Many clinical and research efforts aim to develop antidepressant drugs for those suffering from major depressive disorder (MDD). Yet even today, the available treatments are suboptimal and unpredictable, with a significant proportion of patients enduring multiple drug attempts and adverse side effects before a successful response; and for many patients, no response at all. Thus, a clearer understanding of the mechanisms underlying MDD is necessary. In the “Brain development and disease” class of our Master’s program in Cognitive Sciences, we ask students to collect data about the expression of a gene whose altered expression and/or function is related to a brain disorder. The students’ final exam assignment consists of writing a research article in which the collected data are discussed in relation to the relevant disorder. In the course of one of these students’ assignments, we identified the FKBP5 gene as a key player uniting two major hypotheses of MDD pathogenesis and treatment response. FKBP5 is involved in biological processes including immunoregulation and glucocorticoid function, both of which are separately implicated in the development and prognosis of MDD. Gene expression analysis from the human and non-human primate Allen Brain Atlases revealed that FKBP5 is expressed in brain regions involved in MDD, particularly at periods of development when early-life stressors would have a great effect. Future works should focus on causal mechanisms of FKBP5 for antidepressant response. Our experience shows that classes engaging students in data collection and analysis projects may effectively result in novel data-driven hypotheses.

A fundamental problem for any sensory system is that the motion of the sensory organ introduces a disturbance in sensory measurement. Nevertheless, animal nervous systems can distinguish and compensate for the sensory consequences of movement by keeping track of them. Hypotheses for estimating the motion of a sensory organ include a corollary discharge associated with motor output and mechanosensory inputs from the moving sense organ. However, how motion-related mechanosensory inputs can reduce motion-dependent sensory ambiguity is unknown. We addressed this question in the olfactory system, where changes in inhalation dynamics alter the sorption of odor molecules by the nasal mucosa and mechanically activate olfactory sensory neurons. We found that changes in inhalation dynamics varied the odor concentration reported by the individual neurons in the piriform cortex. However, the actual odor concentration could be decoded from the population activity regardless of the inhalation dynamics. Dynamics-invariant decoding was possible because piriform cortex neurons have a heterogeneously mixed sensitivity for odor concentration and inhalation dynamics. Thus, the mechanosensory input generated by inhalation expanded the dimensionality of the olfactory encoding space and disentangled the decoding of the actual odor concentration from the disturbance caused by the sniff. Orthogonal integration of motion-related mechanosensory information during the encoding stage may be a canonical strategy adopted by nervous systems to mitigate the sensory ambiguities generated by the motion of the sensory organ.