- Ghizelle Jane E. Abarro
A Novel, Model-derived Hypothesis on the Regulation of BRUTUS Activity in Plants’ Iron (Fe) Homeostasis
Authors and Affiliations:
Ghizelle Jane E. Abarro1, Dipali Srivastava 2 , Terri A. Long 2, and Belinda S. Akpa 1, 3
1 Chemical & Biomolecular Engineering, University of Tennessee, Knoxville, TN, USA
2 Department of Plant and Microbial Biology, North Carolina State University, Raleigh, NC, USA
3 Biosciences Division, Oak Ridge National Laboratory, Oak Ridge, TN, USA
Iron (Fe) is a critical nutrient, and we depend on plants to act as our conduit of Fe from the earth. As plants assimilate Fe as they grow, they function constantly at the edge of iron deficiency. In response to iron deficiency, plants activate a complex system of signaling events to increase Fe uptake. However, Fe is a potent redox agent, and its excess causes oxidative damage. Thus, Fe deficiency responses must be regulated to ensure nutrient availability without exposing cells to a potent cytotoxin.
A key player in the deficiency response of Arabidopsis Thaliana is BRUTUS (BTS), an Fe-binding enzyme responsible for ubiquitinating proteins involved in Fe handling. BTS knock-down mutants accumulate more Fe, suggesting that BTS limits Fe uptake. However, wild-type plants respond to Fe deficiency by upregulating BTS – suggesting that BTS might be implicated in activating Fe uptake. Increased Fe has been observed to destabilize BTS – again supporting the notion that BTS should be active under deficiency and inactive when Fe is abundant. It is unclear why plants lacking a key deficiency response protein would exhibit robustness to Fe deficiency. And if BTS acts as a “brake” on Fe uptake, why upregulate BTS precisely when Fe is needed?
Herein, we present a study of the Fe-dependent activity of BTS as viewed through the lens of a mathematical model. Embedding known subcellular events and their putative Fe-dependence into a system of differential equations, we generated simulations predicting BTS accumulation in the nucleus of a cell, as a function of cytosolic Fe. We then used simulation-based inference to determine what kinetics would be consistent with known emergent behaviors of this cellular system. In doing so, we identified a new role for Fe in regulating BTS and thereby posited a non-monotonic relationship between cytoplasmic Fe and the BTS “brake”.
2. Javid Azimi Boulali
Multiscale Computational Modeling of Aortic Valve Calcification
Javid Azimi Boulali1*, Gretchen Mahler2, Bruce Murray1, Peter Huang 1
1Dept. of Mechanical Engineering, 2Dept. of Biomedical Engineering, Binghamton University, Binghamton, NY
Introduction:
Calcific aortic valve disease (CAVD) is a chronic condition that affects a significant portion of the elderly population in the United States. CAVD is characterized by the progressive remodeling of fibrotic tissue and mineralization that results in severe stenosis. This process begins with the formation of nanoscale calcific nodules that gradually progress to micro- and macro-scale nodules. Despite the prevalence of CAVD, there is currently no known treatment for valvular calcification. Patients often do not experience symptoms until the disease has advanced significantly, which can make treatment difficult. Once the disease has progressed to an advanced stage, valvular replacements are required to restore healthy physiology. The calcification process involves various cellular mechanisms, including fibrosis, endothelial to mesenchymal transition (EndMT), and differentiation of resident cells to an osteoblastic phenotype, leading to nucleation of calcium nodules and calcification. Recent studies have indicated that transforming growth factor beta (TGF) is one of the drivers of the calcification process in aortic valve disease. To simulate the tissue remodeling that occurs during disease progression, we have developed a multiscale model that incorporates both subcellular and tissue scale models. This approach provides a more comprehensive simulation of the calcification process in aortic valve disease.
