Some contributions to the statistical modeling of genomics data that use prior biological knowledge to improve interpretability