Abstract
Biomolecular condensates exhibit numerous emergent properties including material properties, reaction rates, and molecular selectivity. In addition to these known properties, it is overwhelmingly likely that there are unknown emergent properties that are related to both function within condensates and the integration of condensates within cellular regulatory mechanisms. Identification of these unknown properties will require an increased emphasis on top-down methods because emergent properties, by definition, are difficult to identify from bottom-up data. This effort will have a tremendous payoff in helping to ameliorate a conceptual gap responsible for the apparent complexity behind some of the greatest challenges in modern medicine.
Introduction: the whole is not equal to the sum of parts
Biomolecular condensation has drawn attention to the importance of emergent properties and, in doing so, illuminated a conceptual gap underlying some of the greatest challenges in modern medicine. Emergent properties are generally defined as the characteristics of large systems that are not seen in the properties of the building blocks (; ; ). This description applies to many aspects of biomolecular condensates such as molecular diffusion, reaction rates, miscibility, and chemical potentials, all of which are very different than what would be expected from dilute solution characterization. Even the formation of the condensate itself and its material properties are emergent properties. This abundance of emergent properties is not surprising; it is an ubiquitous phenomenon across the sciences that “more is different”, to borrow the words of the Nobel-winning physicist Philip Anderson (). Thus, as strikingly demonstrated by biomolecular condensation, collections of biomolecules do not necessarily behave according to the additive effects of specific binding sites. In the Anderson view, biomolecular emergent properties open a frontier in cellular biology at a scale where data are abundant, but understanding is poor. Thus, the discovery of biomolecular condensation should be treated like the first glimmer of light in the dark unknown; what other emergent properties are yet to be discovered, and what methods are needed to find them?
Emergent properties are invisible to bottom-up methods
Scientific investigation relies on both top-down and bottom-up methods (; ). Top-down methods examine large scale phenomenology to infer the mechanisms at smaller scales. A classic example is the development of thermodynamics where systematic measurements of engines and ideal gasses led to universal principles like internal energy and entropy. A more recent example of top-down investigation is the proliferation of phase diagrams that have proven indispensable to the field of biomolecular condensation. In contrast, bottom-up investigations study the properties of small-scale components to learn how the collective system behaves. Returning to the thermodynamics example, bottom-up application of Newtonian physics yielded the kinetic theory of gasses, which provides complementary insights to the thermodynamic potentials (E, S, H, G, etc.) discovered by the top-down approach. Bottom-up methods dominate molecular and cellular biology, in which most studies look for molecular interactions that are pieced together into mechanistic pathways. Unfortunately, the bottom-up approach is problematic when emergent properties are present because, by definition, an emergent property describes situations where a system behaves different than the sum of its parts (). For example, efforts in the early 2000s to predict the properties of regulatory networks by integrating kinetic rate equations could not account for how rates might change due to the unknown (at the time) phenomenon of biomolecular condensation. Similarly, while kinetic theory correctly describes many properties of gasses, the bottom-up Newtonian approach could not have discovered entropy, a concept that is applicable to systems ranging from protein folding to black holes.
To further illustrate the blind spot of bottom-up methods, it is useful to conduct a thought experiment using a situation where the mechanism depends on an emergent property. One such case is enzyme function, which arises from the emergent properties of the folded state. Without prior knowledge of protein folding, a bottom-up investigation of an enzyme might look for interactions between the component amino acids. The resulting network of interactions for lysozyme would look similar to Figure 1A. This representation provides very little insight as to the presence of a specific binding pocket and catalytic site, which are the emergent properties that drive protein function (Figure 1B). In fact, the catalytic residues, highlighted in red, are well separated on the graph.
FIGURE 1
How much caution should be taken from the examples of protein folding and reaction kinetics within condensates? An optimistic view comes from the success of whole-cell computational models (), which suggests that the bottom-up approach can capture most of cellular biology. An opposing view comes from strong arguments that there are undiscovered biochemical mechanisms, particularly in the areas of IDPs and condensates (; ; ). Regardless of which of these viewpoints ultimately proves correct, Anderson gives another compelling reason to extend beyond bottom-up methodology ().