Methods:
A multi-scale approach of modeling with various simulation instruments are employed. At the tissue scale, Cellular Potts Model and partial differential equations are utilized to simulate the interactions between cells and the extracellular matrix (ECM), which lead to cell migration, differentiation, and mitosis. The ECM is modeled as fiber bundles arranged randomly within a three-dimensional lattice. CompuCell3D is used as a hybrid discrete-continuum model, where the discrete model describes cell behaviors, and the continuum transport model employs diffusion equations to describe the evolution, interaction, and transport of chemical species. At the subcellular scale, ordinary differential equations are used to represent signaling pathways, such as the TGFβ pathway, which drives EndMT and calcification. RoadRunner is used to analyze this pathway by simulating the expression of proteins, including PECAM-1, α-SMA, and MMP-9.
Results, Discussion and Conclusion:
The model consists of multiple cell types, each representing an actual cell phenotype or state. Cellular differentiation is determined by protein expression analysis at the subcellular level. Simulation results demonstrate that endothelial cells lose their adhesion and undergo EndMT upon TGFβ stimulation. It also activates interstitial cells, which increases their motility, fibrosis activity, secretion of matrix degrading enzymes and TGFβ secretion. Prolonged activity of interstitial cells in the active state leads to differentiation into osteoblastic-like cells, initiating calcium nodule nucleation and growth (Figure 1). The simulation tracks cell population and differentiation and shows that the majority of active cells eventually die and become trapped in higher fibrotic or calcified areas due to the lack of nutrients. Dead cell body fragments act as feeding sites for calcified nodules to grow at higher rates. The results indicate that calcification occurs within shallower regions of the tissue where the activity of cells is higher. The size of calcified nodules formed inside the tissue is comparable to experimental data [Mendoza et al. (2022). Lab on a Chip, 22(7), 1374-1385.].
In conclusion, this multiscale model can help to unravel the aortic valve calcification process and identify important underlying causes, ultimately leading to the development of nonsurgical therapeutic methods. It can also be applicable to other complex biological systems such as cancer development.
3. Camara L Casson
Developing novel geospatial metrics to evaluate the tumor immune microenvironment
Authors: Camara L Casson, Gregory J Kimmel, Philipp M Altrock, Brandon J Manley, Meghan C Ferrall-Fairbanks
Renal cell carcinoma is the most common type of kidney cancer, with clear cell (ccRCC) making up approximately 75% of cases. Ten percent of ccRCC patients present with tumor thrombus, a condition where the tumor extends into the local vasculature and is associated with worsened prognosis. The use of immune checkpoint inhibitors has led to the expanded study of the tumor immune microenvironment (TIME), which consists of immune cells, stromal cell, cytokines, and extracellular proteins. The interactions and locations of these components have long been under-characterized until the expansion of multiplex imaging. However, the analysis methodologies have not progressed as much as the technology. Novel methods of TIME characterization are needed to understand the relationships between TIME components. To develop these novel metrics, we used digital image analysis to extract single-cell resolution data from multiplex immunohistochemistry images of at least two regions-of-interest from the tumor, stroma, and interface of tumors collected from 126 ccRCC patients. Immunohistochemistry was accessed to identify expression of CD3, CD8, FOXP3, EGFR, TBET, CD68, CD163, PDL1, CD206, and CD20. We then applied metrics from ecology and statistics to quantify cell type distributions (such as mean, median, variance) and underlying spatial relationships between cell types (such as Kolmogorov-Smirnov (KS) and Chebyshev (CBS) distances). These characteristics were then compared across 30 clinicopathological variables to assess if any of the metrics correlated with prognostic markers. Our analysis among ccRCC patients revealed that metrics from ecology and statistics could be used to quantify the differences among a theoretical (random) distribution of intercellular distances and the empirical distributions of intercellular distances of immune and tumor cells in a tumor sample. For example, in CD68+CD20 comparison, which looked at the distances between monocytes and B-cells, higher grade tumors and the presence of sarcomatoid pathology, which is indicative of a more aggressive tumor, demonstrated higher KS distance between the theoretical and empirical distribution. This suggests that the location of monocytes and B-cells in the tumor is not due to random chance, rather has a higher order structure (e.g., clustering) that may be exploited to determine who will have better response to immune-modulating therapies. Similar trends were also seen for higher grade and sarcomatoid tumors with the CBS distance. In investigating the interdistance between monocytes (CD68) and PD-L1 tumor cells, higher grade and sarcomatoid pathology tumors exhibited similar trends of increased KS and CBS distances. While the tests show that the differences between the observed distances (empirical distribution) and the theoretical random distributions are larger for higher grade tumors and sarcomatoid status, it is not yet clear if the observed distances indicate that these cells are more clustered or dispersed than expected and further analysis is looking into metric combinations to differentiate this aspect.
4. Priyojit Das
Alteration of the structural properties of the chromosome X during inactivation: a polymer simulation-based approach
In mammals, the females carry two copies of chromosome X (chrX), one of which gets transcriptionally silenced through a process known as X-chromosome inactivation (XCI). During the XCI, the transcription factors are excluded, and repressive marks are incorporated on the chrX. This leads to the formation of a spherically shaped compact inactivated chrX compared to its active counterpart. Though the molecular basis of the processes involved during the XCI has been studied extensively, the associated changes in the 3D structural organization remain elusive. Recent advances in super-resolution imaging and molecular biology techniques found that the inactive chrX is mostly devoid of structural features and it organizes into two megadomains1. However, there are plenty of open questions regarding the structural reorganization of chrX those are needed to be answered. For example, some imaging studies showed that the inactive chrX exhibits a comparable level of local compaction to the active chrX even though the inactive one is more compacted at the whole chromosome level. It is still not known what structural factors might contribute to this specific organization of inactive chrX. Intrigued by these questions, here, we decide to study the role of different structural organization mechanisms to the specific folding mechanics of the active and inactive chrX. To do that, we perform coarse-grained molecular dynamics simulation of whole chrX for both active and inactive copies in mouse neural progenitor cells (NPC). We find that the chrXi organizes in megadomains structure while chrXa exhibits a compartmental organization. We also find that the chrXi is more compact at the global scale compared to chrXa.
5. Adriana Del Pino Herrera
Population dynamics model of ovarian cancer resistance for adaptive therapy strategies
Authors: Adriana Del Pino Herrera, Meghan Ferrall-Fairbanks
Ovarian cancer is the second most common gynecological cancer accounting for 13,270 deaths in the United States in 2023. The difficulty in ovarian cancer diagnosis is due to its unnoticeable symptoms resulting in late detection and a 5-year survival rate of 50.8%. The standard of care for the disease involves neoadjuvant and adjuvant chemotherapy and surgery. However, around 80% of the patients develop resistance to platinum-based chemotherapies leading to cancer recurrence. The current treatment modality aims to kill treatment-sensitive cells allowing resistant clones to dominate in the tumor population. To combat the development of resistance, clinicians can leverage mathematical models of tumor population dynamics based on eco-evolutionary concepts to determine alternative treatment schedules and doses known as adaptive therapy. This therapeutic regime allows clinicians to control the overall size of the tumor and reduce treatment toxicities. With this application in mind, we have utilized pure population ecology concepts to analyze the growth dynamics of two ovarian cancer cell lines (A2780 and Tyk-nu) and their cisplatin-resistant counter parts (A2780cis and Tyk-nu cp.r) undergoing different continuous treatment conditions and estimated their growth rate under exponential and logistic model assumptions. Growth curves were computed for each growth model and compared by R2 values that were calculated to evaluate accuracy of model estimates. R2 values for the logistic growth curves were >0.9, therefore the logistic model explains cells growth dynamics more accurately than the exponential growth model which resulted in low R2 values (<0.6).
Assuming that cancer cells are able to access unlimited resources from their host and that only sensitive cells can convert to the resistant phenotype, the exponential growth rates found here were used to simulate a Lotka-Volterra system at different initial conditions. To identify cellular dominance, a population residual was then calculated by subtracting the resistant cell counts from the sensitive and taking the average. A positive population residual indicates higher average sensitive counts and thus an optimal tumor composition for adaptive therapy. Simulations under treatment show higher sensitive cell counts throughout a 10-day period when the initial cell concentrations are at a 90/10 ratio. This resulted in a positive population residual suggesting that at a 90/10 sensitive/resistant composition, the tumor had enough sensitive cells to still dominate and control the resistant population at a 10 µM cisplatin dose. For initial conditions of 50/50 and 10/90 cells, the population residuals were negative, indicating that the resistant population could outcompete the sensitive population.
In conclusion, we showed the population dynamics of two ovarian cancer cell lines and their resistant counterparts under different treatments and fitted their growth to population ecology models. Here, the logistic model was the best fit for the data collected in 2-D culture. The Lotka-Volterra framework simulated interactions between sensitive and resistant populations to determine potential treatment strategies.
6. Nolan English
Towards physiology and synthesis informed generative modeling in Drug Discovery
Abstract: The promise of generative AI models for drug design lies in being able to explore a vast, largely uncharted chemical space. Current estimates propose the existence of up to 10^60 chemical compounds with potential therapeutic effect. Only ~10^8 of these have ever been synthesized. Though generative models can propose and assess novel drug candidates in a high throughput manner, most candidates fail two criteria crucial for a successful drug. First: generative models frequently propose molecules that are not synthesizable. The integration of chemical knowledge into a generative model often relies on embedding “chemical language” into the model by using an autoencoder. Unfortunately, the syntax and grammar of chemistry often get lost in the encoding process. Consequently, proposed structures may violate fundamental rules of organic chemistry, rendering them invalid. Alternatively, valid molecules may lack known synthesis pathways. Second: most generative workflows are designed to optimize drug properties that do not represent human-level outcomes. This limits the likelihood that the proposed molecules will exhibit efficacy in clinical trials.
Coupling generative AI with graph-based retrosynthetic and differential equation-based human systems models could address these shortcomings. Retrosynthetic pathway models establish whether a path exists to synthesize a given molecule from known building blocks, and these models can score synthesis pathways based on important factors such as cost. Human systems models predict drug disposition in specific tissues and their potential effect on disease. While retrosynthesis and human systems models could address two key barriers to successful molecular design, these models are computationally expensive, thus they create bottlenecks in generative modeling workflows. This work shares insights from integrating retrosynthetic models and human systems models into a generative modeling framework.
7. Junhao Gu
Due to the low capture rate in single cell RNA sequencing, mRNA distributions of gene expression are often considered zero-inflated. This leads to extra uncertainty when inferring kinetic parameters from data. Here, we compute the likelihood over the biophysical parameter space and demonstrate that the identifiability of certain region of the parameter space is very low. And in some cases, little improvement in the identifiability with increasing number of cells in experiment, and in general, capture rate is the most important factor for trust-worthy inference. There are limitations in parameter inference for single cell RNA sequencing data due to technical challenges, and people should pay extra attention to the inferred parameter in the low identifiability region.
8. Charles Hodgens
Fusing knowledge from multiple scales of life: from qualitative constraints to predictions of cell behavior
Stomata are leaf pores that facilitate critical functions including gas-exchange, transpiration, and temperature regulation. Each pore, defined by two guard cells, opens in a regulated fashion that requires fusion of vacuoles. This process depends on several key proteins. The homotypic fusion and vacuole protein sorting (HOPS) complex tethers pairs of vacuoles together and chaperones membrane-associated SNARE proteins into a trans-SNARE complex. The trans-SNARE complex spans the two vacuole membranes and zippers to create the force required for fusion.
Recruitment of the HOPS complex requires the membrane lipid phosphatidylinositol-3-phosphate (PI3P). Thus, one would expect the removal of PI3P to impede fusion. Instead, in Arabidopsis thaliana, chemical depletion of PI3P from the cell triggers fusion. To resolve this apparent paradox, we have used systems modeling to establish and indicate an experimentally tractable path to validation of a novel hypothesis for the molecular events governing stoma vacuole fusion.
We leveraged qualitative, phenotypic observations to constrain the rate constants of a differential equation model and determine (i) if the model is consistent with prior observations and (ii) what kinetic regimes recapitulate our observations. Using an Approximate Bayesian Computation-Sequential Monte Carlo method, we identified regions in an eight-dimensional parameter space yielding plausible emergent dynamics. Our model findings suggest that HOPS has a third function in regulating guard cell vacuole fusion: impeding fusion activity of the trans-SNARE complex. The plausible kinetics we identified indicate that HOPS chaperones SNAREs but stalls their function until HOPS is passively or actively removed from the super-complex – meaning that stable HOPS:trans-SNARE complexes should be observed in closed stomata. After validating this prediction, we extrapolated across scales to predict how molecular signaling events impact evolving vacuole morphologies. By connecting morphological predictions to live imaging, we can iteratively integrate new observations and refine our understanding of this critical physiological function in plants.
9. Yacoub Innabi and Ju-Won Lee
All biological organizational levels comprise independent agents, such as cells in an organism, that
interact in a complex network. These individuals contribute value to the system, gaining net benefits
from its stability. Paradoxically, unchecked uncooperative agents can exploit the system for more
significant benefits. Enforcers regulating these disruptive agents are crucial for system stability. Our
study applies this conceptual framework using a computer simulation that mirrors a cancer cell’s
uncooperative nature. Utilizing R, the model investigates how interventions akin to fasting(inducing
resource scarcity) and chemotherapy (halting cell division) can re-establish system order. The simulation
captures the distinct metabolic and proliferative traits of cancer cells, offering theoretical confirmation
for the effectiveness of fasting and chemotherapy independently and synergistically. By illuminating
these dynamics, our research presents insights into cancer treatment mechanisms and extends the
biological economy metaphor to broader cooperation levels.
10. Alanda Kelly
Turning cancer cells into neurons: A role for the transcription factor INSM1
Alanda Kelly, Ankur Saxena
Department of Biological Sciences, University of Illinois Chicago
Neuroblastoma (NB) is the most common cancer in the first year of life. Finding widely effective treatments for NB remains difficult due to tumor heterogeneity and acquired resistance in some NB cells post-treatment. Interestingly, NB may be able to spontaneously resolve by differentiating into neurons. The metabolite retinoic acid (RA) is used often to resolve NB along with chemotherapy; however, some high-risk tumors are resistant to this treatment, highlighting the need to identify and develop more effective ways to differentiate NB. We previously found several factors that are necessary, but not sufficient, to promote the differentiation of NB cells in vivo, suggesting that there are other key factors driving this process. One potential contributor is the transcription factor INSM1, which we observed to have increased mRNA levels in neuron-like human NB cells compared to the levels in more progenitor-like NB cells. To determine its role in differentiating NB, we overexpressed or inhibited INSM1 in human NB SK-N-AS cells and injected them into streams of migrating neural crest cells in developing zebrafish embryos. INSM1 overexpression led to premature differentiation in vivo, while knockdown led to reduced differentiation, suggesting that INSM1 is essential for this process. Moving forward, we will determine if INSM1 works with other factors, including RA, to promote NB differentiation. Our xenotransplantation system may reveal novel mechanisms that promote the in vivo differentiation of NB and provide new targets for therapy.
11. Ryan LeFebre
Degrading a gradient: information transmission in yeast mating
Ryan LeFebre1, Joseph Landsittel1,2, David Stone3, Andrew Mugler1
1Department of Physics and Astronomy, University of Pittsburgh
2Department of Mathematics, University of Pittsburgh
3Department of Biological Sciences, University of Illinois at Chicago
Chemical gradient sensing is ubiquitous in biology. From development to migration, it plays a significant role in both multi-cellular and single-celled organisms. In haploid yeast cells of two types, the detection of chemical gradients is used to find a suitable mating partner. Each mating type secretes a pheromone that is sensed by the partner type. Paradoxically, one of the mating types also secretes an enzyme that degrades the attractant pheromone of its partner type. This degradation is vital for efficient mating. It is thought that degradation leads to a steepened gradient, but the roles of noise and information transmission are poorly understood. Can destroying part of a signal you detect increase the amount of information you receive? Using both stochastic spatiotemporal modeling and tools from information theory, we find that the answer is yes: both the signal-to-noise ratio and information transmission increase with degradation. Our work helps explain a counterintuitive signaling strategy in yeast and offers insights into optimal sensory strategies more generally.
