Figure 1.

Predicted probabilities of GWAS signal for conserved and non-conserved lincRNAs.

Predicted probabilities and corresponding 95% prediction intervals are calculated based on multivariable models using average of observed median values for GC content and TE coverage and observed trait-specific median number of SNPs. The predicted probability of GWAS signal is greater for non-conserved lincRNAs than conserved lincRNAs for all traits considered except height based on the primary definition of conservation. The results based on the secondary definition of conservation are consistent though in this case, the predicted probably of GWAS signal is greater for non-conserved lincRNAs than conserved lincRNAs for all traits including height (results not shown). The consistently overlapping confidence intervals suggest that the likelihood of GWAS association for conserved and non-conserved lincRNAs is comparable and, therefore, the traditional metrics of conservation for prioritizing lncRNAs for functional studies needs to be reconsidered.