Talk Abstracts

Transcription is the initial step in gene activation that underlies organismal development and cellular responses. RNA polymerase II (Pol II) transcribes all protein-coding genes and the majority of non-coding transcripts. Pol II transcription involves multiple highly regulated steps, and the deregulation of any of these can lead to various human diseases and developmental anomalies. Recent advances in cryo-electron microscopy have enabled high-resolution structural insights into key regulatory steps during Pol II transcription, such as the initiation and elongation of RNA transcripts. However, the mechanisms of Pol II termination remain poorly understood.

In this talk, I will present our recent structural and biochemical insights into how the metazoan-specific Integrator complex regulates transcription through premature termination. The Integrator is a highly conserved factor with endonuclease activity that binds tightly to protein phosphatase 2A. I will discuss how the interplay of these enzymatic activities facilitates the precise removal of Pol II from the DNA template, thereby regulating gene activity and ensuring genome stability. Additionally, I will emphasize the consequences of failures in this process, which can lead to human diseases. The mechanisms uncovered here have broad applications for understanding transcription termination across eukaryotic organisms.

Synthetic polymeric membranes have become the critical components of a broad range of sustainability and biomanufacturing related applications including 1) water purification and resource recovery, 2) energy generation, conversion and storage, 3) gas separations and chemical manufacturing, 4) CO2 capture, conversion and storage, and 5) downstream bioprocessing to produce protein therapeutics (e.g. monoclonal antibodies) and plasmid DNA for vaccines and gene therapy. However, current commercial polymeric membranes are not very effective at addressing existing and emerging separation and manufacturing challenges in sustainability and biomanufacturing. First, these membranes exhibit a permeability-selectivity trade-off; i.e. highly permeable membranes have low selectivity and vice versa. Second, commercial polymeric membranes have a high fouling propensity. Third, they cannot perform multiple functions (e.g. rejection, ion transport, adsorption and catalysis) without lengthy and costly surface and/or matrix modifications. In my presentation, I will summarize the work that my group has carried out during the last 10 years to develop, optimize, and validate a one-pot and single-step phase inversion casting process to fabricate a new generation of mixed matrix (MM) polymeric membranes with in-situ synthesized functional polymer particles that can carry out multiple functions including solute retention, protein and metal adsorption, and catalytic hydrogenation.1-6 As an illustration of the translation potential of our MM membrane research, I will focus on two applications including 1) fouling resistant ultrafiltration (UF) membranes for water treatment and resource recovery and 2) flow though anion exchange (AEX) microfiltration (MF) membrane adsorbers for monoclonal antibody (mAb) product polishing during downstream bioprocessing.

References:

1. Bateman, O.; Kornfield, J. A. and Diallo, M. S.* Mixed Matrix PVDF Microfiltration Membranes with In-situ Synthesized Polyethyleneimine Particles as a Platform for Flow Through, High Capacity, Weak Base and Salt Tolerant Anion Exchange Membrane Adsorbers for Downstream Bioprocessing. Journal of Membrane Science. Volume 716, February 2025, 123488.

2. Kotte, M. R.; Kuvarega, K. T.; Talapaneni, S. N.; Cho, M. and Diallo, M. S.* A Facile and Scalable Route to the Preparation of Catalytic Membranes with In-situ Synthesized Dendrimer Encapsulated Pt(0) Nanoparticles Using a Low Generation PAMAM Dendrimer (G1-NH2) as Precursor. ACS Applied Materials and Interfaces. 2018, 10, 33238-33251.

3. Kotte, M. R.; Kuvarega, A.; Mamba, B. B; Cho, M. and Diallo, M. S.* Mixed Matrix PVDF Membranes with In-Situ Synthesized PAMAM Dendrimer-Like Particles: A New Class of Sorbents for Cu(II) Recovery from Aqueous Solutions by Ultrafiltration. Environmental Science & Technology 2015, 49, 9431-9442.

