LAMARC - Likelihood Analysis with Metropolis Algorithm using Random Coalescence

LAMARC is a program which estimates population-genetic parameters such as population size, population growth rate, recombination rate, and migration rates. It approximates a summation over all possible genealogies that could explain the observed sample, which may be sequence, SNP, microsatellite, or electrophoretic data. LAMARC and its sister program Migrate are successor programs to the older programs Coalesce, Fluctuate, and Recombine, which are no longer being supported. The programs are memory-intensive but can run effectively on workstations; we support a variety of operating systems.

The LAMARC package is not in any immediate sense derived from the work of Jean Baptiste Pierre Antoine de Monet, Chevalier de Lamarck (1744-1829). The similarity of names is mostly accidental. But all evolutionary biologists do owe him a debt. Lamarck is an unfairly maligned figure. In addition to being one of the greatest figures of invertebrate biology, he was one of the founders (with Buffon) of the theory of evolution, and the first to propose a mechanism for evolution. You may want to read more about his life and work.

News and Updates

July 14th, 2008	Lamarc 2.1.3, This release fixes a bug introduced into the code used to guess haplotype resolution. Analyses using this feature in LAMARC versions 2.1.2 and 2.1.2b should be re-run. Additional improvements include: Tracer output now renders step counts as integers instead of reals. Mapping output now writable to its own file. Upgraded default wxWidgets installation to 2.8.8.
October 9th, 2007	Lamarc 2.1.2b, minor fixes to bugs affecting user experience only. You do not need to re-run any analyses performed with Lamarc 2.1.2.
October 4th, 2007	Lamarc 2.1.2, the program. LAMARC can jointly estimate population sizes, growth rates, migration rates, and recombination rate of single or multiple populations, using DNA, SNP, microsatellite and/or electrophoretic data in any combination. New for version 2.1.2: Greatly increased the length of sequence possible when estimating recombination: now can handle ~0.2 cM with relative ease, and ~0.5 cM with patience. (Previously, the limit was around ~0.01 cM.) Improved Tracer support, and other smallish improvements.
June 2007	Migrate 2.3. Migrate allows to estimate migrate rates, and population sizes from a wide variety of migration matrix models using maximum likelihood or Bayesian inference. It can use many loci (with different mutation rates) of infinite allele data, DNA, SNP, or microsatellite data. It runs on Desktop and parallel computers (loci: MPI-parallel, multiple heated chains: multi-threading).

General

All methods presented here use the coalescence theory of Kingman (1982a, b) and estimate population parameters like effective population size, growth rate, migration rate. To achieve this, the methods sample random genealogies of sequences (or alleles) and calculate the parameters. Since this is an integration over almost an infinitely large tree space, the methods use a Metropolis Monte Carlo sampling technique to concentrate the sampling in regions which contribute to the final result.

LAMARC is written in C++; the programs Fluctuate, Recombine and Migrate are available in C source code. All are also available as executables for Windows, MacOS and Linux. They will compile on most workstations. You can find more information to the individual programs following the links below:

Programs available:

Lamarc estimates the effective population sizes, population exponential growth rates, migration rates, and per-site recombination rate of n populations using DNA, SNP, microsatellite, or electrophoretic data. It can also perform trait mapping. (Last update: July 2008)

Migrate estimates the effective population sizes and migration rates of n constant populations using nonrecombining sequence, microsatellite or enzyme electrophoretic data. Migrate is now maintained by Peter Beerli at Florida State University. (Last update: June 2007)

Older programs (for reference):

	`Recombine` estimates the effective population size and per-site recombination rate of a single constant-size population using sequence or SNP data. (last updated: June 2002)
	`Fluctuate` estimates the effective population size and a exponential growth rate of a single growing population using nonrecombining sequence data. (last updated: September 2002)
	`Coalesce` estimates the effective population size of a single constant population using nonrecombining sequences. (last updated: September 1995)

Papers and Posters available:

		Paper introducing LAMARC 2.0 (also available from the Bioinformatics web site).
		Poster presented at the EVO-WIBO meeting 2006 in Port Townsend, Washington .
		Paper about `Migrate` (n-population version).
		Paper about `Migrate` (2-population version).
		Paper about `Fluctuate`.
		Paper about `Coalesce`.

Go back to:

PHYLIP home page
WWW-directory of evolution.gs.washington.edu

Mary K. Kuhner, Jon Yamato (mkkuhner@gs.washington.edu), and Joe Felsenstein, Department of Genome Sciences, University of Washington, Box 355065, Seattle, WA 98195-5065, USA. Phone: (206) 543-8751, fax: (206) 543-0754

	`Lamarc` estimates the effective population sizes, population exponential growth rates, migration rates, and per-site recombination rate of n populations using DNA, SNP, microsatellite, or electrophoretic data. It can also perform trait mapping. (Last update: July 2008)
	`Migrate` estimates the effective population sizes and migration rates of n constant populations using nonrecombining sequence, microsatellite or enzyme electrophoretic data. Migrate is now maintained by Peter Beerli at Florida State University. (Last update: June 2007)