Whole Genome Prediction of Bladder Cancer Risk With the Bayesian LASSO

Abstract

To build a predictive model for urothelial carcinoma of the bladder (UCB) risk combining both genomic and nongenomic data, 1,127 cases and 1,090 controls from the Spanish Bladder Cancer/EPICURO study were genotyped using the HumanHap 1M SNP array. After quality control filters, genotypes from 475,290 variants were available. Nongenomic information comprised age, gender, region, and smoking status. Three Bayesian threshold models were implemented including: (1) only genomic information, (2) only nongenomic data, and (3) both sources of information. The three models were applied to the whole population, to only nonsmokers, to male smokers, and to extreme phenotypes to potentiate the UCB genetic component. The area under the ROC curve allowed evaluating the predictive ability of each model in a 10-fold cross-validation scenario. Smoking status showed the highest predictive ability of UCB risk (AUCtest = 0.62). On the other hand, the AUC of all genetic variants was poorer (0.53). When the extreme phenotype approach was applied, the predictive ability of the genomic model improved 15%. This study represents a first attempt to build a predictive model for UCB risk combining both genomic and nongenomic data and applying state-of-the-art statistical approaches. However, the lack of genetic relatedness among individuals, the complexity of UCB etiology, as well as a relatively small statistical power, may explain the low predictive ability for UCB risk. The study confirms the difficulty of predicting complex diseases using genetic data, and suggests the limited translational potential of findings from this type of data into public health interventions.

Whole genome prediction and heritability of childhood asthma phenotypes

“…These include inclusion criteria requiring mild or moderate persistent childhood asthmatics; and exclusion of both those with very mild asthma and those with severe asthma. While the measurement of accuracy with hold-out test-sets or cross-validation schemes is typical in Whole Genome Prediction studies [22,35,63], the greatest test of these prediction methodologies is to apply them prospectively in an independent cohort; our results indicate that such test should result in significant prediction of a number of longitudinal spirometric phenotypes. Comprehensive longitudinal lung function pattern phenotypes are difficult to assess in additional cohorts, although it would be of great interest to investigate them further.…”

Statistical considerations for genomic selection

“…The theory of GS can also be applied to human disease risk prediction. Case studies have been conducted on Celiac disease [98,99] , type I diabetes [100] , coronary heart disease [101] , breast cancer [102] , bladder cancer [103] , skin cancer [104] , and others. As Visscher [105] predicted, personalized genetics and genomics will become an integral part of health care and clinical practice in future.…”

A multiple-trait Bayesian Lasso for genome-enabled analysis and prediction of complex traits

“…GWAS is dominant in human genetics; Visscher et al (2017) present a perspective and Gianola et al (2016) formulate a statistically orientated critique. WGR was developed mostly in animal and plant breeding (e.g., Lande and Thompson 1990; Meuwissen et al 2001; Gianola et al 2003) primarily for predicting future performance, but it has received some attention in human genetics as well (e.g., Lee et al 2011; Yang et al 2010; de los Campos et al 2011; Makowsky et al 2011; López de Maturana et al 2014). de los Campos et al (2013), Gianola (2013) and Isik et al (2017) reviewed an extensive collection of WGR approaches.…”