4. Kotte, M. R.; Hwang, T.; Han, J-I. and Diallo, M. S.*; A One-Pot Method for the Preparation of Mixed Matrix Polyvinylidene Fluoride Membranes with In-Situ Synthesized and PEGylated Polyethyleneimine Particles. Journal of Membrane Science 2015, 474, 277–287.

5. Hwang, T.; Kotte, M. R.; Han, J-I.; Oh, Y-K and Diallo, M. S.* Microalgae Recovery by Ultrafiltration Using Novel Fouling-Resistant PVDF Membranes with In-Situ PEGylated Polyethyleneimine Particles. Water Research 2015, 73, 181-192.

6. Kotte, M. R.; Cho, M. and Diallo, M. S.* A Facile Route to the Preparation of Mixed Matrix Polyvinylidene Fluoride Membranes with In-Situ Generated Polyethyleneimine Particles. Journal of Membrane Science 2014, 450, 93-102.

Collectively, the many mitochondria within each cell form a dynamic network that spans the cytoplasm and is responsive to metabolic signals and energy demands on the micron scale. Mitochondria are the only organelles inside our cells that contain their own genome, which must be expressed for cells to meet their energy demands. While geneticists and molecular cell biologists have revealed incredible insights about the regulation of the nuclear genome, regulation of the human mitochondrial genome remains understudied. In this talk I will describe principles of mitochondrial network form and function in terms of information processing, and how the residency of mitochondrial genomes within the mitochondria necessitates a systems view of the flow of genetic information, relevant to the biomedical implications of mitochondrial dysfunction. I will describe our forthcoming work using spatial transcriptomics to define where and when genetic information is transmitted within mitochondrial networks. We recently discovered that mitochondrial transcripts are consolidated into micron-scale hubs for translation, which are remodeled during stress to facilitate attenuation of translation needed for cellular recovery. I will conclude with a model in which the spatial organization of mitochondrial gene expression into discrete domains serves to throttle the flow of genetic information to support mitochondrial quality control.

Large language model (LLM)–based agents are increasingly being deployed in multi-agent environments, introducing unprecedented risks of coordinated harmful behaviors. While individual LLMs have already demonstrated concerning capabilities for deception and manipulation, scaling to multi-agent systems could enable qualitatively distinct and more dangerous emergent behaviors. Despite these pressing concerns, there remains a critical gap in our ability to understand and predict how multiple LLM agents might collaborate in harmful ways, such as orchestrating coordinated deception campaigns or amplifying local misinformation into global crises. In the first part of this talk, I will describe work from the UCLA Misinformation, AI & Responsible Society (MARS) lab on measuring persuasive capabilities of debating LLM agents. In the second part, I will introduce a new multi-agent social-simulation environment to enable evaluation of coordinated LLM deception risks by AI researchers, social scientists, and industry partners. This simulation combines advanced LLM agents with game-theoretic modeling to analyze emergent deception behaviors. I will conclude by discussing concrete intervention strategies for disrupting harmful content amplification before it reaches critical mass. In the long term, our research establishes a foundation for responsible scaling of multi-agent AI systems.

STEM professionals today have more career pathways than ever, yet navigating the transition from academia to industry, entrepreneurship, and leadership requires adaptability, strategic thinking, and a willingness to explore uncharted territory. In this talk, I will share my journey from academic research in robotics, machine learning, and plasma physics to applying these technologies in industry settings, including product management roles at Airbnb and Roblox.

We will explore how AI and emerging technologies—such as natural language processing, image recognition, and blockchain—are driving innovation across industries, particularly in consumer marketplaces. I will also provide insights into the future of AI applications and the opportunities they create for professionals looking to bridge technical expertise with business impact.

For those interested in entrepreneurship, I will share lessons learned as an angel investor in AI/ML startups, discussing key qualities of successful STEM-driven ventures and how to navigate the intersection of technology, business, and investment. Additionally, I will highlight how STEM professionals can transition into product management roles, leveraging their technical skills while balancing user-centered design and business strategy.

