UMass ADVANCE is pleased to announce that three research teams are recipients of ADVANCE Collaborative Research Seed Grant awards for 2021-2022. These competitive grants aim to foster the development of innovative and equitable collaborative research projects among faculty. Recognizing longstanding gender gaps in the academy, the National Science Foundation (NSF) funds universities to build institutional transformation programs in order to advance gender equity for faculty in science and engineering. Through the power of collaboration, UMass ADVANCE cultivates faculty equity and inclusion—especially for women and minorities in science and engineering. Three winning teams demonstrated innovative research and well thought-out and equitable collaborations.
Investigating Amygdala Circuit Dysfunctions in a Mouse Model Relevant to Schizophrenia
The principal investigators for this team are:
- Karine Fenelon, assistant professor, department of biology
- David Moorman, associate professor, psychological and brain sciences
The filtering of sensorimotor information is a fundamental brain mechanism that, if reduced, is associated with and often predictive of psychosis, attention impairment and cognitive over-load. In humans and translational models, sensorimotor filtering can be measured using the prepulse inhibition (PPI) of the auditory startle response task. Acoustic PPI occurs when a weak sound presented prior to a loud startling sound, inhibits startle. Reduced PPI is a hallmark of schizophrenia but is also seen in other neurological and psychiatric disorders. Currently, the reversal of PPI deficits in animal models is widely used in pre-clinical research for antipsychotic drug screening. Yet, the neurotransmitter systems and synaptic mechanisms underlying PPI deficits are still not resolved. Amygdalar dysfunctions alter PPI and are common to pathologies displaying sensorimotor filtering deficits, including schizophrenia. Therefore here, we aim to identify amygdala mechanisms that cause PPI deficits as promising drug ta-gets, using a mouse model of schizophrenia. To do so, the team will perform in vitro (Fenelon group) and in vivo (Moorman group) electrophysiological recordings of neurons central to PPI.
Accelerating Fragment-Based Quantum Chemistry via Machine Learning
The principal Investigators for the team are:
- Zhou Lin, assistant professor, department of chemistry
- Hui Guan, assistant professor, college of information and computer sciences
The high-throughput design of multi-fragment complex systems with compelling electronic properties is a driving force behind scientific and technological advancement. First-principles quantum chemical calculations are gradually replacing tedious and labor-intensive experiments to promote design productivity. For example, the construction of high-quality doped organic electronic materials requires a computational search of polymer-dopant composites with fast electron-hole separations and slow electron-hole recombination. However, first-principles evaluations of these properties are generally infeasible due to the difficulty in balancing efficiency and accuracy in quantum chemistry. Lin and Guan’s collaborative team proposes developing data-driven computational methods for modeling energetics and dynamics of composite systems with enhanced accuracy and efficiency. Methodologically, we will establish a fragment-based, quantum embedding framework and integrate advanced machine learning algorithms into otherwise expensive evaluations of inter-fragment interactions. As the first step, we will construct graph learning-based models to capture quantitative relationships between effortless molecular descriptors and targeted many-fragment interactions based on van der Waals molecular aggregations. This strategy will remove the computational bottleneck of composite systems without compromising their accuracy and provide insights into predicting and interpreting their energetics and dynamics. Both aspects will make our study significant and unique in the high-throughput rational design of composite systems.
Portable, Robotic Footwear to Actively Modulate Foot-ground Stiffness in Real-time
The principal investigators for this team are:
- Meghan Huber, assistant professor, department of mechanical and industrial engineering
- Wouter Hoogkamer, assistant professor, department of kinesiology
Impairments to upright balance and locomotion can present in humans for a multitude of reasons, including injury, neurological disease, or even simply aging, and when they do, one’s quality of life is often markedly reduced. Thus, highly effective methods for gait and balance rehabilitation are greatly needed. The control of foot-ground interaction dynamics plays a pivotal role in maintaining standing balance and locomotion during many activities of human living. Human feet serve as interfaces through which the body and ground simultaneously act upon each other and through which the body can sense the physical world around it. Currently, the existence of tools to study, and ultimately assist or re-train, how humans manage interaction with the ground through their feet is limited. To address this gap, the overall goal of the proposed project is to design, build, and evaluate portable, robotic footwear that can actively modulate foot-ground stiffness and measure the ground reaction forces under each foot. The added advantage of such a portable research tool is that it can be used to study or modify human behavior not only in tightly controlled laboratory tasks, but also in a wide variety of tasks that require whole-body control in real-world contexts.