Investigating the functions of genes under various conditions and their interacting networks is necessary for understanding the fundamental biological processes. In bacteria, large-scale genome-wide screening techniques like Transposon insertion sequencing (TIS) can link genes to phenotypes on a comprehensive level, thus a valuable tool in the functional annotation of genetic elements. Through this work, we propose a model-based framework that uses regularized negative binomial regression to estimate the change in transposon insertions attributable to gene-environment changes in a genetic interaction study without transformations or uniform normalization. We also propose a systematic multilevel analysis approach to dissect the genetic modulators of protein homeostasis in Caulobacter crescentus.