BACKGROUND: Heritability and genetic correlation can be estimated from genome-wide single-nucleotide polymorphism (SNP) data using various methods. We recently developed multivariate genomic-relatedness-based restricted maximum likelihood (MGREML) for statistically and computationally efficient estimation of SNP-based heritability ([Formula: see text]) and genetic correlation ([Formula: see text]) across many traits in large datasets. Here, we extend MGREML by allowing it to fit and perform tests on user-specified factor models, while preserving the low computational complexity. RESULTS: Using simulations, we show that MGREML yields consistent estimates and valid inferences for such factor models at low computational cost (e.g., for data on 50 traits and 20,000 individuals, a saturated model involving 50 [Formula: see text]'s, 1225 [Formula: see text]'s, and 50 fixed effects is estimated and compared to a restricted model in less than one hour on a single notebook with two 2.7 GHz cores and 16 GB of RAM). Using repeated measures of height and body mass index from the US Health and Retirement Study, we illustrate the ability of MGREML to estimate a factor model and test whether it fits the data better than a nested model. The MGREML tool, the simulation code, and an extensive tutorial are freely available at https://github.com/devlaming/mgreml/ . CONCLUSION: MGREML can now be used to estimate multivariate factor structures and perform inferences on such factor models at low computational cost. This new feature enables simple structural equation modeling using MGREML, allowing researchers to specify, estimate, and compare genetic factor models of their choosing using SNP data.

Traditional extensions of the binary support vector machine (SVM) to multiclass problems are either heuristics or require solving a large dual optimization problem. Here, a generalized multiclass SVM is proposed called GenSVM. In this method classification boundaries for a $K$-class problem are constructed in a $(K-1)$-dimensional space using a simplex encoding. Additionally, several different weightings of the misclassification errors are incorporated in the loss function, such that it generalizes three existing multiclass SVMs through a single optimization problem. An iterative majorization algorithm is derived that solves the optimization problem without the need of a dual formulation. This algorithm has the advantage that it can use warm starts during cross validation and during a grid search, which significantly speeds up the training phase. Rigorous numerical experiments compare linear GenSVM with seven existing multiclass SVMs on both small and large data sets. These comparisons show that the proposed method is competitive with existing methods in both predictive accuracy and training time, and that it significantly outperforms several existing methods on these criteria.

BACKGROUND: Genome-wide association studies have become very popular in identifying genetic contributions to phenotypes. Millions of SNPs are being tested for their association with diseases and traits using linear or logistic regression models. This conceptually simple strategy encounters the following computational issues: a large number of tests and very large genotype files (many Gigabytes) which cannot be directly loaded into the software memory. One of the solutions applied on a grand scale is cluster computing involving large-scale resources. We show how to speed up the computations using matrix operations in pure R code. RESULTS: We improve speed: computation time from 6 hours is reduced to 10-15 minutes. Our approach can handle essentially an unlimited amount of covariates efficiently, using projections. Data files in GWAS are vast and reading them into computer memory becomes an important issue. However, much improvement can be made if the data is structured beforehand in a way allowing for easy access to blocks of SNPs. We propose several solutions based on the R packages ff and ncdf.We adapted the semi-parallel computations for logistic regression. We show that in a typical GWAS setting, where SNP effects are very small, we do not lose any precision and our computations are few hundreds times faster than standard procedures. CONCLUSIONS: We provide very fast algorithms for GWAS written in pure R code. We also show how to rearrange SNP data for fast access.

Bibliographic data and classifications of all the ERIM reports are also available on the ERIM website: www.erim.eur.nl