Gravar-mail: A regression framework for assessing covariate effects on the reproducibility of high-throughput experiments