Introduction

LDAK 5 (beta) now out! – Click on Downloads

To accompany Version 5, we have written a step-by-step guide for computing SNP heritability, starting with raw data, as well as additional features.
_ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _

LDAK is back, and better than ever! Version 5 is being released to accompany our latest submission (see BiorXiv). Basically, LDAK5 currently does the following:

Calculates LDAK SNP weights (now quicker than before)
Computes weighted and unweighted kinships
Performs kinship manipulation (add/subtracting/converting/decomposing)

Generalised REML Solver – allows for both kinships and regions
Calculation of BLUP effect size estimates
Haseman-Elston Regression
Super-efficient mixed-model analysis
Genetic risk scoring – allows for multiple profiles
Principal component analysis
Generates simulated genotypes and phenotypes
Can accommodate / convert to a wide-range of data formats

I have not yet recoded the gene-based analysis, so for MultiBLUP, please continue to use Version 4. LDAK5 is still beta, so please email doug<dot>speed<at>ucl<dot>ac<dot>uk with any bug reports.

The original use of LDAK was to adjust for linkage disequilibrium (LD) by calculating SNP weightings which downweight the contribution of SNPs in highly tagged regions. The idea is outlined here or see the AJHG paper for full details.

LDAK now also includes MultiBLUP, a method for creating linear prediction models. This improves upon standard BLUP by allowing for multiple kinship matrices (see our Genome Research paper here). Either the user can provide these kinship matrices, or they can be determined using Adaptive MultiBLUP.
_ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _

To get started, download the executables and test datasets from here.