Andrew Pyo

All Publications

• Evolutionary Benefits of Fitness-Dependent Mutation Rates PHYSICAL REVIEW LETTERS Pyo, A. G. T., Yu, Q., Merkenschlager, J., Wingreen, N. S. 2025; 135 (4): 048402

  Abstract

  Motivated by recent observations that mutation rates can be correlated with individual fitness, we analyze an evolutionary hill-climbing model with fitness-dependent mutation rates. Our results show that a mutation rate that decreases with increasing relative fitness can greatly accelerate the accumulation of beneficial mutations. Moreover, we show that a lower mutation rate for fitter individuals can prevent "mutational meltdown" of small populations by decreasing the probability of fixation of deleterious mutations. These findings suggest potential strategies for accelerating the adaptation of populations to environmental changes.

  View details for DOI 10.1103/szpk-sx1c

  View details for Web of Science ID 001538755800007

  View details for PubMedID 40794095

• Membrane wetting by biomolecular condensates is facilitated by mobile tethers. bioRxiv : the preprint server for biology GrandPre, T., Yu, Q., Pyo, A. G., Kosmrlj, A., Wingreen, N. S. 2025

  Abstract

  Biomolecular condensates frequently rely on membrane interactions for localization, recruitment, and chemical substrates. These interactions are often mediated by membrane-anchored mobile tethers, a feature overlooked by traditional wetting models. Here, we propose a general theoretical framework to study how mobile tethers impact both equilibrium and dynamic properties of condensate wetting. We show that a favorable tether-condensate interaction leads to tether enrichment at the condensate-membrane interface, which reduces the surface tension with the membrane and modifies the equilibrium contact angle. Increasing tether abundance on the membrane can drive transitions between wetting regimes, with only a modest binding energy required for biologically relevant scenarios. Furthermore, by helping condensates coat membranes, mobile tethers can facilitate condensate localization to junctions of membrane structures, such as the reticulated membranes inside the algal pyrenoid. These results provide a framework to study the implications of tether-mediated condensate-membrane interactions for cellular organization and function.

  View details for DOI 10.1101/2024.12.04.626804

  View details for PubMedID 39677715

• Regulated somatic hypermutation enhances antibody affinity maturation NATURE Merkenschlager, J., Pyo, A. G. T., Silva Santos, G. S., Schaefer-Babajew, D., Cipolla, M., Hartweger, H., Gitlin, A. D., Wingreen, N. S., Nussenzweig, M. C. 2025; 641 (8062): 495-502

  Abstract

  Germinal centres are specialized microenvironments where B cells undergo affinity maturation. B cells expressing antibodies whose affinity is improved by somatic hypermutation are selected for expansion by limiting numbers of T follicular helper cells. Cell division is accompanied by mutation of the immunoglobulin genes, at what is believed to be a fixed rate of around 1 × 10-3 per base pair per cell division1. As mutagenesis is random, the probability of acquiring deleterious mutations outweighs the probability of acquiring affinity-enhancing mutations. This effect might be heightened, and even become counterproductive, in B cells that express high-affinity antibodies and undergo the greatest number of cell divisions2. Here we experimentally examine a theoretical model that explains how affinity maturation could be optimized by varying the rate of somatic hypermutation such that cells that express higher-affinity antibodies divide more but mutate less per division. Data obtained from mice immunized with SARS-CoV-2 vaccines or a model antigen align with the theoretical model and show that cells producing high-affinity antibodies shorten the G0/G1 phases of the cell cycle and reduce their mutation rates. We propose that these mechanisms safeguard high-affinity B cell lineages and enhance the outcomes of antibody affinity maturation.

  View details for DOI 10.1038/s41586-025-08728-2

  View details for Web of Science ID 001447838200001

  View details for PubMedID 40108475

  View details for PubMedCentralID PMC12058521

• Contrastive learning of T cell receptor representations CELL SYSTEMS Nagano, Y., Pyo, A. G. T., Milighetti, M., Henderson, J., Shawe-Taylor, J., Chain, B., Tiffeau-Mayer, A. 2025; 16 (1): 101165

  Abstract

  Computational prediction of the interaction of T cell receptors (TCRs) and their ligands is a grand challenge in immunology. Despite advances in high-throughput assays, specificity-labeled TCR data remain sparse. In other domains, the pre-training of language models on unlabeled data has been successfully used to address data bottlenecks. However, it is unclear how to best pre-train protein language models for TCR specificity prediction. Here, we introduce a TCR language model called SCEPTR (simple contrastive embedding of the primary sequence of T cell receptors), which is capable of data-efficient transfer learning. Through our model, we introduce a pre-training strategy combining autocontrastive learning and masked-language modeling, which enables SCEPTR to achieve its state-of-the-art performance. In contrast, existing protein language models and a variant of SCEPTR pre-trained without autocontrastive learning are outperformed by sequence alignment-based methods. We anticipate that contrastive learning will be a useful paradigm to decode the rules of TCR specificity. A record of this paper's transparent peer review process is included in the supplemental information.