Finally, I will provide actionable advice for students and early-career professionals on forging unique career paths, embracing interdisciplinary opportunities, and driving impact at the intersection of academia, industry, and innovation.

Vocal communication is an essential tool for survival and human interaction. Yet, each year, hundreds of thousands of people will join the millions who have lost their ability to communicate as a result of stroke, injury or neurodegenerative diseases such as ALS. Brain-computer interfaces (BCI)—electronic devices that decode a person’s desired actions directly from neural activity, thereby bypassing the injured part of the nervous system—stand to radically improve the quality of life for individuals with lost motor, speech, and language function. However, there has yet to be a clinically viable communication BCI that can fully meet patients’ needs. Traditional BCI studies limit their subjects’ vocabulary and externally control when the subject initiates speaking. This has yielded promising results within limited conversation-like scenarios—such as fixed responses to simple questions—but are far from restoring natural conversational speech. To move the field beyond its reliance on rigid behavioral control and toward a more naturalistic, conversation-based paradigm, I’ve begun piloting dialogue-driven tasks that influence participants' responses by strategically structuring the semantic flow of their interactions. Unlike traditional BCI experiments, this question-answer paradigm eliminates the go/no-go cue and instead relies on the participant’s natural ability to recognize the end of a question, determine the correct response, provide the correct answer, and prepare for the next question. In this talk, I will introduce this novel experimental paradigm, present key findings, and discuss the impact of naturalistic approaches in advancing speech BCIs capable of fluid conversation.

There is an alarmingly high prevalence of respiratory disease in the underprivileged, minority communities residing around California’s Salton Sea. My studies aim to understand this high incidence of asthma in the region by generating in-vivo mouse exposure studies. We noted that lung inflammation induced by aerosol exposures to dust material collected from the region matched characteristics of bacterial lipopolysaccharide (LPS)/endotoxin exposure, identifying this component as the likely trigger. Additionally, we gathered data on LPS concentrations and other environmental factors in addition to performing a clinical symptom survey across the region; colocalization analysis is consistent with a model in which nutrient-driven growth of bacteria in Salton Sea leads to endotoxin entrainment in dust, promoting lung inflammation in nearby residents.

Plant-parasitic nematodes (PPNs) threaten global food production, with Meloidogyne incognita (root-knot nematodes, RKNs) being a major concern due to their ability to damage plant roots, reducing crop yields. Current control methods, including synthetic chemical fumigants, are harmful to humans and the environment. Chalcones, plant-derived precursors to flavonoids, have shown promising nematicidal effects, but their mechanism of action remains unknown. Our hypothesis suggests that chalcones inhibit a nematode protein by binding to its active site, preventing its function. Mutations in the corresponding gene may alter protein conformation, preventing chalcone binding and leading to resistance. To investigate this, our lab used Caenorhabditis elegans as a model organism due to its cost-effectiveness, ease of maintenance, and genetic similarity to RKNs. C. elegans, with its short lifespan, rapid reproduction, and genetic resources, is well-suited for genetic studies and shares biological characteristics with M. incognita, facilitating research on nematode control. Previously, a whole-genome sequencing followed by bioinformatics approach on ethyl methanesulfonate (EMS) generated resistant mutant lines from the VC2010 parental strain of C. elegans, identified potential genes linked to resistance. The current study aimed to validate candidate genes responsible for C. elegans resistance to Chalcone 17 (the candidate genes are sly-1, Y62H9A.8, Y53F4B.21, C30D11.5 and C50E10.13) and 30 (the candidate genes are clec-9, cyp-13A10, gst-5, ifb-2, and K06A9.1) using CRISPR-Cas9. Single guide RNAs (sgRNAs) were designed, editing efficiency of each guide was tested in vitro, and visualized via agarose gel electrophoresis to detect shifts in gene sequence size after digestion. Microinjection facilitated in vivo delivery of sgRNAs for gene editing. Our findings identified ifb2, an intermediate filament (IF) protein gene expressed in the intestine, as responsible for Chalcone 30 resistance. For Chalcone 17 resistance, we identified sly-1, a gene encoding a Sec1 family domain protein involved in vesicle transport from the endoplasmic reticulum to the Golgi. Understanding how chalcones affect these genes will guide future studies on the structural and functional differences between mutant and wild-type proteins, advancing the development of targeted nematode control strategies.

