How to Learn Biomolecular Condensates: From Phase Separation to Stress Granules, Dynamic Composition and Material-State Control

Wait, What? Cells Can Build Compartments Without Membranes

A nucleus needs a membrane.

A lysosome needs a membrane.

But the nucleolus, stress granules and many other cellular assemblies can concentrate selected molecules without surrounding them with a lipid bilayer.

These structures are often called biomolecular condensates.

Many form through phase-separation-related physics, but not every molecular cluster is a simple liquid droplet.

membraneless does not mean structureless—and droplet-like does not automatically prove liquid–liquid phase separation.

The One-Sentence Answer

Learn biomolecular condensates by asking how many weak interactions collectively concentrate molecules above a threshold, how the resulting compartment exchanges material with its surroundings, and how cells tune its composition and material state through RNA, post-translational modification, energy use and protein quality control.

Stage 1: Start With Mixing

Oil and water can separate into different phases because molecules prefer some neighbours over others.

Cells contain much more complex mixtures of proteins, RNA, DNA, lipids, ions and metabolites.

Under suitable conditions, selected biomolecules can become enriched in one region while remaining less concentrated elsewhere.

Stage 2: Phase Separation Creates Coexisting Concentration States

In a simplified liquid–liquid phase-separation model, one dense phase coexists with a dilute phase.

The dense phase is not defined merely by looking round under a microscope.

It is defined by a thermodynamic and compositional relationship between phases.

Stage 3: Multivalency Makes Weak Interactions Powerful

One weak molecular interaction may be transient.

Dozens of weak interactions acting together can create a stable collective assembly.

Multivalency can arise from:

multiple binding domains;
repeated short motifs;
intrinsically disordered regions;
RNA with many binding sites;
oligomerisation.

Stage 4: Intrinsically Disordered Regions Expand the Interaction Grammar

Intrinsically disordered regions do not settle into one rigid three-dimensional fold.

They can present many transient interaction motifs.

That makes them common contributors to condensate formation, although folded domains and nucleic acids can be equally important.

Stage 5: A Saturation Threshold Matters

Below a critical concentration, components may remain broadly dispersed.

Above a threshold, a dense phase can become favourable.

This leads to a powerful professional question:

did the system cross a concentration threshold, or did a local scaffold simply recruit more molecules?

Stage 6: A Phase Diagram Is Better Than a Single Picture

Condensate formation can depend on:

protein concentration;
RNA concentration;
salt;
pH;
temperature;
post-translational modifications.

Mapping these variables reveals an operating region rather than one yes/no observation.

Stage 7: RNA Can Promote or Suppress Condensation

RNA is multivalent and negatively charged.

At some ratios, it can nucleate protein–RNA condensates.

At higher ratios, it can dissolve or reshape them by saturating binding sites.

Therefore:

more RNA does not always mean more condensation.

Stage 8: Condensates Are Selective

A condensate enriches some molecules and excludes others.

This selectivity is described using partition coefficients and interaction preferences.

The biological function may depend less on the visible droplet and more on exactly which molecules enter.

Stage 9: Composition Can Change While the Condensate Persists

A major 2025–2026 Nature Reviews Molecular Cell Biology synthesis emphasises that condensates can maintain an identifiable compartment while continuously changing molecular composition with time and stimulus.

Identity is therefore dynamic.

Stage 10: The Nucleolus Is a Multiphase Condensate

The nucleolus contains spatially distinct regions associated with rRNA transcription, processing and ribosomal assembly.

It demonstrates that one condensate can contain subphases with different material properties and compositions.

The canonical owner of ribosome manufacture remains nucleolar/ribosome-biogenesis biology; this article owns the general physical principle.

Stage 11: Stress Granules Assemble During Translation Stress

When translation initiation is inhibited, untranslated mRNAs and RNA-binding proteins can accumulate into stress granules.

G3BP1/2 proteins are important organisers in many mammalian stress-granule systems.

The granules can temporarily redistribute RNA and signalling factors during stress.

Stage 12: Stress Granules Are Not Simple Storage Lockers

They exchange components rapidly, interact with translation machinery and differ according to stress type and duration.

A granule after heat shock may not be compositionally identical to one after oxidative stress.

Stage 13: P-Bodies Are Related but Distinct

Processing bodies, or P-bodies, are cytoplasmic RNA–protein assemblies enriched in factors linked to mRNA repression, decapping and decay.

Stress granules and P-bodies can interact, but they are not synonyms.

Stage 14: Condensates Can Organise Signalling

Receptor and signalling proteins can form high-local-concentration assemblies.

This can increase reaction rates, exclude inhibitors or create thresholds.

Condensation therefore can behave like a biochemical logic gate.

Stage 15: DNA and Chromatin Can Participate Too

Transcription factors, coactivators, RNA polymerase and chromatin proteins can form concentrated assemblies at selected genomic regions.

However, the field has learned to be cautious:

seeing a bright transcriptional focus does not by itself prove equilibrium liquid–liquid phase separation.

Stage 16: Membrane Domains and Protein Condensates Share Physics

A 9 April 2026 Nature Reviews Molecular Cell Biology review emphasises that phase separation can organise both:

lateral membrane domains;
protein/nucleic-acid condensates.

The molecular systems differ, but the general principle of coexisting phases provides a useful bridge to Membrane Biophysics and Lipid Bilayers.

Stage 17: Condensates Have Material Properties

A condensate can be:

fluid-like;
viscoelastic;
gel-like;
solid-like.

