SCMB Final Symposium
April 10 – 11, 2025
Titles and Abstracts
Natasha Jonoska
University of South Florida (SCMB Senior Personnel)
&
Francesca Storici
Georgia Tech (SCMB Senior Personnel)
“RNA-mediated double-strand break repair by end-joining mechanisms”
Abstract: Double-strand breaks (DSBs) in DNA are challenging to repair. Cells employ at least three DSB-repair mechanisms, with a preference for non-homologous end joining (NHEJ) over homologous recombination (HR) and microhomology-mediated end joining (MMEJ). While most eukaryotic DNA is transcribed into RNA, providing complementary genetic information, much remains unknown about the direct impact of RNA on DSB-repair outcomes and its role in DSB-repair via end joining. Here, we show that both sense and antisense-transcript RNAs impact DSB repair in a sequence-specific manner in wild-type human and yeast cells. Depending on its sequence complementarity with the broken DNA ends, a transcript RNA can promote repair of a DSB or a double-strand gap in its DNA gene via NHEJ or MMEJ, independently from DNA synthesis. The results demonstrate a role of transcript RNA in directing the way DSBs are repaired in DNA, suggesting that RNA may directly modulate genome stability and evolution.
Svetlana Poznanovik
Clemson University (Guest Speaker)
“Predicting R-loop formation with a formal grammar”
Abstract: R-loops, transient three-stranded nucleic acids, emerge during transcription when newly formed RNA binds with the template DNA, releasing the coding strand of DNA. Not much is known about the formation process and the three-dimensional structure of R-loops. In this study, we represent an R-loop as a term in a formal grammar and utilize this grammatical framework to predict R-loop formation. Our model is trained using experimental data from SMRF-seq. Although R-loop formation is influenced by both DNA sequence and topology, our grammar, which does not include explicit topological details, accurately predicts R-loop formation for plasmids with varying starting topologies.
Hector Banos
Cal State University, San Bernardino (former SCMB Junior Researcher)
“How Much Is Transcription-associated Mutagenesis Driving tRNA Microevolution?”
Abstract: Transfer RNAs (tRNAs) are highly conserved and frequently transcribed. Recent studies have identified transcription-associated mutagenesis (TAM) as a key driver of sequence variation around tRNA loci. However, TAM's impact on tRNA remains unclear, as its mutational patterns often preserve structure, making it difficult to separate TAM effects from the strong selection pressure to maintain structural integrity.
To investigate this, we analyze tRNA variation in Caenorhabditis elegans and introduce a TAM-driven microevolution model (incorporating TAM-specific mutational biases within a continuous Markov framework). Our TAM-biased model outperforms standard models and aligns with the observed secondary structure characteristics. These findings highlight TAM’s significant role in shaping tRNA allelic variation within populations.
Bill Kath
Northwestern University (Guest Speaker)
“Opportunities at the NITMB”
Abstract: Are you curious about the new NSF-Simons National Institute for Theory and Mathematics in Biology (NITMB) in Chicago? This informational talk will cover the ways you can be part of the institute, including external research projects, scientific workshops, affiliate memberships, the visiting scholars program, and more.
Scott McKinley
Tulane University (SCMB Senior Personnel)
“Robust inference and model selection for particle tracking in live cells”
Abstract: There is now an expansive collection of mathematical work on building models for the transport of intracellular cargo by molecular motors. Commonly studied cargo undergo “saltatory” motion (bidirectional ballistic motion, intermixed with periods of stationarity) along often unobserved microtubules. Traditionally microparticle transport is quantified in terms of mean-squared displacement, but this ubiquitous statistic averages over periods of motion and pauses, eliminating important biophysical information. In this talk, I will discuss our group’s approach to segmentation analysis: an in-house changepoint detection algorithm coupled with a focus on summary statistics that are robust with respect to the inevitable mistakes that changepoint detections algorithms make.
Catera Wilder
University of California, San Francisco (Guest Speaker)
“Specificity of cellular responses through dynamic control of signaling networks”
Abstract: Interferon (IFN)-mediated immunity is a vital component of the innate immune response initiated during pathogenic exposure. Type I IFNs induce powerful anti-viral responses via the activation of IFN stimulated genes (ISGs) by the transcription factor, IFN-stimulated gene factor (ISGF3). In some pathological contexts type I IFNs are responsible for exacerbating inflammation. While the IFN-mediated cell-intrinsic anti-viral response is well known, it is less clear how IFNs may contribute to the cell-extrinsic inflammatory response. We hypothesized that IFN type-specific functions may be the result of IFN type-specific control of ISGF3 dynamics; hence we sought to develop a quantitative understanding of the mechanisms that control ISGF3 activity.
Using quantitative biochemical assays and next generation sequencing, we found that a high dose of IFN-β activates an anti-viral and inflammatory gene expression program in contrast to IFN-λ3, a type III IFN, which elicits only the common anti-viral gene program. We show that the inflammatory gene program depends on a second, potentiated phase in ISGF3 activation. Iterating between mathematical modeling and experimental analysis we show that the ISGF3 activation network may engage a positive feedback loop with its subunits IRF9 and STAT2. This network motif mediates stimulus-specific ISGF3 dynamics that are dependent on ligand, dose, and duration of exposure, and when engaged activates the inflammatory gene expression program.
Our results reveal a previously underappreciated dynamical control of the JAK-STAT/IRF signaling network that may produce distinct biological responses, and suggest that studies of type I IFN dysregulation, and in turn therapeutic remedies, may focus on feedback regulators within it.
Alex Ruys de Perez
Cal Poly, San Luis Obispo (Former SCMB Junior Researcher)
&
Eunbi Park
National Institutes of Health (Former SCMB Junior Researcher)
"Topological data analysis of human induced pluripotent stem cell colonies: Pattern formation and tissue fate prediction".
Abstract: The organizational structure of multicellular bodies plays a crucial role in determining cell fate decisions and differentiation during the developmental phase. However, this structure is often difficult to analyze, due to it being characterized by qualitative properties like shape or connectedness. To this problem we apply topological data analysis, a method that can be thought of as taking these qualitative properties and turning them into quantitative figures. We show in two separate examples the usefulness of this approach. First, we demonstrate how neural networks can use topological data to identify stem cell colonies given separate differentiation protocol treatments. Second, we describe a pipeline that processes topological data in microscopy images to find pattern formations. We also show a specific achievement of this pipeline, in its identification of the loss of pluripotency in certain stem cell colonies."
Julie Mitchell
University of Wisconsin (SCMB Senior Personnel)
&
Matt Torres
Georgia Tech (SCMB Senior Personnel)
“Accessing Phosphorylation-Induced Conditional Protein Folds Through Generative Machine Learning”
Abstract: We will discuss our ongoing progress in designing peptide sequences able to undergo disorder to order transition upon phosphorylation. Our approach employs generative artificial intelligence to produce sequences with the desired properties. Unlike the protein folding problem, for which there is ample structural data for training, no such data exist for the question we have posed. We have pursued a strategy that utilizes synthetic data, wherein the results of molecular dynamics simulation is aligned to NMR data for specific examples. We also utilize descriptors in our approach (in contrast to a large language type model), from which insights are possible into the biophysical properties driving conformational transformation.
Elizabeth Munch
Michigan State University (Guest Speaker)
“Measuring Shape with the Directional Transform and Topological Descriptors”
Abstract: Quantifying the shape of complex structures is a fundamental challenge, particularly in fields like plant morphology, where traditional methods rely on manual measurements with calipers and scales. These approaches, while effective, often reduce shape to simplistic attributes such as length, width, and weight. The advent of high-resolution imaging and 3D scanning provides richer data, but extracting meaningful, computable shape representations remains an open problem. In this talk, we explore techniques from Topological Data Analysis (TDA) that enable robust shape encoding by combining the directional transform with topological descriptors such as the Euler characteristic, persistent homology, and merge trees. This framework offers complete representational information, efficient computation, and adaptable comparison metrics. We demonstrate the power of these tools in plant morphology, where they facilitate machine learning models for automated phenotypic classification.
Johnathan Bush
James Madison University (former SCMB Junior Researcher)
“Topological Feature Selection for Time Series”
Abstract: I will describe how tools from applied topology may be used to identify components of time series most responsible for cyclic dynamics observed in orbits of an underlying dynamical system. In this setting, we will consider C. elegans neuronal data collected by Dr. Lu’s lab and use topology to identify subsets of neurons driving global cyclic brain dynamics in the spirit of dimensionality reduction.
Sihoon Moon
Georgia Tech (SCMB Junior Researcher)
“Topological feature selection for single-cell resolved whole-brain neural activity timeseries”
Abstract: Generating complex behaviors requires coordination of neural activity at many scales, from individual neurons to brain-wide. This has motivated the development of large-scale neural activity recording methods, such as whole-brain imaging in C. elegans, where all head neurons are recorded in parallel with single cell resolution. While the neural network of C. elegans is extremely compact (only 302 neurons, compared to ~86 billion in humans), it can generate complex behaviors like foraging and learning. While individual neurons and circuits underlying complex behaviors have been identified, the role of whole-brain dynamics in processing sensory information to generate these behaviors is less well understood. To investigate brain-wide recruitment of neural activity by sensory stimuli we have combined advances in whole-brain imaging, microfluidic devices, and computer vision. In the spirit of dimensionality reduction, we have adapted tools from applied topology to identify subsets of neurons driving global cyclic brain dynamics in whole-brain imaging datasets. These tools from applied topology can be generalized more broadly to identify components of time series most responsible for cyclic dynamics observed in orbits of an underlying dynamical system.
