Assistant Professor, Department of Electrical Engineering and Computer Sciences
Title: Multimodal Intelligent Interfaces for Assistive Communication
Abstract: Several neurological conditions (like cerebral palsy, ALS) result in paralysis of all voluntary movements, including those of the vocal apparatus, causing total loss of spoken communication. Current assistive technologies for communication are based on residual functions like eye gaze, blinks or twitch of the cheek muscle to move a cursor on a screen and select characters/icons on an on-screen display. These solutions are often erroneous, unintuitive and laborious for paralyzed users. We propose to improve the current state of assistive communication technologies by integrating multiple neural and behavioral sensing modalities, and tightly integrating the graphical interfaces, and personalizing them to the user’s context. We will use state-of-the-art neural engineering and artificial intelligence to develop novel communication interfaces. Among the neural sensing modalities we will use are Electrocorticography (ECoG, invasive recordings from the surface of the cortex), and the non-invasive modalities include in-ear Electroencephalography (EEG) sensors and functional near infrared spectroscopy (fNIRS). These neural sensors augment eye trackers and Electromyography (EMG) sensors. On the software side, we will use on-device speech recognition and dialog management to incorporate the acoustic context of the user (speech directed towards the user), to provide an adaptive interface including word/ phrase level autocomplete suggestions driven by language models. Where available, we will create a digital voice from the subject’s own voice recordings, so the system can synthesize speech that sounds like the user, enhancing their sense of embodiment. We believe these strategies will lead to better information rates and robust communication solutions for paralyzed users.
Profile: Gopala Anumanchipalli is an Assistant Professor in Electrical Engineering and Computer Sciences at UC Berkeley and an Adjunct Assistant Professor at Dept. of Neurosurgery at UC San Francisco. His lab works at the intersection of Speech Processing, Neuroscience, and Artificial Intelligence with an emphasis on human-centered Speech and Assistive technologies. His work involves developing computational paradigms for bio-inspired spoken language processing and automated methods for early diagnosis, characterization and rehabilitation for disordered speech. He also develops methods to advance understanding of the neural mechanisms underlying speech/ language function in healthy individuals and creates algorithms for Brain-Computer Interfacing to decode these signals directly from the brain to externally restore communication in paralyzed patients.
Assistant Professor, Department of Molecular & Cell Biology
Title: Psychedelics-inspired plasticity therapeutics
Abstract: A rigid mind limits cognition. In neurological disorders, our ability to move, remember, empathize, learn, or recover from trauma is not limited by will but by the capacity of our neurons to generate components for mental flexibility. Plasticity in the brain counters this rigidity, yet we lack insight into opening plasticity windows in the brain. Psychedelics, such as psilocybin, have been used in Indigenous healing practices for millennia to engage cognitive connection and open pathways to healing. Likewise, intensive clinical exploration of psychedelics since the mid-20th century reveals their potential for treating neuropsychiatric conditions. While we know our genome contains the instructive cues for plasticity, the genome, by itself, cannot directly facilitate plasticity. Instead, it relies on RNA to coordinate instructive cues to alter neural circuit function. Andrea Gomez's research synthesizes two rapidly evolving fields: RNA biology and psychedelics. Advances in genome sequencing technologies highlight the diversity of RNA expressed in the brain, yet linking RNA to neural function remains a core challenge in neurobiology. Likewise, psychedelics' therapeutic potential comes at a critical time for innovation in mental health, as current treatments for neurological and psychiatric disorders are marginally effective. Her unique strategy using psychedelics to identify molecular plasticity programs will enable the discovery of novel RNA-based therapeutic targets to treat neurological conditions associated with defective neural plasticity.
Profile: Andrea Gomez (Laguna Pueblo/Chicana) is an Assistant Professor in the Department of Molecular and Cell Biology and the Helen Wills Neuroscience Institute at the University of California, Berkeley. She is also a member of the Executive Committee at the UC Berkeley Center for the Science of Psychedelics. Gomez received her Ph.D. in Developmental Genetics from New York University and conducted postdoctoral research at the University of Basel, Switzerland. Her work is devoted to understanding the instructive cues that sculpt patterns of brain activity. Her efforts led to the discovery of RNA-based programs that are critical for synaptic organization and plasticity. The Gomez lab uses state-of-the-art techniques to decode the brain's modular nature, including molecular biology, electrophysiology, and functional imaging. The robust and widespread neural plasticity induced by psychedelics motivates the Gomez lab to decode the synaptic mechanisms that underlie cognitive flexibility. Gomez started her lab at UC Berkeley in 2020 and has received several awards, including the European Molecular Biology Organization Advanced Fellowship, a Rennie Fund Fellow, a C.J. Herrick Early-Career Investigator, and is a Brain and Behavior Research Foundation Young Investigator.
Assistant Professor, Department of Bioengineering
Title: Decoding the perfect message: mRNA design for vaccines and therapeutics
Abstract: The first approved mRNA vaccines, administered less than a year after the COVID pandemic began, have changed the face of medicine. I have focused my career on understanding how mRNA sequences specify a precise outcome: the right amount of protein made at the right time. My work has uncovered unanticipated constraints on sequence design. Now, I am applying my research to learn the rules to design the perfect synthetic mRNAs for therapeutic purposes in a wide range of different cellular contexts. Our powerful combination of machine learning and new genome-scale experiments will allow us to synthesize a new model of the parameters governing protein output and efficacy from therapeutic mRNAs.
Profile: Liana Lareau works to decipher the layers of information encoded in the genome that specify cellular form and function. Using machine learning and other computational approaches, she investigates how sequence determines the output of the genome as it is transcribed into RNA then translated into protein. She is a Shirl and Kay Curci Foundation Faculty Scholar and recently won the Excellence in Research award from the Laboratory for Genomics Research in support of her work sequencing SARS-CoV-2 virus genomes for pandemic surveillance. Before starting her faculty position in Berkeley’s Bioengineering department in 2019, she was supported by a Damon Runyon Postdoctoral Research Fellowship while working in the lab of Pat Brown at Stanford, where she used the new technique of ribosome profiling to understand translation of mRNAs. She is a proud Berkeley alumna, having completed her PhD here at Berkeley after undergraduate degrees in mathematics and biology from MIT.