Funding: NSF MCB-2003415
12. Keren Li
DNAcycP: A Deep Learning Attempt at Mechanical Properties of DNA
Deep learning methods have had a significant impact on both industry and academia in many fields, including the mechanical property study of DNA. DNA mechanical properties play a critical role in every aspect of DNA-dependent biological processes. Recently a high throughput assay named loop-seq has been developed to quantify the intrinsic bendability of a massive number of DNA fragments simultaneously. Using the loop-seq data, we develop a software tool, DNAcycP, for intrinsic DNA cyclizability prediction utilizing a deep learning model mixed with an Inception-ResNet structure and an LSTM layer. We demonstrate DNAcycP predicts intrinsic DNA cyclizability with high fidelity compared to the experimental data. Using an independent dataset from in vitro selection for enrichment of loopable sequences, we further verified the predicted cyclizability score, termed C-score, can well distinguish DNA fragments with different loopability. We applied DNAcycP to multiple species and compared the C-scores with available high-resolution chemical nucleosome maps. Our analyses showed that both yeast and mouse genomes share a conserved feature of high DNA bendability spanning nucleosome dyads. Additionally, we extended our analysis to transcription factor binding sites and surprisingly found that the cyclizability is substantially elevated at CTCF binding sites in the mouse genome. We further demonstrate this distinct mechanical property is conserved across mammalian species and is inherent to CTCF binding DNA motif.
13. Lynne Nacke
A pro-neurogenic role for Aβ42 in olfactory development
Lynne M. Nacke, Sriivatsan G. Rajan, and Ankur Saxena
Department of Biological Sciences, The University of Illinois Chicago
Alzheimer’s disease (AD) is a crippling neurodegenerative condition that afflicts millions of people worldwide. The loss of smell is an early potential biomarker for AD, which is perplexing given that the olfactory system is highly regenerative, with a population of basal stem cells responsible for the continuous renewal of olfactory sensory neurons (OSNs). Building on our previous developmental findings of a Notch signaling-Insm1a feedback loop that drives olfactory neurogenesis, we hypothesized that AD-associated Aβ42 peptide disrupts this signaling mechanism early in the onset of AD. To test this hypothesis, we first treated zebrafish embryos with exogenous Aβ42 and discovered temporally-dynamic changes in the number of basal stem cells and OSNs. Next, we generated and injected a transgenic construct to endogenously express Aβ42 protein. In vivo comparisons between individual Aβ42-expressing cells uncovered transcriptional changes that suggest a pro-neurogenic role for this peptide. In sum, our preliminary findings suggest that Aβ42 cell autonomously shifts the olfactory stem cell-neuron balance towards neuronal differentiation. Given that previous research on Aβ42 has focused predominantly on neurodegeneration, we hope to uncover new insights into how neurogenic mechanisms might be harnessed to improve outcomes for neurodegenerative diseases.
14. Tanya Pelayo
Temperature-Induced Stress Responses in Soybean: Unraveling Gene Expression Patterns and Regulatory Networks
Dr. Yoshie Hanzawa
Tanya Pelayo
California State University of Northridge
Soybean (Glycine max) is an economically important crop globally, serving as a
significant source of protein, oil, and livestock feed. Investigating how soybean plants respond to
environmental changes, particularly temperature shifts, is crucial for ensuring agricultural
productivity. To better understand the influence of environmental fluctuations to soybean’s
flowering and maturity, we characterized global gene expression patterns under different
photoperiod and temperature regimes in a time-series experiment using the mutant plants
carrying recessive alleles of the major maturity loci, E1, E2, and E3, and the control plants, as
well as the wild soybean Glycine soja. Our RNA-seq samples showed clear clustering patterns
for different temperatures and time points. Differential expression analyses elucidated genes that
were affected by photoperiod, temperature, genotype, and time points. Photoperiod affected
more flowering genes than other conditions did. As a next step towards clarifying soybean’s
environmental responses at the gene network level, we are exploring MAGINE (Multi-Omics
Analyzing Software) which presents a sophisticated computational framework for multi-omics
analysis in soybean research. By integrating our RNA-seq data with various genomic datasets,
MAGINE enables the comprehensive study of gene-to-gene interactions underlying soybean’s
molecular responses to diverse temperature conditions and facilitates the identification of key
regulatory genes, pathways, and molecular mechanisms involved in temperature-induced stress
tolerance and adaptation in soybean, providing valuable insights for crop improvement and
agricultural practices in the face of changing climates.