Concepts are the primary tool of complexity reduction
Scientific concepts are based on emergent properties
Anderson wrote “More is Different” in opposition to the notion that the adjective “fundamental” applied to physics only at the small (sub-atomic particles) and large (cosmology) extremes. He argued that it cannot be claimed that any scale is more or less fundamental than adjacent scales because, due to emergent properties, different principles and concepts would apply at every scale. For example, membranes are described using properties like permeability and flexibility that are meaningless at the scale of individual phospholipids or even small assemblies like micelles. Similarly, the emergent properties of protein folding and active sites do not exist for polypeptides shorter than about 50 residues.
It is instructive to examine modern molecular and cellular biology to see how the principles and concepts described by Anderson enable understanding at each scale. Figure 2 shows an (incomplete) list of concepts ranging from electron wavefunctions to tissues. Consistent with Anderson’s claim, each concept in Figure 2 is based on an emergent property at that scale.
FIGURE 2
The conceptual hierarchy shown in Figure 2 has two features that are worthy of further attention. The first is the close spacing between concepts at scales ranging from 10-10 to 10-8 m. This density of concepts enables a remarkable ability to reason across scales. For example, consider a protein containing a14C atom that decays to 14N. What effect would this perturbation have on the function of the protein? Applying the concepts in Figure 2 suggests that the decay would change the valence and electronegativity of the atom. This change, in turn, would alter amino acid properties like charge, hydrophobicity, and flexibility. Finally, depending on the location of the affected amino acid (surface, interior, or active site), the amino acid properties provide intuition about whether the change would be negligible or alter the protein’s ability to fold, bind substrate, or catalyze a reaction. This ability to rationalize the effects of a femtometer-scale nuclear change on the function of a nanometer scale protein, a million-fold range in scales, is only possible because of a conceptual hierarchy that clarifies the most essential features at each scale while ignoring irrelevant details at smaller scales. In practice, each conceptual step defines a small number of emergent variables that explain the effects of a much greater number of microscopic variables (; Haken, 1977). For example, in protein folding the effects of 20 amino acid substitutions can be mostly understood by the property of hydrophobicity (), and the detailed atomic coordinates in a protein active site can be understood by the effect on a transition state energy. The simplification enabled by mapping microscopic variables to mesoscopic variables gives concepts great power as a tool of complexity reduction.
The mesoscale concept gap inhibits our understanding of cell regulation
A second prominent feature of Figure 2 is the lack of concepts on scales larger than simple feedback circuits (i.e., the lac operon) but smaller than organelles, a range of scales that has been referred to as the cellular mesoscale (). Some of the most challenging problems in modern medicine, like cancer and neurodegeneration, occur due to malfunction at the mesoscale. This is not a coincidence; much of the difficulty in treating these diseases follows from our inability to connect phenomena on scales ranging from molecules to tissues. The paucity of complexity-reducing concepts at the cellular mesoscale means that we are confronted by the apparent complexity that comes from an overwhelming number of microscopic variables. Thus, there is much to be gained from advancing the conceptual frontier at the mesoscale.
The above arguments suggest an auspicious role for biomolecular condensates, which are an emergent property occurring at a scale within the concept gap. However, biomolecular condensation is more often used a step in a biochemical pathway than a mapping to an emergent variable. As a result, biomolecular condensation has not yet led to complexity reduction in cellular functions and, therefore, it is premature to add biomolecular condensation to Figure 2. Furthermore, explaining the effect of 14C decay required 3-4 conceptual steps per decade in length scale, which suggests that the mesoscale gap is too large for a single concept to fill. Therefore, building a smoothly connected conceptual hierarchy will likely require the discovery of emergent functions both within condensates and within the cellular mechanisms that condensates participate in.
