No abstract

Major depressive disorder is a debilitating disease affecting 264 million people worldwide. While many antidepressant medications are available, few clinical guidelines support choosing among them. Decision support tools (DSTs) embodying machine learning models may help improve the treatment selection process, but often fail in clinical practice due to poor system integration.We use an iterative, co-design process to investigate clinicians' perceptions of using DSTs in antidepressant treatment decisions. We identify ways in which DSTs need to engage with the healthcare sociotechnical system, including clinical processes, patient preferences, resource constraints, and domain knowledge. Our results suggest that clinical DSTs should be designed as multi-user systems that support patient-provider collaboration and offer on-demand explanations that address discrepancies between predictions and current standards of care. Through this work, we demonstrate how current trends in explainable AI may be inappropriate for clinical environments and consider paths towards designing these tools for real-world medical systems. CCS CONCEPTS• Human-centered computing → User centered design; • Applied computing → Health care information systems; • Information systems → Decision support systems.

The human T-cell leukemia retrovirus type-1 (HTLV-1) p30II protein is a multifunctional latency-maintenance factor that negatively regulates viral gene expression and deregulates host signaling pathways involved in aberrant T-cell growth and proliferation. We have previously demonstrated that p30II interacts with the c-MYC oncoprotein and enhances c-MYC-dependent transcriptional and oncogenic functions. However, the molecular and biochemical events that mediate the cooperation between p30II and c-MYC remain to be completely understood. Herein we demonstrate that p30II induces lysine-acetylation of the c-MYC oncoprotein. Acetylation-defective c-MYC Lys→Arg substitution mutants are impaired for oncogenic transformation with p30II in c-myc−/− HO15.19 fibroblasts. Using dual-chromatin-immunoprecipitations (dual-ChIPs), we further demonstrate that p30II is present in c-MYC-containing nucleoprotein complexes in HTLV-1-transformed HuT-102 T-lymphocytes. Moreover, p30II inhibits apoptosis in proliferating cells expressing c-MYC under conditions of genotoxic stress. These findings suggest that c-MYC-acetylation is required for the cooperation between p30II/c-MYC which could promote proviral replication and contribute to HTLV-1-induced carcinogenesis.

Recent years have seen a boom in interest in interpretable machine learning systems built on models that can be understood, at least to some degree, by domain experts. However, exactly what kinds of models are truly human-interpretable remains poorly understood. This work advances our understanding of precisely which factors make models interpretable in the context of decision sets, a specific class of logic-based model. We conduct carefully controlled human-subject experiments in two domains across three tasks based on human-simulatability through which we identify specific types of complexity that affect performance more heavily than others–trends that are consistent across tasks and domains. These results can inform the choice of regularizers during optimization to learn more interpretable models, and their consistency suggests that there may exist common design principles for interpretable machine learning systems.

Organophosphorus (OP) nerve agents are deadly chemical weapons that pose an alarming threat to military and civilian populations. The irreversible inhibition of the critical cholinergic degradative enzyme acetylcholinesterase (AChE) by OP nerve agents leads to cholinergic crisis. Resulting excessive synaptic acetylcholine levels leads to status epilepticus that, in turn, results in brain damage. Current countermeasures are only modestly effective in protecting against OP-induced brain damage, supporting interest for evaluation of new ones. (-)-Phenserine is a reversible AChE inhibitor possessing neuroprotective and amyloid precursor protein lowering actions that reached Phase III clinical trials for Alzheimer's Disease where it exhibited a wide safety margin. This compound preferentially enters the CNS and has potential to impede soman binding to the active site of AChE to, thereby, serve in a protective capacity. Herein, we demonstrate that (-)-phenserine protects neurons against soman-induced neuronal cell death in rats when administered either as a pretreatment or post-treatment paradigm, improves motoric movement in soman-exposed animals and reduces mortality when given as a pretreatment. Gene expression analysis, undertaken to elucidate mechanism, showed that (-)-phenserine pretreatment increased select neuroprotective genes and reversed a Homer1expression elevation induced by soman exposure. These studies suggest that (-)-phenserine warrants further evaluation as an OP nerve agent protective strategy.