15. HaoSheng Sun
De-orphanizing neuropeptide-GPCR interaction using structural analysis and machine learning techniques?
Neuropeptides play important roles in neuronal communication, and can signal either via diffusion or via long-range signaling as circulating hormones to activate downstream GPCRs. Across post-embryonic animal development, there are pervasive and dynamic changes in neuropeptide expression. This suggests that altering the repertoire of neuromodulatory peptides could be a conserved maturation mechanism in the animal kingdom that defines behavioral state transitions across development. Taking advantage of the compact nervous system of the nematode C. elegans, we are currently characterizing the dynamic changes in neuropeptide and their cognate GPCR expression across development of the entire nervous system in single neuron resolution, and establishing neuropeptidergic connectome maps across development. Recently, system-wide in vitro reverse pharmacology has deorphanized many of the peptide-GPCR pairs. With the advance of putative GPCR protein structure by AlphaFold, I am seeking structural and computational modeling collaborators that can take advantage of these putative AlphaFold protein structures to model their interactions with peptide ligands. My hope is that we can take advantage of many experimentally validated peptide-GPCR pairs, and use machine learning approaches to deorphanize the rest of the interactions. With the C. elegans system, we can very easily test the computationally-predicted neuropeptide-GPCR pairs. This will help us understand whether they are generalized logic to neuropeptide-GPCR pairing/interaction, and help us to better understand the role of neuropepidergic system in neuronal maturation.
16. Ghocho Terasaki
Merging Traditional Scientific Computing with Data Science to Develop a New Prediction Engine for Brain Cancer
Glioblastoma multiforme (GBM) is one of the fastest-growing brain tumors and it has very low survival rates. Mathematical modeling can be used to predict the growth and treatment of brain cancer. However, one of the difficulties lies in the ability to estimate patient-specific parameters in the mathematical model from magnetic resonance imaging (MRI) data. We constructed a numerical solver to simulate tumor growth over a realistic 3D brain geometry derived from segmented-MRI. Then, using information about the size of the different glioma sub-regions, we are developing a method that estimates the patient-specific model parameters to inform the forward simulation. Ultimately, we hope to predict the overall survival of a patient from a single pre- operative scan.
17. Lingyun (Ivy) Xiong
Are physiological oscillations physiological?
Most, if not all, of the body’s systems are oscillatory. Despite widespread and striking examples of these physiological oscillations, their functional role is often unclear. Even glycolysis, the paradigm example of oscillatory biochemistry, has seen questions about its function. Here, we take a systems approach to summarize evidence that oscillation plays critical physiological roles. Oscillatory behavior enables systems to avoid desensitization, to avoid chronically high and therefore toxic levels of chemicals, and to become more resistant to noise. Oscillation also enables complex physiological systems to reconcile incompatible conditions such as oxidation and reduction, by cycling between them, and to synchronize the oscillations of many small units into one large effect. In pancreatic beta cells, we show that glycolytic oscillations are in synchrony with calcium and mitochondrial oscillations to drive pulsatile insulin release, which is pivotal for the liver to regulate blood glucose dynamics. In addition, oscillation can keep biological time, essential for embryonic development in promoting cell diversity and pattern formation. The functional importance of oscillatory processes requires a re-thinking of the traditional doctrine of homeostasis, holding that physiological quantities are maintained at constant equilibrium values, which has largely failed us in the clinic. A more dynamic approach will enable us to view health and disease through a new light and initiate a paradigm shift in treating diseases, including depression and cancer. This modern synthesis also takes a deeper look into the mechanisms that create and sustain oscillatory processes, which requires the language of nonlinear dynamics, well beyond the linearization techniques of equilibrium control theory.
