Penalized regression and model selection methods for polygenic scores on summary statistics

AbstractPolygenic scores (PGS) summarize the genetic contribution of a person’s genotype to a disease or phenotype. They can be used to group participants into different risk categories for diseases, and are also used as covariates in epidemiological analyses. A number of possible ways of calculating polygenic scores have been proposed, and recently there is much interest in methods that incorporate information available in published summary statistics. As there is no inherent information on linkage disequilibrium (LD) in summary statistics, a pertinent question is how we can make use of LD information available elsewhere to supplement such analyses. To answer this question we propose a method for constructing PGS using summary statistics and a reference panel in a penalized regression framework, which we call lassosum. We also propose a general method for choosing the value of the tuning parameter in the absence of validation data. In our simulations, we showed that pseudovalidation often resulted in prediction accuracy that is comparable to using a dataset with validation phenotype and was clearly superior to the conservative option of setting the tuning parameter of lassosum to its lowest value. We also showed that lassosum achieved better prediction accuracy than simple clumping and p-value thresholding in almost all scenarios. It was also substantially faster and more accurate than the recently proposed LDpred.

Download Full-text

Polygenic scores via penalized regression on summary statistics

Genetic Epidemiology ◽

10.1002/gepi.22050 ◽

2017 ◽

Vol 41 (6) ◽

pp. 469-480 ◽

Cited By ~ 58

Author(s):

Timothy Shin Heng Mak ◽

Robert Milan Porsch ◽

Shing Wan Choi ◽

Xueya Zhou ◽

Pak Chung Sham

Keyword(s):

Penalized Regression ◽

Summary Statistics ◽

Polygenic Scores

Download Full-text

Faculty Opinions recommendation of Polygenic scores via penalized regression on summary statistics.

Faculty Opinions – Post-Publication Peer Review of the Biomedical Literature ◽

10.3410/f.727593680.793563010 ◽

2019 ◽

Author(s):

John Nurnberger

Keyword(s):

Penalized Regression ◽

Summary Statistics ◽

Polygenic Scores

Download Full-text

Contrasting Broad- and Clinically- defined Polygenic Indicators of Depression and Depression-related Phenotypes in Adults and Children

10.31234/osf.io/pn9vb ◽

2020 ◽

Author(s):

John E. McGeary ◽

Chelsie Benca-Bachman ◽

Victoria Risner ◽

Christopher G Beevers ◽

Brandon Gibb ◽

...

Keyword(s):

Suicidal Ideation ◽

Cognitive Reappraisal ◽

Twin Studies ◽

European Ancestry ◽

Summary Statistics ◽

Depression Severity ◽

Uk Biobank ◽

Polygenic Scores ◽

Adults And Children ◽

The Uk

Twin studies indicate that 30-40% of the disease liability for depression can be attributed to genetic differences. Here, we assess the explanatory ability of polygenic scores (PGS) based on broad- (PGSBD) and clinical- (PGSMDD) depression summary statistics from the UK Biobank using independent cohorts of adults (N=210; 100% European Ancestry) and children (N=728; 70% European Ancestry) who have been extensively phenotyped for depression and related neurocognitive phenotypes. PGS associations with depression severity and diagnosis were generally modest, and larger in adults than children. Polygenic prediction of depression-related phenotypes was mixed and varied by PGS. Higher PGSBD, in adults, was associated with a higher likelihood of having suicidal ideation, increased brooding and anhedonia, and lower levels of cognitive reappraisal; PGSMDD was positively associated with brooding and negatively related to cognitive reappraisal. Overall, PGS based on both broad and clinical depression phenotypes have modest utility in adult and child samples of depression.

Download Full-text

Bayesian Model Averaging to Account for Model Uncertainty in Estimates of a Vaccine's Effectiveness

10.1101/2021.05.12.21257126 ◽

2021 ◽

Author(s):

Carlos R Oliveira ◽

Eugene D Shapiro ◽

Daniel M Weinberger

Keyword(s):

Model Selection ◽

Model Uncertainty ◽

Bayesian Model ◽

Bayesian Model Averaging ◽

Model Averaging ◽

Selection Methods ◽

Final Model ◽

Negative Case ◽

Confounder Selection ◽

Control Study

Vaccine effectiveness (VE) studies are often conducted after the introduction of new vaccines to ensure they provide protection in real-world settings. Although susceptible to confounding, the test-negative case-control study design is the most efficient method to assess VE post-licensure. Control of confounding is often needed during the analyses, which is most efficiently done through multivariable modeling. When a large number of potential confounders are being considered, it can be challenging to know which variables need to be included in the final model. This paper highlights the importance of considering model uncertainty by re-analyzing a Lyme VE study using several confounder selection methods. We propose an intuitive Bayesian Model Averaging (BMA) framework for this task and compare the performance of BMA to that of traditional single-best-model-selection methods. We demonstrate how BMA can be advantageous in situations when there is uncertainty about model selection by systematically considering alternative models and increasing transparency.

Download Full-text

Learning Coefficient of Vandermonde Matrix-Type Singularities in Model Selection

Entropy ◽

10.3390/e21060561 ◽

2019 ◽

Vol 21 (6) ◽

pp. 561

Author(s):

Miki Aoyagi

Keyword(s):

Model Selection ◽

Information Criteria ◽

Learning Systems ◽

Vandermonde Matrix ◽

Selection Methods ◽

Learning Models ◽

Matrix Type ◽

Log Canonical Threshold ◽

Log Canonical ◽

Blowing Up

In recent years, selecting appropriate learning models has become more important with the increased need to analyze learning systems, and many model selection methods have been developed. The learning coefficient in Bayesian estimation, which serves to measure the learning efficiency in singular learning models, has an important role in several information criteria. The learning coefficient in regular models is known as the dimension of the parameter space over two, while that in singular models is smaller and varies in learning models. The learning coefficient is known mathematically as the log canonical threshold. In this paper, we provide a new rational blowing-up method for obtaining these coefficients. In the application to Vandermonde matrix-type singularities, we show the efficiency of such methods.

Download Full-text