Discovering target and off-target effects of specific compounds is critical to drug discovery and development. We generated a compendium of "chemical-genetic interaction" profiles by testing the collection of viable yeast haploid deletion mutants for hypersensitivity to 82 compounds and natural product extracts. To cluster compounds with a similar mode-of-action and to reveal insights into the cellular pathways and proteins affected, we applied both a hierarchical clustering and a factorgram method, which allows a gene or compound to be associated with more than one group. In particular, tamoxifen, a breast cancer therapeutic, was found to disrupt calcium homeostasis and phosphatidylserine (PS) was recognized as a target for papuamide B, a cytotoxic lipopeptide with anti-HIV activity. Further, the profile of crude extracts resembled that of its constituent purified natural product, enabling detailed classification of extract activity prior to purification. This compendium should serve as a valuable key for interpreting cellular effects of novel compounds with similar activities.
Affinity propagation (AP) was recently introduced as an unsupervised learning algorithm for exemplar-based clustering. We present a derivation of AP that is much simpler than the original one and is based on a quite different graphical model. The new model allows easy derivations of message updates for extensions and modifications of the standard AP algorithm. We demonstrate this by adjusting the new AP model to represent the capacitated clustering problem. For those wishing to investigate or extend the graphical model of the AP algorithm, we suggest using this new formulation since it allows a simpler and more intuitive model manipulation.
Affinity propagation (AP) was recently introduced as an unsupervised learning algorithm for exemplar-based clustering. We present a derivation of AP that is much simpler than the original one and is based on a quite different graphical model. The new model allows easy derivations of message updates for extensions and modifications of the standard AP algorithm. We demonstrate this by adjusting the new AP model to represent the capacitated clustering problem. For those wishing to investigate or extend the graphical model of the AP algorithm, we suggest using this new formulation since it allows a simpler and more intuitive model manipulation.
scite is a Brooklyn-based organization that helps researchers better discover and understand research articles through Smart Citations–citations that display the context of the citation and describe whether the article provides supporting or contrasting evidence. scite is used by students and researchers from around the world and is funded in part by the National Science Foundation and the National Institute on Drug Abuse of the National Institutes of Health.