Microenvironment coupling creates biochemical pipelines
A candidate for an emergent function in biomolecular condensates, as well as a complexity-reducing concept, is “microenvironment coupling”, which has been proposed to explain the mechanism of ribosome maturation in the granular component of the nucleolus (). In microenvironment coupling the state of a biochemical pathway (in this case ribosome maturation) alters the local macromolecular environment to facilitate the recruitment of proteins appropriate for the next pathway step. This feedback between the biochemical cascade and protein recruitment effectively creates a pipeline for sequential steps. An important distinction is that the spatial coordinate implied by a pipeline (or an assembly line (; )) is replaced by a chemical potential coordinate. Thus, much like the atomic coordinates of a protein active site determines binding affinity, microscopic features of a condensate like density, hydrophobicity, binding site concentrations, and molecular positions can be understood by their effect on an emergent chemical potential gradient. Importantly, microenvironment coupling need not be limited to condensates and, therefore, could be a generic mechanism to pipeline multi-step cascades throughout the cell.
Emergent properties require top-down investigative methods
Since bottom-up methods are poorly suited to identify emergent properties, identification of emergent functions and new concepts must come from top-down approaches (; ; ). The blind spot of bottom-up methods to emergent properties may explain the persistence of a concept gap in spite of an abundance of experimental methods capable of manipulating cells with almost arbitrary precision. In particular, molecular manipulations like knockouts and mutations tend to give binary results () or, at best, one-dimensional response curves. To illustrate the inadequacy of such data, consider the phase behavior of N2 and CO2. A one-dimensional investigation of the effect of reduced temperature would reveal qualitatively different behavior; nitrogen condenses into a liquid while carbon dioxide forms a solid. However, expanding the investigation to include a second dimension, pressure, reveals that both molecules have nearly identical phase diagrams. This shared phase diagram was an important clue that phase behavior is not a feature of a particular molecular structure but, instead, a universal phenomenon resulting from competition between energy and entropy. This example evokes comparisons to fusion oncoproteins and amyloid-driven neurodegeneration, both of which are cases where a large number of different molecular drivers cause macroscopically similar diseases. A multi-dimensional, systematic comparison of these molecules would likely reveal emergent variables that clarify how microscopic factors drive cellular networks into dysregulated states.
Emergent variables are not the same as collective variables that, in many cases, do not promote understanding. For example, while root mean squared deviation (RMSD) is a useful metric to assess protein conformation, it rarely provides mechanistic insight. Similarly, dimensionality reduction methods like principal component analysis (PCA) yield collective variables with little conceptual value.
An example of an emergent variable at the cellular mesoscale comes from E. coli where the effects on growth from the microscopic variables of carbon source, translation inhibitors, and foreign expression vectors can all be understood by a mesoscopic variable describing the cytoplasmic volume allocated to carbon catabolism at the expense of ribosomes (; ; ). An illustrative non-biological example comes from economics where microscopic variables ranging from weather, to politics, to consumer fads, can be understood by their effects on the emergent variables of “supply” and “demand”. Importantly, a microscopic variable that decreases supply will have the same qualitative effect as a variable that increases demand. Determining whether two microscopic variables are manipulating a common emergent variable requires a quantitative comparison between the two response curves. This quantitative comparison is the basis of “data collapse”, a method that reveals emergent behavior when multiple response curves can be scaled to fall on a common master curve. Curve scaling has revealed emergent principles in diverse fields including critical phenomena (), polymers (), economics (Yakovenko and Rosser, 2009), and metabolic trends across organisms (West and Brown, 2005). An intriguing example in the cellular mesoscale is the recent finding that the duration of “on” and “off” event in transcription bursts in drosophila, yeast, and human cells are uniquely determined by the probability the gene is “on” (). This one-to-one correspondence casts doubt on reaction-diffusion models of transcription factor binding, which only determine the on/off ratio.