  View details for DOI 10.1016/j.cels.2024.12.006

  View details for Web of Science ID 001422685200001

  View details for PubMedID 39778580

• The exchange dynamics of biomolecular condensates ELIFE Zhang, Y., Pyo, A. G. T., Kliegman, R., Jiang, Y., Brangwynne, C. P., Stone, H. A., Wingreen, N. S. 2024; 12

  Abstract

  A hallmark of biomolecular condensates formed via liquid-liquid phase separation is that they dynamically exchange material with their surroundings, and this process can be crucial to condensate function. Intuitively, the rate of exchange can be limited by the flux from the dilute phase or by the mixing speed in the dense phase. Surprisingly, a recent experiment suggests that exchange can also be limited by the dynamics at the droplet interface, implying the existence of an 'interface resistance'. Here, we first derive an analytical expression for the timescale of condensate material exchange, which clearly conveys the physical factors controlling exchange dynamics. We then utilize sticker-spacer polymer models to show that interface resistance can arise when incident molecules transiently touch the interface without entering the dense phase, i.e., the molecules 'bounce' from the interface. Our work provides insight into condensate exchange dynamics, with implications for both natural and synthetic systems.

  View details for DOI 10.7554/eLife.91680

  View details for Web of Science ID 001320601500001

  View details for PubMedID 39320949

  View details for PubMedCentralID PMC11424094

• Effects of linker length on phase separation: lessons from the Rubisco-EPYC1 system of the algal pyrenoid. bioRxiv : the preprint server for biology GrandPre, T., Zhang, Y., Pyo, A. G., Weiner, B., Li, J., Jonikas, M. C., Wingreen, N. S. 2023

  Abstract

  Biomolecular condensates are membraneless organelles formed via phase separation of macromolecules, typically consisting of bond-forming "stickers" connected by flexible "linkers". Linkers have diverse roles, such as occupying space and facilitating interactions. To understand how linker length relative to other lengths affects condensation, we focus on the pyrenoid, which enhances photosynthesis in green algae. Specifically, we apply coarse-grained simulations and analytical theory to the pyrenoid proteins of Chlamydomonas reinhardtii: the rigid holoenzyme Rubisco and its flexible partner EPYC1. Remarkably, halving EPYC1 linker lengths decreases critical concentrations by ten-fold. We attribute this difference to the molecular "fit" between EPYC1 and Rubisco. Varying Rubisco sticker locations reveals that the native sites yield the poorest fit, thus optimizing phase separation. Surprisingly, shorter linkers mediate a transition to a gas of rods as Rubisco stickers approach the poles. These findings illustrate how intrinsically disordered proteins affect phase separation through the interplay of molecular length scales.

  View details for DOI 10.1101/2023.06.11.544494

  View details for PubMedID 37333342

• Proximity to criticality predicts surface properties of biomolecular condensates PROCEEDINGS OF THE NATIONAL ACADEMY OF SCIENCES OF THE UNITED STATES OF AMERICA Pyo, A. G. T., Zhang, Y., Wingreen, N. S. 2023; 120 (23): e2220014120

  Abstract

  It has recently become appreciated that cells self-organize their interiors through the formation of biomolecular condensates. These condensates, typically formed through liquid-liquid phase separation of proteins, nucleic acids, and other biopolymers, exhibit reversible assembly/disassembly in response to changing conditions. Condensates play many functional roles, aiding in biochemical reactions, signal transduction, and sequestration of certain components. Ultimately, these functions depend on the physical properties of condensates, which are encoded in the microscopic features of the constituent biomolecules. In general, the mapping from microscopic features to macroscopic properties is complex, but it is known that near a critical point, macroscopic properties follow power laws with only a small number of parameters, making it easier to identify underlying principles. How far does this critical region extend for biomolecular condensates and what principles govern condensate properties in the critical regime? Using coarse-grained molecular-dynamics simulations of a representative class of biomolecular condensates, we found that the critical regime can be wide enough to cover the full physiological range of temperatures. Within this critical regime, we identified that polymer sequence influences surface tension predominately via shifting the critical temperature. Finally, we show that condensate surface tension over a wide range of temperatures can be calculated from the critical temperature and a single measurement of the interface width.

