News: People

12/01/2024

Trainee Highlight – Samar Alqatari

Samar is a grad student in the Nagel group. She received her bachelor’s from Stanford, then completed a master’s in computational science at MIT, where she performed experiments and simulations on the Rayleigh-Taylor instability under confinement. At UChicago, she has expanded her work on fluid instabilities into the Saffman-Taylor viscous fingering that is observed under applied rotational shear. More recently, she has been studying function and adaptation using the framework of evolvable mechanical networks. Samar’s current work is centered on the evolution of strain in elastic mechanical networks that are tuned via selective pruning to exhibit an “allosteric” function: applying a strain to a pair of “source” nodes induces a strain at a pair of “target” nodes, analogous to allostery in proteins. The networks are tuned for either an in-phase or an out-of-phase strain response between the source and target. The network can be evolved from one function to a another with a single-step mutation by adding or removing a link. The statistics of this evolving strain behavior show similarities to biological evolution, such as epistatic effects, i.e. the result of each mutation is dependent on the nature of previous mutations. This analysis has led to the identification of several systematic features of evolution. To make the most of this system, she hopes to collaborate with biologists, biophysicists, and anyone interested in evolution and the emergence of function in matter.

12/01/2024

Trainee Highlight – Nico Romeo

Nico is a CLS fellow working with Vincenzo Vitelli, Elizabeth Jerison, Aaron Dinner and Noah Mitchell. He obtained his PhD at MIT, where he researched the mechanics of biological development and methods for its representation . His work at Chicago focuses on developing methods to study the complex dynamics of innate immune responses and biological development, drawing on tools from statistical physics, geometry, and machine learning. In his current project, he and his collaborators have developed a contrastive learning technique to identify dynamical behaviors from trajectory data and characterize systems beyond topological features, using transition path theory. These techniques allow them to draw a ‘phase diagram’ of dynamical systems and reveal structure in excitable systems. As he continues this work, Nico would love to learn about simple systems that exhibit complex spatial dynamics.

12/01/2024

Congratulations Jasmine Nirody!

Please join us in congratulating Jasmine Nirody on being elected as a Member-at-Large in the Division of Biological Physics in the American Physical Society! She will take office at this upcoming APS Global Physics Summit in March 2025. 

11/01/2024

Welcome CLS Postdoctoral Fellow Sang Hyun Choi

Sang Hyun is currently working on exploring phase transitions in a XY model where two species of spins interact non-reciprocally, exploring how dynamics are affected by non-reciprocity, space, and noise.

10/01/2024

Welcome Noah Mitchell!

Please welcome Noah Mitchell as a CLS Investigator! Previously an affiliate, he will be working with the MA2 team on encoding information in morphogenetic systems. Noah’s lab is a part of the Department of Molecular Genetics and their current research focus is collective cell behaviors and mechanical interactions between mesodermal and epithelial tissue layers that link genes to geometry.

10/01/2024

Congratulations Stephanie Palmer!

Please join us in congratulating Stephanie Palmer as one of the new Schmidt Science Polymaths! This program offers support to researchers looking to explore different focuses with their work. Stephanie has several projects she is interested in pursuing including, examine the evolution of neural computation in color vision for butterflies and the circadian rhythms of bacteria, specifically how their environment affects computation and migration. Read the full article here.

10/01/2024

Trainee Highlight – Cal Floyd 

Cal is a postdoc in the Vaikuntanathan and Dinner groups who obtained his Ph.D. working with Garyk Papoian and Chris Jarzynski at the University of Maryland, where he researched the non-equilibrium thermodynamics of the cytoskeleton. At Chicago, he is studying cellular computation and control, using simulation and theory to analyze the computational functions of biochemical processes. He recently collaborated with Suri Vaikuntanathan, Arvind Murugan, and Aaron Dinner to investigate non-equilibrium biophysical systems that perform a classification function, with the goal of examining the physical limitations on the systems’ computational expressivity. The target systems were modeled as discrete-state Markov processes, and the systems’ steady state distributions effectively partitioned their input space into distinct computational outputs. Through analytical derivations and supporting numerical work, he showed that these effective input-output functions have specific inherent limitations that prevent them from classifying arbitrarily complicated input distributions.  However, he also demonstrated that these limitations can be systematically overcome.  More complex computations can be achieved by feeding inputs into the system through multiple paths.  Further analysis suggests that the mathematical structure underlying these biophysical computation mechanisms resembles the transformer architectures used in machine learning, such as those found in systems like ChatGPT.  As these findings are fairly general, they have the potential to provide insight into a range of biological processes, such as how cells encode their state through membrane glycoproteins, pathogen recognition in immune responses, and quorum sensing by bacterial populations.

