Current release: ALDER v1.03 (March 5, 2013)
The ALDER software computes the weighted linkage disequilibrium (LD) statistic for making inference about population admixture described in:
- Loh P-R, Lipson M, Patterson N, Moorjani P, Pickrell JK, Reich D, and Berger B. Inferring Admixture Histories of Human Populations Using Linkage Disequilibrium. Genetics, 2013.
Abstract (excerpt): Long-range migrations and the resulting admixture between populations have been an important force shaping human genetic diversity. Most existing methods for detecting and reconstructing historical admixture events are based on allele frequency divergences or patterns of ancestry segments in chromosomes of admixed individuals. An emerging new approach harnesses the exponential decay of admixture-induced linkage disequilibrium (LD) as a function of genetic distance. Here, we comprehensively develop LD-based inference into a versatile tool for investigating admixture. We present a new weighted LD statistic that can be used to infer mixture proportions as well as dates with fewer constraints on reference populations than previous methods. We define an LD-based three-population test for admixture and identify scenarios in which it can detect admixture that previous formal tests cannot. We further show that we can discover phylogenetic relationships between populations by comparing weighted LD curves obained using a suite of references. Finally, we describe several improvements to the computation and fitting of weighted LD curves that greatly increase the robustness and speed of the computation.
Weighted LD decay curves for Mbuti Pygmies using various reference populations.
Admixture analysis using weighted LD decay curves was first proposed by Moorjani et al. (PLoS Genetics, 2011), whose ROLLOFF software package, described more fully in Patterson et al. (Genetics, 2012), infers admixture dates from weighted LD curves. ALDER extends the general methodology of ROLLOFF, the most notable advances being:
- a new form of the weighted LD statistic that is more robust and makes the amplitude of the weighted LD curve interpretable
- a variant of the statistic that calculates unbiased weighted LD using the test population itself as one reference
- a statistical test for admixture
- automatic determination of the minimum genetic distance at which to start curve fitting (to avoid confounding signal from background LD)
- calculation of the affine term (i.e., horizontal asymptote of the curve) using inter-chromosome SNP pairs
- a fast Fourier transform (FFT) algorithm for computing weighted LD that provides speedup of multiple orders of magnitude.
For reference, the ROLLOFF software is included in the ADMIXTOOLS package on the Reich Lab software page.
Our source code and a Linux executable can be downloaded here for academic and non-profit use:
The package contains a detailed README explaining how to install and use the software.
Version 1.03 (Mar 5, 2013):
- Applied compatibility patch for Mac OS X contributed by Ryan Rauum. (Program behavior is identical to v1.02.)
Version 1.02 (Feb 9, 2013): alder_v1.02.tar.gz
- Added support for input in "packed geno" format (2 bits/genotype).
- Added option 'print_jackknife_fits' to display decay and amplitude fits corresponding to individual jackknife reps.
- Fixed bug causing crash in cases with long-range LD.
- Added check for empty parameters; exit with error message instead of crashing.
- Fixed bug in LD correlation check output when parameter 'approx_ld_corr: NO' is set.
- Fixed bug causing failure to set LD correlation threshold to infinity in cases where long-range correlated LD is discovered on a pass after the first.
Version 1.01 (Nov 15, 2012; internal):
- Added 'snp_outname' param: filename to which to write IDs of SNPs used in the analysis.
Version 1.0 (Oct 31, 2012): alder_v1.0.tar.gz
- Initial release.
We welcome feedback, questions and suggestions. Contact information is available at the primary authors' websites: