When we consider intracellular organization, we often picture different regions of the cell forming tiny membrane-bound compartments, like ribosomes and mitochondria. However, many important micro-environments inside the cell are not enclosed by membranes. For decades, scientists have wondered how do these "membraneless" organelles (more generally termed biomolecular condensates) arise, and how are their structures maintained without physical membranes?

Recent ground-breaking experiments have proposed the transformative paradigm of "liquid-liquid phase separation", which suggests that the physical chemistry of phase-separation of multicomponent mixtures sustains such micro-environments: Analogous to the separation of oil and water into distinct liquid phases, biomolecules (proteins, RNA, DNA) in the cytoplasm and nucleoplasm exhibit attractive interactions with each other that drive them to condense and then undergo phase separation, forming "oil-like droplets" inside the cells.

In this area, we are interested in:

• Discovering the molecular mechanisms, alongside liquid—liquid phase separation, that give rise to biomolecular condensates
• Probing how environmental stressors tune the phase behaviors of biomolecules
• Elucidating the physical determinants of biomolecular condensate structure, transport and mechanical properties, and function

Biomolecular condensates are often regarded as liquid-like compartments and, more recently, have been described as viscoelastic fluids. Generally, these compartments exist as dynamic micro-environments that form and dissolve in response to internal and external cues and the overall needs of the cell. However, overtime condensates have been shown to age and transform from their normal reversible liquid-like states into solid-like aggregates. A leading hypothesis is that these solid-like aggregates are linked to neurodegenerative diseases (including ALS and Frontotemporal dementia).

Interestingly, several proteins that are prone to phase separation inside cells are also implicated in neurodegenerative diseases, suggesting that the condensed phase may be a precursor to such disease states. Apart from these disorders, there is also evidence that condensates may function as sites of aberrant cellular signaling (e.g., in cases where condensate regulatory mechanisms become defective), leading to certain cancers. However, the precise links between biomolecular phase separation and disease are still not well understood.

Here, we are interested in:

• Uncovering the molecular signatures of condensates that are prone to aggregation
• Understanding the role of specific and non-specific interactions in driving condensate aging
• Determining the underlying mechanisms involved in condensate aging, and contrasting those to classical protein aggregation models

Resolving the behavior of individual biomolecules in intracellular compartments cannot be done by experiments alone. Indeed, computational modeling offers an ideal complement to the many exciting experimental advances and have the power to provide us with closeup views that are inaccessible in experiments. However, there are several non-trivial prerequisites for achieving realistic modeling of biomolecular phase separation and ensuing compartments: (1) to study collective behaviors we need to model systems that are large enough (i.e., hundreds to thousands of copies of the biomolecules involved); (2) to achieve quantitative accuracy we must include all the relevant molecular details of the biomolecules (including structural information); (3) to mimic intracellular environments we need to appropriately account for effects of ionic salt, pH, and crowding.

To this end, we exploit techniques such as molecular dynamics, Monte Carlo simulations and principles from statistical mechanics, soft condensed matter physics, and polymer physics, to develop modeling and simulation approaches for studying collective behaviors of biomolecules. We adopt a multiscale simulation framework, where we take advantage of key information at different spatiotemporal scales. Specifically, we exploit atomistic simulations (bottom-up) and leverage experimental data (top-down) to design accurate coarse-grained and minimal models for interrogating in intracellular compartments.

Here our focus encompasses developing models and computer simulation approaches for:

• High-resolution structural characterization of compartments
• Accurate prediction of thermodynamic properties
• Measuring transport and mechanical properties

An overarching goal of our research group is to control and tune biomolecular self-assembly and organization, with qualitative and quantitative precision. An immediate goal is to control biomolecular condensate formation and dissolution on demand. By achieving exquisite control of these processes, we can engineer new functions in condensates (e.g., creating new metabolic hubs), as well as bypass unwanted functions (e.g., in the case of disease states).

Accordingly, we combine computer simulations, machine learning approaches, and experiments to design intracellular compartments with prescribed properties and to tune the properties of existing condensates.

In this area, our focus is to leverage our understanding of intracellular compartmentalization for:

• Designing therapeutics for condensate-linked diseases
• Developing new soft materials based on biomolecular self-assembly and organization, including stimuli responsive polymers
• Engineering intracellular compartments with improved and novel functions, including for sustainability applications