  View details for DOI 10.1073/pnas.2220014120

  View details for Web of Science ID 001038060100005

  View details for PubMedID 37252985

  View details for PubMedCentralID PMC10266063

• Surface tension and super-stoichiometric surface enrichment in two-component biomolecular condensates ISCIENCE Pyo, A. G. T., Zhang, Y., Wingreen, N. S. 2022; 25 (2): 103852

  Abstract

  Cells can achieve internal organization by exploiting liquid-liquid phase separation to form biomolecular condensates. Here we focus on the surface properties of condensates composed of two multivalent associative polymers held together by one-to-one "sticker" bonds. Using coarse-grained molecular-dynamics simulations, we study the influence of component stoichiometry on condensate surface properties. We find that unequal stoichiometry results in enrichment of the majority species at the interface and a sharp reduction of surface tension. To relate these two effects, we show that the reduction in surface tension scales linearly with the excess concentration of free binding sites at the interface. Our results imply that each excess free site contributes an approximately fixed additional energy and entropy to the interface, with the latter dominating such that enrichment of free majority sites lowers the surface tension. Our work provides insight into novel physical mechanisms by which cells can regulate condensate surface properties.

  View details for DOI 10.1016/j.isci.2022.103852

  View details for Web of Science ID 000760353900001

  View details for PubMedID 35198903

  View details for PubMedCentralID PMC8851291

• Motif-pattern dependence of biomolecular phase separation driven by specific interactions PLOS COMPUTATIONAL BIOLOGY Weiner, B. G., Pyo, A. G. T., Meir, Y., Wingreen, N. S. 2021; 17 (12): e1009748

  Abstract

  Eukaryotic cells partition a wide variety of important materials and processes into biomolecular condensates-phase-separated droplets that lack a membrane. In addition to nonspecific electrostatic or hydrophobic interactions, phase separation also depends on specific binding motifs that link together constituent molecules. Nevertheless, few rules have been established for how these ubiquitous specific, saturating, motif-motif interactions drive phase separation. By integrating Monte Carlo simulations of lattice-polymers with mean-field theory, we show that the sequence of heterotypic binding motifs strongly affects a polymer's ability to phase separate, influencing both phase boundaries and condensate properties (e.g. viscosity and polymer diffusion). We find that sequences with large blocks of single motifs typically form more inter-polymer bonds, which promotes phase separation. Notably, the sequence of binding motifs influences phase separation primarily by determining the conformational entropy of self-bonding by single polymers. This contrasts with systems where the molecular architecture primarily affects the energy of the dense phase, providing a new entropy-based mechanism for the biological control of phase separation.

  View details for DOI 10.1371/journal.pcbi.1009748

  View details for Web of Science ID 000738832800002

  View details for PubMedID 34965250

  View details for PubMedCentralID PMC8751999

• Memory effects in single-molecule force spectroscopy measurements of biomolecular folding PHYSICAL CHEMISTRY CHEMICAL PHYSICS Pyo, A. G. T., Woodside, M. T. 2019; 21 (44): 24527-24534

  Abstract

  Folding is generally assumed to be a Markov process, without memory. When the molecular motion is coupled to that of a probe as in single-molecule force spectroscopy (SMFS) experiments, however, theory predicts that the coupling to a second Markov process should induce memory when monitoring a projection of the full multi-dimensional motion onto a reduced coordinate. We developed a method to evaluate the time constant of the induced memory from its effects on the autocorrelation function, which can be readily determined from experimental data. Applying this method to both simulated SMFS measurements and experimental trajectories of DNA hairpin folding measured by optical tweezers as a model system, we validated the prediction that the linker induces memory. For these measurements, the timescale of the induced memory was found to be similar to the time required for the force probe to respond to changes in the molecule, and in the regime where the experimentally observed dynamics were not significantly perturbed by probe-molecule coupling artifacts. Memory effects are thus a general feature of SMFS measurements induced by the mechanical connection between the molecule and force probe that should be considered when interpreting experimental data.

