When the Toxic Substances Control Act was enacted in 1976 roughly 60,000 existing chemicals were considered safe for use and grandfathered in. Today there are approximately 85,000 chemicals on the list, with around 2,000 new chemicals introduced each year. It is impossible to screen all of these chemicals via full-organism in vivo studies. High-throughput toxicity screening (HTS) programs allow for the relatively cheap and fast collection of dose-response information in vitro, which can provide clues as to chemicals' potential toxic effects. We propose to use these HTS dose-response curves as supervision information of sparse linear dimension reduction of structural features. Specifically, we propose a Bayesian partially shared latent factor joint model imposing sparsity on chemical structure loadings and smoothness on dose-response loadings. We show simulation performance compared to existing methods, and preliminary results using high-throughout screening data. We also illustrate how this model can be used to generate a coherent pairwise distance metric informed by both chemical structure and toxicity.