Material state affects molecular diffusion, fusion, deformation and biochemical access.

“Condensate” therefore describes organisation, not one universal viscosity.

Stage 18: Liquid-to-Solid Transitions Can Be Functional

Some assemblies become more elastic or solid as they mature.

This can stabilise structures or create long-lived memory.

Solidification is not automatically pathological.

Stage 19: But Aberrant Hardening Can Promote Disease

RNA-binding proteins such as FUS and TDP-43 can form condensates whose material properties change with mutation, concentration, RNA binding, ageing or stress.

Persistent liquid-to-gel or liquid-to-solid transitions can increase aggregation risk in selected neurodegenerative contexts.

This connects to, but does not replace, Protein Folding and Proteostasis.

Stage 20: Phase Separation and Fibril Formation Are Not the Same

A protein can:

phase-separate without forming amyloid fibrils;
form fibrils without a long-lived liquid-condensate stage;
use condensation as one route that alters aggregation kinetics.

Recent 2026 work directly examines how phase separation and fibrillisation can be mechanistically decoupled.

Stage 21: Post-Translational Modifications Tune Interaction Strength

Phosphorylation, methylation, acetylation, SUMOylation and ubiquitination can alter:

charge;
binding motifs;
valency;
partner recognition.

A cell can therefore dissolve or assemble a condensate without changing total protein abundance.

Stage 22: ATP Can Influence Condensates in More Than One Way

ATP powers helicases, chaperones and remodelling enzymes that alter condensate composition.

At high concentrations it can also influence solubility and intermolecular interactions directly.

Living condensates are therefore not always passive equilibrium droplets.

Stage 23: Chaperones Help Maintain Material Quality

Protein-quality-control systems can prevent inappropriate hardening or aggregation within condensates.

The boundary between condensate biology and proteostasis is therefore operationally important.

Stage 24: FRAP Measures Exchange, Not “Liquidness” by Itself

Fluorescence recovery after photobleaching measures how fluorescent molecules repopulate a bleached region.

Fast recovery can indicate rapid exchange.

Slow recovery can reflect strong binding, low diffusion or a more solid-like state.

But:

FRAP recovery alone does not prove phase separation.

Stage 25: Fusion and Roundness Are Also Insufficient Alone

Two droplets merging and becoming round is consistent with surface tension.

Yet active cellular structures can sometimes mimic these behaviours.

Strong evidence combines multiple independent tests.

Stage 26: Optogenetic Systems Test Causality

Light-controlled oligomerisation systems such as optoDroplet-type approaches can drive local condensation on demand.

Researchers can ask whether forming an assembly is sufficient to alter signalling or transcription.

Induced condensates remain engineered systems and may not perfectly reproduce native biology.

Stage 27: High-Throughput Methods Are Moving Beyond Hand-Picked Images

A 2026 Nature Communications study developed flow-cytometry-based approaches to measure phase-separation behaviour at scale.

This helps convert qualitative microscopy into population-level quantitative data.

Stage 28: Single-Molecule Methods Reveal Hidden Heterogeneity

Tracking individual molecules can estimate:

residence times;
diffusion;
binding states;
exchange between dense and dilute phases.

A visually uniform droplet may contain several molecular behaviours.

Stage 29: Professional Condensate Science Is an Evidence-Standards Problem

The professional question becomes:

What physical mechanism generated this molecular enrichment, what is the phase or material state, how selective and dynamic is its composition, and which experiment separates phase separation from ordinary clustering or aggregation?

Misconceptions Worth Hunting

Every cellular punctum is a condensate.
Every condensate is a liquid droplet.
Round shape proves liquid–liquid phase separation.
FRAP recovery proves liquid behaviour.
Intrinsically disordered regions are required for every condensate.
More RNA always increases phase separation.
Solidification is always pathological.
Phase separation and protein aggregation are identical.

Transfer Check

A fluorescent protein forms round puncta that fuse.

Have you proved liquid–liquid phase separation?

No.

Now concentration crosses a reproducible threshold and dilute-phase concentration buffers while a dense phase grows.

Is the case stronger?

Yes.

Finally, the condensate shows slow FRAP recovery.

Does that mean it is definitely a solid?

No. Binding kinetics and viscoelasticity need additional measurement.

How We Know the Learning Has Held

A learner should be able to explain:

multivalency and saturation concentration;
how RNA changes phase behaviour;
partitioning and selectivity;
stress granules versus P-bodies;
liquid, viscoelastic and solid-like states;
why disease can involve material-state transitions;
why FRAP, fusion and roundness are insufficient individually;
how optogenetics and single-molecule methods add causal and kinetic evidence.

Model Limits

In-vitro phase diagrams simplify the crowded, energy-consuming cell. Overexpressed fluorescent proteins can create non-physiological condensates. The term “phase separation” is sometimes used too broadly. Condensates can be actively maintained and compositionally heterogeneous, so equilibrium liquid-droplet models are useful but incomplete.

Professional condensate biology keeps:

concentration + interaction valency + composition + material state + molecular exchange + causal perturbation

visible together.

Connect This to the eduKate Science Estate

Research Sources and Further Learning

The Quiet Ending

The beginner asks, “Why did these proteins form a droplet?”

The developing biophysicist asks, “What interaction network and concentration threshold produced the dense phase?”

And the professional asks:

Which physical state, molecular composition and causal experiment justify calling this assembly a functional biomolecular condensate rather than simply a cluster?

Small Group Tutorials