Michael Lavigne
North Carolina School of Science and Math
(former SCMB Communication, Education and Outreach Asst. Director)
“Priming the Pipeline: Education, Outreach, and Interactional Expertise at SCMB”
Abstract: SCMB’s mission to advance science at the interface of mathematics and biology demanded more than just the incubation of research collaborations. It demanded that we learn to talk to one another. The cross-disciplinary relationships in SCMB’s seven Seed Projects sought to explicitly train early-career researchers in Interactional Expertise—a constellation of skills, both communicative and scientific, needed to productively engage with practitioners of another discipline. It worked. Outside of the Seed Projects, SCMB’s outreach mission aimed to “prime the pipeline” for future generations of researchers to take interest in the immense potential of mathematics to address bio-systems challenges. To this end, the Center piloted summer programs, workshops, and courses that have reached students from 9th grade to grad school. This is the story of what we tried, what we learned, and where we go from here.
Jennifer Rieser
Emory University (Guest Speaker)
“Impressive feats without feet: the multi-scale physics of snake climbing”
Abstract: Some snake species can navigate diverse arboreal habitats with ease, propelling their limbless bodies over tree trunks and branches of varying diameter, orientation, flexibility, and surface roughness. However, other species struggle or even fail to climb. We take a multi-faceted approach and explore behavioral strategies that lead to success as well as skin adaptations that may enhance frictional interactions. For behavioral studies, we investigate the movement strategies for both frequent (corn snakes) and occasional (kingsnakes) climbers on our smooth vertical climbing wall, from which evenly spaced force-sensing protrusions and serve as the only “footholds”. Combining time-resolved 2D force data with 3D kinematics for vertical ascents and descents, we analyzed how body shapes and distributions of applied forces vary with the length and spacing of surface features. We found that corn snakes alter their behavior as conditions become more difficult (e.g., shorter or fewer footholds), but they were successful in a wide range of conditions. Inspired by these results, we built a simple robotic model and found that, when it was programmed to mimic a simple corn snake strategy observed in our experiments, it climbed easily. Surprisingly, our robot outperformed the kingsnakes, which climbed poorly in all conditions. To allow for broad parameter variation, we developed a mathematical model to compare the diverse strategies in the context of trade-offs between stability and speed. Examining small-scale skin adaptations, our lab has recently found that the structures present on the surfaces of snakeskin are unchanged by the preservation process that specimens undergo when added to a museum collection, allowing us much broader access to samples across the phylogenetic tree. Using this result, we examine the skin texture and textural variation in the context of potential adaptations for different forms of movement. We observed longer spike-like features in preliminary AFM scans of the ventral scales of arboreal snakes, which we hypothesize may enhance lateral friction.
Daniel Irvine
Kennesaw State University (former SCMB Junior Researcher)
“Neuromechanical Dispersion for Undulatory Swimming".
Abstract: In this talk I will discuss how geometric mechanics can be applied to study animal locomotion. I will focus, specifically on the nematode C. elegans, which moves itself by undulating. The main result of this work is identifying a relationship between the temporal frequency and the spatial wavenumber of swimming worms. This relationship holds for organisms in diverse rheologies, and we have chosen to call this a neuromechanical dispersion relationship. I will give experimental evidence for this relationship in the species of nematode Caenorhabditis elegans. A model treating the organisms as actively driven viscoelastic beams reproduces the experimentally observed scaling.
Baxi Zhong
Georgia Tech (SCMB Junior Researcher)
“Embodied Intelligence in Highly-capable Mesoscale Robots”
Abstract: Mesoscale robots (1-10 cm in height, 0.01-2 kg in weight) offer unique advantages over their macro and micro counterparts, particularly their ability to navigate confined spaces while maintaining high mobility. Despite their potential, mesoscale robotics remains an underexplored area of research. In the first part of my talk, I will outline the challenges associated with mesoscale locomotion, such as terrain uncertainty and complex dynamics. To address these challenges, I will introduce the paradigms I developed in my prior work to study embodied intelligence (EI) at the mesoscale. Specifically, I will discuss how mechanical intelligence (MI) emerges from passive dynamics in reptile- and centipede-like robots, enabling novel capabilities such as guaranteed robustness, sand-swimming, and obstacle-aided locomotion. I will also present my research vision to integrate computational intelligence with MI, inspired by theories and algorithms from signal transmission, to further advance EI in mesoscale robots. In the second part of my talk, I will explore a co-design problem: how variations in morphology can enhance MI for specialized tasks such as burrowing, load-carrying, and climbing. Using tools from geometric mechanics, I will demonstrate how these insights can inform robot design. I will conclude by discussing the broader implications of mesoscale robotics, including advancing our understanding of evolutionary biology and enabling modular, reconfigurable robot systems for diverse real-world applications.