Top-down experiments (primarily phase diagrams) and theory of simplified in vitro biomolecular condensates have revealed diverse emergent properties with functional implications including conformational conversions (; ), tunable selectivity (), and biomolecular signaling (). In each of these cases, modeling revealed mesoscale order parameters describing how the microscopic details of the condensate relate to the macroscopic phenomenology (; Haken, 1977). The challenge is to adapt this approach in vivo to identify emergent functions and complexity-reducing concepts at the cellular mesoscale. This will require the collection of systematic curves quantifying cellular response to various perturbations (). Modern methods allow for rigorous quantification of cellular experiments, but in many cases it is not clear how to systematically perturb cells in one dimension, let alone the multiple dimensions required to reveal emergent variables. This introduces a challenge to identify microscopic variables that can be varied in a continuous manner. One possibility is the titration of inhibitors (; ). Similarly, extracellular conditions like nutrient concentrations or stressors (temperature, osmolality, oxidants, etc.) may be useful. Alternatively, a series of discrete mutations to a protein of interest could be projected onto a continuous variable like binding affinity, catalytic rate, saturation concentration, IDR length, or phosphorylation level. Another accessible variable is the concentration of intracellular proteins, which can be passively observed using cell-to-cell variability () or directly manipulated at the expression level ().
Comparing the response curved generated by different continuous variables will provide candidate emergent concepts that can be tested by traditional bottom-up methods. Upon confirmation, bottom-up methods can be used to study how emergent variables change between systems. For example, the pairwise interaction map in Figure 1A could be used to reconstruct the folded structure. Furthermore, the new emergent principles can then be used to streamline additional bottom-up investigations, as demonstrated in examples like implicit solvent representations of the hydrophobic effect, bead-spring models of polymers, and kinetic Monte Carlo representations of reaction-diffusion systems.
Conclusion: the mesoscale frontier
The phenomenon of biomolecular condensation is large and slow relative to the traditional scales of molecular biology. This reality runs contrary to the technological frontier that emphasizes resolution gains at short times, small sizes, and ever-more fine-grained molecular control. But the central premise of “More is different” is that small and fast is not the only frontier in science (). In medical science and cellular biology a more significant frontier lies in the treatment of systematic diseases driven by malfunction at the cellular mesoscale. The emergent phenomenon of biomolecular condensation has fortuitously provided illumination into this unknown, as well as a reminder that systems can behave very different than the sum of their parts.
Statements
Data availability statement
The original contributions presented in the study are included in the article/supplementary material, further inquiries can be directed to the corresponding author.
Author contributions
JS: Writing – original draft, Writing – review and editing.
Funding
The author(s) declared that financial support was not received for this work and/or its publication.
Acknowledgments
I would like to thank Nick Wallace for stimulating discussions and Ned Wingreen and Josh Riback for comments on the manuscript.
Conflict of interest
The author(s) declared that this work was conducted in the absence of any commercial or financial relationships that could be construed as a potential conflict of interest.
Generative AI statement
The author(s) declared that generative AI was not used in the creation of this manuscript.
Any alternative text (alt text) provided alongside figures in this article has been generated by Frontiers with the support of artificial intelligence and reasonable efforts have been made to ensure accuracy, including review by the authors wherever possible. If you identify any issues, please contact us.
Publisher’s note
All claims expressed in this article are solely those of the authors and do not necessarily represent those of their affiliated organizations, or those of the publisher, the editors and the reviewers. Any product that may be evaluated in this article, or claim that may be made by its manufacturer, is not guaranteed or endorsed by the publisher.
