is called bpp.exe.

Compiling from source You can either download the source distribution for a particular version or clone the repository.

Source distribution To download the source distribution from a release and build the executable and documentation, use the following commands:

wget https://github.com/bpp/bpp/archive/v4.8.7.tar.gz
tar zxvf v4.8.7.tar.gz
cd bpp-4.8.7/src
make

Cloning the repo Instead of downloading the source distribution as a compressed archive, you could clone the repo and build it as shown below.

git clone https://github.com/bpp/bpp.git
cd bpp/src
make

Compiling BPP requires that your system has GCC version 4.7 or newer, as AVX and AVX-2 optimized functions are compiled even if your processor does not support them. This is fine, as BPP will automatically select the right instruction set that your processor supports at run-time. This means, you can compile on one system, and run BPP on any other compatible system.

However, if your compiler is older than 4.7, you will get errors such as:

cc1: error: unrecognized command line option "-mavx2"

or

cc1: error: unrecognized command line option "-mavx"

If your compiler is GCC 4.6.x then you can compile BPP using:

make clean
make -e DISABLE_AVX2=1

In case your compiler is older than GCC 4.6 then compile using:

make -e DISABLE_AVX2=1 DISABLE_AVX=1

You can check your compiler version with:

gcc --version

Running BPP

After creating the control file, one can run BPP as follows:

bpp --cfile [CONTROL-FILE]

If you would like to resume a checkpoint file, please run:

bpp --resume [CHECKPOINT-FILE]

If you would like to run the simulator (previously MCcoal), please run:

bpp --simulate [CONTROL-FILE]

If you would like to run the MSci network generator, please run:

bpp --msci-create [DEFS-FILE]

If you would like to only summarize the results of an analysis, please run:

bpp --summary [CONTROL-FILE]

For an example of a DEFS-FILE see the MSci generator notes

More documentation regarding control files, will be available soon on the wiki.

Documentation

The most up-to-date documentation for BPP is now available through the BPP Documentation Project. The original manual bpp4DOC.pdf is found in the current BPP distribution.

A tutorial on BPP was recently published as a book chapter: A Tutorial on the Use of BPP for Species Tree Estimation and Species Delimitation

For information on the MSci model please read:

• Flouri T., Jiao X., Rannala B., Yang Z. (2020) A Bayesian Implementation of the Multispecies Coalescent Model with Introgression for Phylogenomic Analysis. Molecular Biology and Evolution, 37(4):1211-1223.

• Jiao X., Flouri T., Rannala B., Yang Z. (2020) The Impact of Cross-Species Gene Flow on Species Tree Estimation. Systematic Biology (in press). doi:10.1093/sysbio/syaa001

Citing BPP

Please cite the following publication if you use BPP:

Flouri T., Jiao X., Rannala B., Yang Z. (2018) Species Tree Inference with BPP using Genomic Sequences and the Multispecies Coalescent. Molecular Biology and Evolution, 35(10):2585-2593. doi:10.1093/molbev/msy147

Please note, citing the corresponding of the four underlying methods, may also be appropriate.

When using the MSC-I model please cite:

Flouri T., Jiao X., Rannala B., Yang Z. (2020) A Bayesian Implementation of the Multispecies Coalescent Model with Introgression for Phylogenomic Analysis. Molecular Biology and Evolution, 37(4):1211-1223. doi:10.1093/molbev/msz296

When using the MSC-M model please cite:

Flouri T., Jiao X., Huang J., Rannala B., Yang Z. (2023) Efficient Bayesian inference under the multispecies coalescent with migration. Proceedings of the National Academy of Sciences, 120(44):e2310708120, 2023. doi:10.1073/pnas.2310708120

When using relaxed clocks please cite:

Flouri T., Huang J., Jiao X., Kapli P., Rannala B., Yang Z. (2022) Bayesian phylogenetic inference using relaxed-clocks and the multispecies coalescent. Molecular Biology and Evolution, 39(8), 2022. doi:10.1093/molbev/msac161

License and third party licenses

The code is curre