< Back to Article

Investigating the effect of dependence between conditions with Bayesian Linear Mixed Models for motif activity analysis

Fig 5

IQR-study on GTEx motif-weight.

(A): Pearson correlation values between the predicted posterior motif influence from the Bayesian Linear Mixed Model and the Ridge Regression of 56 selected motif values over all tissues. (B): extreme cases of different motif scores per method. Each box shows the predicted motif scores per sample, separated by model assumptions used to predict the scores (dependence, denoted BLMM, colored in red, and indepdence between samples, named RIDGE, colored in blue). Per sample, the first, second and third quantile (as in a boxplot) of the overall motif activity of all motifs on that sample are depicted in gray.

doi: https://doi.org/10.1371/journal.pone.0231824.g005