  View details for DOI 10.1039/c9cp04197d

  View details for Web of Science ID 000498220500026

  View details for PubMedID 31663550

• Measuring the average shape of transition paths during the folding of a single biological molecule PROCEEDINGS OF THE NATIONAL ACADEMY OF SCIENCES OF THE UNITED STATES OF AMERICA Hoffer, N. Q., Neupane, K., Pyo, A. G. T., Woodside, M. T. 2019; 116 (17): 8125-8130

  Abstract

  Transition paths represent the parts of a reaction where the energy barrier separating products and reactants is crossed. They are essential to understanding reaction mechanisms, yet many of their properties remain unstudied. Here, we report measurements of the average shape of transition paths, studying the folding of DNA hairpins as a model system for folding reactions. Individual transition paths were detected in the folding trajectories of hairpins with different sequences held under tension in optical tweezers, and path shapes were computed by averaging all transitions in the time domain, 〈t(x)〉, or by averaging transitions of a given duration in the extension domain, 〈x(t|τ)〉 τ Whereas 〈t(x)〉 was close to straight, with only a subtle curvature, 〈x(t|τ)〉 τ had more pronounced curvature that fit well to theoretical expectations for the dominant transition path, returning diffusion coefficients similar to values obtained previously from independent methods. Simulations suggested that 〈t(x)〉 provided a less reliable representation of the path shape than 〈x(t|τ)〉 τ , because it was far more sensitive to the effects of coupling the molecule to the experimental force probe. Intriguingly, the path shape variance was larger for some hairpins than others, indicating sequence-dependent changes in the diversity of transition paths reflective of differences in the character of the energy barriers, such as the width of the barrier saddle-point or the presence of parallel paths through multiple barriers between the folded and unfolded states. These studies of average path shapes point the way forward for probing the rich information contained in path shape fluctuations.

  View details for DOI 10.1073/pnas.1816602116

  View details for Web of Science ID 000465363700012

  View details for PubMedID 30952784

  View details for PubMedCentralID PMC6486767

• Transition-path properties for folding reactions in the limit of small barriers JOURNAL OF CHEMICAL PHYSICS Pyo, A. G. T., Hoffer, N. Q., Neupane, K., Woodside, M. T. 2018; 149 (11): 115101

  Abstract

  Transition paths are of great interest because they encapsulate information about the mechanisms of barrier-crossing reactions. Analysis of experiments measuring biomolecular folding reactions has relied on expressions for properties of transition paths such as transition-path times and velocities that hold in the limit of large harmonic barriers, but real molecules often have relatively small barriers. Recent theoretical work presented more general expressions for transition-path properties. Here we extend this work, deriving expressions from the general case that can be applied to small harmonic barriers. We first compared the performance of small-barrier, large-barrier, and general solutions when applied to simulated transitions, focusing on improvements in estimates of the diffusion coefficient determined from transition times and velocities. We then applied these expressions to experimental data from force spectroscopy measurements of DNA hairpins. We found that the low-barrier approximation and exact solution reduced or resolved the small but systematic inconsistencies that had arisen from assuming large harmonic barriers, demonstrating the practical utility of the new equations for analyzing experimental data.

  View details for DOI 10.1063/1.5046692

  View details for Web of Science ID 000445368000019

  View details for PubMedID 30243275

• Probing Position-Dependent Diffusion in Folding Reactions Using Single-Molecule Force Spectroscopy BIOPHYSICAL JOURNAL Foster, D. A. N., Petrosyan, R., Pyo, A. G. T., Hoffmann, A., Wang, F., Woodside, M. T. 2018; 114 (7): 1657-1666

  Abstract

  Folding of proteins and nucleic acids involves a diffusive search over a multidimensional conformational energy landscape for the minimal-energy structure. When examining the projection of conformational motions onto a one-dimensional reaction coordinate, as done in most experiments, the diffusion coefficient D is generally position dependent. However, it has proven challenging to measure such position-dependence experimentally. We investigated the position-dependence of D in the folding of DNA hairpins as a simple model system in two ways: first, by analyzing the round-trip time to return to a given extension in constant-force extension trajectories measured by force spectroscopy, and second, by analyzing the fall time required to reach a given extension in force jump measurements. These methods yielded conflicting results: the fall time implied a fairly constant D, but the round-trip time implied variations of over an order of magnitude. Comparison of experiments with computational simulations revealed that both methods were strongly affected by experimental artifacts inherent to force spectroscopy measurements, which obscured the intrinsic position-dependence of D. Lastly, we applied Kramers's theory to the kinetics of hairpins with energy barriers located at different positions along the hairpin stem, as a crude probe of D at different stem positions, and we found that D did not vary much as the barrier was moved along the reaction coordinate. This work underlines the difficulties faced when trying to deduce position-dependent diffusion coefficients from experimental folding trajectories.

  View details for DOI 10.1016/j.bpj.2018.02.026

  View details for Web of Science ID 000430214500015

  View details for PubMedID 29642035

  View details for PubMedCentralID PMC5954564