References
1
AckerJ.WangX.PardalA. J.DesiròD.AgarwalT.ColyerA.et al (2026). Phosphorylation tunes strain-specific protein condensation during rotavirus replication organelle assembly. EMBO J.45, 4733–4765. 10.1038/s44318-026-00814-z
2
AndersonP. W. (1972). More is different. Science. 177, 393–396. 10.1126/science.177.4047.393
3
BhandariK.A. CottenM.KimJ.RosenK.SchmitD. (2021). Structure–function properties in disordered condensates. J. Phys. Chem. B125, 467–476. 10.1021/acs.jpcb.0c11057
4
BondosS. E.DunkerA. K.UverskyV. N. (2022). Intrinsically disordered proteins play diverse roles in cell signaling. Cell. Commun. Signal.20, 20. 10.1186/s12964-022-00821-7
5
ChenP. T.LevoM.ZollerB.GregorT. (2025). A conserved coupling of transcriptional ON and OFF periods underlies bursting dynamics. Nat. Struct. and Mol. Biol.32:10, 1959–20251971. 10.1038/s41594-025-01615-4
6
ColemanP. (2007). Frontier at your fingertips. Nature446, 379. 10.1038/446379a
7
De GennesP. G. (1979). Brownian motions of flexible polymer chains. Nature282, 367–370. 10.1038/282367a0
8
DollingerC.PotolitsynaE.MartinA. G.AnandA.DatarG. K.SchmitJ. D.et al (2025). Nanometer condensate organization in live cells derived from partitioning measurements. bioRxiv, 2025. 02. 10.1101/2025.02.26.640428
9
GalagederaS. K. K.DaoT. P.EnosS. E.ChaudhuriA.SchmitJ. D.CastañedaC. A. (2023). Polyubiquitin ligand-induced phase transitions are optimized by spacing between ubiquitin units. Proc. Natl. Acad. Sci.120, e2306638120. 10.1073/pnas.2306638120
10
HakenH. (1977). Synergetics. Phys. Bull.28, 412. 10.1088/0031-9112/28/9/027
11
HollandJ. H. (1998). Emergence: From Chaos to Order. Oxford University Press.
12
KhanT.KandolaT. S.WuJ.VenkatesanS.KetterE.LangeJ. J.et al (2018). Quantifying nucleation in vivo reveals the physical basis of prion-like phase behavior. Mol. Cell.71, 155–168.e7. 10.1016/J.MOLCEL.2018.06.016
13
LafontaineD. L. J.RibackJ. A.BascetinR.BrangwynneC. P. (2021). The nucleolus as a multiphase liquid condensate. Nat. Rev. Mol. Cell. Biol.22, 165–182. 10.1038/s41580-020-0272-6
14
LazebnikY. (2002). Can a biologist fix a radio? - or, what I learned while studying apoptosis. Prepr. A. T. Cell. Press, 2, 179, 182. 10.1016/S1535-6108(02)00133-2
15
Luthey-SchultenZ.ThornburgZ. R.GilbertB. R. (2022). Integrating cellular and molecular structures and dynamics into whole-cell models. Curr. Opin. Struct. Biol.75, 102392. 10.1016/j.sbi.2022.102392
16
MckenzieR. H. (2025). Emergence: from physics to biology, sociology, and computer science. 10.48550/arXiv.2508.08548
17
NassarR.DignonG. L.RazbanR. M.DillK. A. (2021). The protein folding problem: the role of theory. J. Mol. Biol.433, 167126. 10.1016/j.jmb.2021.167126
18
NovikoffA. B. (1945). The concept of integrative levels and biology. Sci. (1979)101, 209–215. 10.1126/science.101.2618.209
19
OttoR. M.Turska-NowakA.BrownP. M.ReynoldsK. A. (2024). A continuous epistasis model for predicting growth rate given combinatorial variation in gene expression and environment. Cell. Syst.15, 134–148.e7. 10.1016/j.cels.2024.01.003
20
RiziA. K.ByE.AbbottD. (2026). What is emergence, after all?PNAS Nexus5 (2), pgag010. 10.1093/PNASNEXUS/PGAG010
21
SchmitJ.DillK. A. (2023). Next-gen biophysics: look to the forest, beyond the trees. Annu. Rev. Biophys.52, v-viii. 10.1146/annurev-bb-52-030923-100001
22