10/01/2024

Trainee Highlight – Kyle Crocker

Kyle is a postdoc in the Kuehn lab. His doctoral research at Ohio State University centered on building thermodynamic models of DNA mismatch repair and DNA origami. At the University of Chicago, he is studying microbial ecology by combining computational modeling, theory, and experimentation to uncover and understand collective metabolic behavior in the soil microbiome. Since microbial communities typically harbor tens to hundreds of distinct species, microbiome dynamics are potentially highly complex, yet understanding how these populations respond to change is critical for predicting the impact of perturbations, interventions, and climate change. Even though the taxonomic diversity of the microbiome is high, its taxa can often be sorted into functional guilds made of strains with similar metabolic traits. These guilds reduce the complexity of the system and promise to simplify our understanding of how communities respond to environmental change. Microbial communities experience fluctuation on a range of timescales, from rapid changes in moisture, temperature, or light levels to long-term seasonal or climactic variation. In the case of nutrient fluctuations, the Kuehn lab have found that the nature of the response depends on the timescale of the change. Rapid changes in nutrient levels bring about correlated fluctuations in the population of each guild (dynamic cohesion), yielding an overall community response that has its dimensionality set by the number of guilds. In contrast, at slower timescales of nutrient variation the species within a guild begin to compete due to similar nutrient preferences. This competition leads to a community response that is not dominated by guild structure. These results provide a route to understanding the relationship between functional guilds and the community response, as well as an experimental approach to identifying functional guilds via designed nutrient perturbations to complex communities. To further this work, Kyle is looking for collaborators knowledgeable in representation learning, soil biogeochemistry, and the Birch effect. 

10/01/2024

Trainee Highlight – Emily Hinds 

Emily is a graduate student in the Ranganathan group who graduated with a B.S. in biochemistry and computer science from the University of Wisconsin, Madison, where she worked on the laboratory evolution of proteins for structure prediction. At UChicago, she continues to study protein evolution and design by building statistical models mapping protein sequence to function, using these models to design new proteins, and testing the activity of the designs with high-throughput experiment. She is interested in other (real or toy) systems that explore questions of dimensionality and connectivity in epistatic networks. Proteins which evolved from a common ancestor to perform the same cellular function can differ in their linear sequence. by more than 75%. A group of related proteins evolves upon a network limited by mutation and natural selection; all sequences in the network must retain molecular function, and they are each connected to the rest of the network via paths of single mutations. The effect of these constraints on the geometry of sequence spaces—their size, dimensionality, and shape, and the extent of connectivity—is not well characterized for real proteins. To better understand these networks, we predicted new proteins within a specified family by building statistical (Potts) models that collectively reproduce first-order sequence statistics and a targeted subset of second-order statistics for that protein family. These models were used to design functional proteins through an evolution-like sampling process. The designs were then verified through high-throughput growth experiments in E. coli. For the chorismate mutase (AroQ) enzyme family, models built using only 0.4% of the pairwise correlations suffice to design functional synthetic proteins, some up to 51% different than any natural protein. Currently, Emily is investigating the properties of this sparse model in silico. This work will broaden the novelty and diversity of proteins designed by evolution-based statistical methods and provide strategies to analyze natural (evolved) protein sequence spaces.

08/26/2024

Welcome CLS Postdoctoral Fellow Mason Rouches

Mason received his PhD in Molecular Biophysics and Biochemistry from Yale University and comes with experience working on phase separation phenomena in biology, especially the phenomenon of pre-wetting. Here at the Center for Living Systems, he will explore information processing and computation through phase separation in multi-component systems with an eye towards condensates that control gene regulation in the nucleus.