Maltz, Evan
Evan is a graduate student in the Biochemistry, Molecular and Structural Biology Program. He received B.S. degrees in Molecular Biology and Biochemistry from the University of Wisconsin, Madison, and then came to UCLA and joined the laboratory of Dr. Roy Wollman. He entered the CMB Training Program in 2018.
Mentor: Dr. Roy Wollman
As an integral component of the innate immune response, NF-κB is one of the most extensively studied models of transcription factor dynamics. The most evidenced mathematical models implicate stimulus-dependent NF-κB translocation dynamics in control of relevant mRNA transcript abundance, though the connection is yet insufficiently characterized. Based on these models, I hypothesize that NF-κB dynamics encode information about instigatory ligands, which is influentially decoded downstream during transcription. Despite literature characterizing NF-κB dynamics, its specific effects on gene expression have remained elusive due to complete cellular heterogeneity in NF-κB response. Thus far, attempts to precisely correlate NF-κB dynamics with gene expression have been significantly limited because of the need to measure both NF-κB relative nuclear and mRNA transcript abundance at the single-cell level over long time periods. While fluorescence microscopy has been the primary tool for characterizing NF-κB translocation dynamics, no bona fide solution exists for quantifying changes in gene expression without confounding cell identities.
Recent approaches for measuring single-cell mRNA transcript abundance compromise either throughput or sequencing depth. However, sequential smFISH (seqFISH) is an emergent technology that can simultaneously locate and quantify all molecules of theoretically unlimited mRNA species, solving both depth and breadth problems. Combining augmented microscopy with seqFISH enables us to use robust, image-based tracking of single-cell measurements to correlate dynamics with transcript abundance. Furthermore, DNA barcoding combined with fluorescent labeling/imaging of downstream protein targets of NF-κB via CD-tagging (central dogma) completes our analysis of the genetic pipeline. The scale of image acquisition and the absence of stark morphological changes presents a problem most feasibly solved by deep learning-based automation and analysis. Despite advancements and increasing adoption, there is no standard methodology for incorporating deep learning into traditional acquisition and analysis pipelines. Convolutional neural networks will be used for tracking and cell segmentation in all subsequent aims, representing the first multi-platform, deep learning-based microscopy analysis pipeline. Correlating single-cell NF-κB dynamics with changes in gene expression and translation promises deeper insight into the general principles of cell signaling.