SchmitJ. D.BouchardJ. J.MartinE. W.MittagT. (2020). Protein network structure enables switching between liquid and gel states. J. Am. Chem. Soc.142, 874–883. 10.1021/jacs.9b10066
23
SchmitJ. D.FericM.DundrM. (2021). How hierarchical interactions make membraneless organelles tick like clockwork. Trends biochem. Sci.46, 525–534. 10.1016/j.tibs.2020.12.011
24
ScottM.HwaT. (2023). Shaping bacterial gene expression by physiological and proteome allocation constraints. Nat. Rev. Microbiol.21, 327–342. 10.1038/s41579-022-00818-6
25
ScottM.GundersonC. W.MateescuE. M.ZhangZ.HwaT. (2010). Interdependence of cell growth and gene expression: origins and consequences. Science. 330, 1099–1102. 10.1126/science.1192588
26
ScottM.HwaT.KishonyR.HatzimanikatisV. (2011). Bacterial growth laws and their applications this review comes from a themed issue on systems biology. Curr. Opin. Biotechnol.22, 559–565. 10.1016/j.copbio.2011.04.014
27
ScottM.KlumppS.MateescuE. M.HwaT. (2014). Emergence of robust growth laws from optimal regulation of ribosome synthesis. Mol. Syst. Biol.10, 747. 10.15252/msb.20145379
28
SearR. P.PagonabarragaI.FlausA. (2015). Life at the mesoscale: the self-organised cytoplasm and nucleoplasm. BMC Biophys.8, 4. 10.1186/s13628-015-0018-6
29
StanleyH. E. (1999). Scaling, universality, and renormalization: three pillars of modern critical phenomena. Rev. Mod. Phys.71, S358–S366. 10.1103/RevModPhys.71.S358
30
TartakoffA.DiMarioP.HurtE.McStayB.PanseV. G.TollerveyD. (2022). The dual nature of the nucleolus. Genes. Dev.36, 765–769. 10.1101/GAD.349748.122
31
WestG. B.BrownJ. H. (2005). The origin of allometric scaling laws in biology from genomes to ecosystems: towards a quantitative unifying theory of biological structure and organization. J. Exp. Biol.208, 1575–1592. 10.1242/jeb.01589
32
YakovenkoV. M.RosserJ. B. (2009). Statistical mechanics of money, wealth, and income. Rev. Mod. Phys.81, 1703–1725. 10.1103/RevModPhys.81.1703
Summary
Keywords
biomolcular condensates, data collapse, emergent properties, molecular biology, top-down, complexity reduction
Citation
Schmit JD (2026) Biomolecular condensates, emergent properties, and the mesoscale concept gap. Front. Biophys. 4:1927727. doi: 10.3389/frbis.2026.1927727
Received
03 July 2026
Revised
29 July 2026
Accepted
03 August 2026
Published
26 August 2026
Volume
4 - 2026
Edited by
Jianhan Chen, University of Massachusetts Amherst, United States
Reviewed by
Davit Potoyan, Iowa State University, United States
Dongheon Lee, FAMU-FSU Department of Chemical and Biomedical Engineering, United States
Updates
Copyright
© 2026 Schmit.
This is an open-access article distributed under the terms of the Creative Commons Attribution License (CC BY). The use, distribution or reproduction in other forums is permitted, provided the original author(s) and the copyright owner(s) are credited and that the original publication in this journal is cited, in accordance with accepted academic practice. No use, distribution or reproduction is permitted which does not comply with these terms.
*Correspondence: Jeremy D. Schmit, schmit@ksu.edu
Disclaimer
All claims expressed in this article are solely those of the authors and do not necessarily represent those of their affiliated organizations, or those of the publisher, the editors and the reviewers. Any product that may be evaluated in this article or claim that may be made by its manufacturer is not guaranteed or endorsed by the publisher.