Search

We develop a scalable and highly efficient algorithm to fit a Cox proportional hazard model by maximizing the $L^1$-regularized (Lasso) partial likelihood function, based on the Batch Screening Iterative Lasso (BASIL) method developed in Qian and others (2019). Our algorithm is particularly suitable for large-scale and high-dimensional data that do not fit in the memory. The output of our algorithm is the full Lasso path, the parameter estimates at all predefined regularization parameters, as well as their validation accuracy measured using the concordance index (C-index) or the validation deviance. To demonstrate the effectiveness of our algorithm, we analyze a large genotype-survival time dataset across 306 disease outcomes from the UK Biobank (Sudlow and others, 2015). We provide a publicly available implementation of the proposed approach for genetics data on top of the PLINK2 package and name it snpnet-Cox., (© The Author 2020. Published by Oxford University Press. All rights reserved. For permissions, please e-mail: journals.permissions@oup.com.)

Motivation: The prediction performance of Cox proportional hazard model suffers when there are only few uncensored events in the training data., Results: We propose a Sparse-Group regularized Cox regression method to improve the prediction performance of large-scale and high-dimensional survival data with few observed events. Our approach is applicable when there is one or more other survival responses that 1. has a large number of observed events; 2. share a common set of associated predictors with the rare event response. This scenario is common in the UK Biobank dataset where records for a large number of common and less prevalent diseases of the same set of individuals are available. By analyzing these responses together, we hope to achieve higher prediction performance than when they are analyzed individually. To make this approach practical for large-scale data, we developed an accelerated proximal gradient optimization algorithm as well as a screening procedure inspired by Qian et al., Availabilityandimplementation: https://github.com/rivas-lab/multisnpnet-Cox., (© The Author(s) 2021. Published by Oxford University Press. All rights reserved. For permissions, please e-mail: journals.permissions@oup.com.)

Context: Recently, 10 novel type 2 diabetes (T2D) susceptibility single nucleotide polymorphisms (SNPs) in ZMIZ1, ANK1, KLHDC5, TLE1, ANKRD55, CILP2, MC4R, BCAR1, HMG20A, and GRB14 loci were discovered in MetaboChip-genotyped populations of European ancestry., Objective: The aim of the present study was to characterize prediabetic quantitative traits underlying these SNP associations and to calculate the amount of interindividual variation in glycemic traits explained by these and previous T2D susceptibility variants., Design and Participants: A total of 5739 Danish individuals naive to glucose-lowering medication were included in quantitative trait studies, and case-control analyses were performed in 1892 patients with T2D and 6603 normoglycemic control subjects. Participants without known T2D underwent an oral glucose tolerance test, and measures of insulin release and sensitivity were estimated from insulinogenic, disposition, BIGTT, and Matsuda indexes., Results: We confirmed associations of ZMIZ1, KLHDC5, CILP2, HMG20A, ANK1, ANKRD55, and BCAR1 with T2D. The risk T allele of BCAR1 rs7202877 associated with decreased disposition index (P = .02). The C allele of ANK1 rs516946 associated with decreased insulinogenic (P = .005) and disposition (P = .002) indexes. The G allele of ANKRD55 rs459193 associated with decreased Matsuda index (P = .02) adjusted for waist circumference. The C allele of GRB14 rs13389219 associated with both increased insulinogenic (P = .04) and decreased Matsuda (P = .05) indexes. All validated European T2D variants still only explained a few percentage points of glycemic trait variation., Conclusions: BCAR1 rs7202877 may mediate its diabetogenic impact through impaired β-cell function, but this finding needs to be replicated in independent studies. In addition, we substantiated previous evidence that ANK1 rs516946 confers impaired insulin release and that ANKRD55 rs459193 and GRB14 rs13389219 associate with insulin resistance.