Biologics such as proteins, peptides, and oligonucleotides are powerful ligands to modulate challenging drug targets that lack readily “ligandable” pockets.  However, the limited membrane permeance of biologics severely restricts their intracellular applications. Moreover, different cell types may exhibit varying levels of impermeability, and some delivery vehicles might be more sensitive to this variance. Erythroid lineage cells are especially challenging to deliver cargo to because of their unique cytoskeleton and the absence of endocytosis in mature erythrocytes. We recently employed a cell permeant miniature protein to deliver bioPROTACs to human umbilical cord blood derived erythroid progenitor cells (HUDEP-2) and primary hematopoietic stem (CD34+) cells (Shen et al., ACS Cent. Sci. 2022, 8, 1695-1703). While successful, the low efficiency of delivery and lack of cell-type specificity limits use of bioPROTACs in vivo. In this work, we thoroughly evaluated the performance of various recently reported cell penetrating peptides (CPPs), CPP additives, bacterial toxins, and contractile injection systems for their ability to deliver cargo to erythroid precursor cells. We also explored how targeting receptors enriched on the erythroid cell surface might improve the efficiencies and specificities of these delivery vehicles. Our results reveal that certain vehicles exhibit improved efficiencies when directed to cell surface receptors while others do not benefit from this targeting strategy. Together, these findings advance our understanding of protein delivery to challenging cell types and illustrate some of the intricacies of cell-surface receptor targeting.

Connected networks of quantum devices are one of the most promising applications of quantum technologies. There are wide range of tasks for which these networks are well suited including distributed sensing, quantum cryptography, blind quantum computing, and election protocols. However, building such a network requires the development of new devices to quickly and efficiently generate and send quantum signals over long distances. In this talk, I will discuss the fundamental building blocks of quantum networks as well as the challenges that still need to be overcome to physically realize these networks. Through out it all, I will give a glimpse into the different materials and platforms being researched for these devices. I will particularly focus on one of the platforms researched at Caltech: rare earth ions doped into solid state hosts. 

Current enzyme function prediction tools rely on multiple sequence alignments to identify conserved residues or functional motifs within an enzyme family, making the implicit assumption that proximity in sequence space equates to proximity in functional space. However, this assumption has yet to be rigorously tested due to the time and cost required to profile multiple aspects of function across multiple enzymes using traditional biochemistry assays. To see if this fundamental assumption is true, I used a microfluidic approach, HT-MEK (High-Throughput Microfluidic Enzyme Kinetics), to express, purify and calculate kinetic and thermodynamic parameters (kcat, KM, kcat/K M, and Ki’s for multiple substrates and inhibitors) for 48 orthologs of the well-characterized PafA enzyme within the Alkaline Phosphatase superfamily. Although all 48 enzymes are highly structurally similar (0.84±0.1 LDDT score), relatively well-conserved in sequence space (41±9%), and have identical active site residues, these measurements reveal a 1000-fold difference in catalytic reactivity for a given substrate . Via additional measurements, I establish that these differences cannot be fully accounted for by differences in either stability or non-equilibrium folding of the enzyme. Measured inhibition constants for the PafA transition state analog tungstate also differ by 100-fold, further supporting the conclusion that differences in the surrounding enzyme scaffold can alter active site residue positioning even for highly similar scaffolds. These results show that sequence proximity is not the same as functional proximity and illustrate the magnitude of the challenge for efficient enzyme design.

Isogeny-based cryptography is a promising form of quantum-safe encryption because it relies on mathematical problems that are distinct from other cryptographic constructions. I will talk about why isogeny-based cryptography is quantum-safe, why we need new, distinct cryptographic constructions in a world of quantum computers, what an isogeny is and how our current work relates to trustless and secure systems in this domain.