Search for: All records

Cells mediate interactions with the extracellular environment through a crowded assembly of transmembrane proteins, glycoproteins and glycolipids on their plasma membrane. The extent to which surface crowding modulates the biophysical interactions of ligands, receptors, and other macromolecules is poorly understood due to the lack of methods to quantify surface crowding on native cell membranes. In this work, we demonstrate that physical crowding on reconstituted membranes and live cell surfaces attenuates the effective binding affinity of macromolecules such as IgG antibodies in a surface crowding-dependent manner. We combine experiment and simulation to design a crowding sensor based on this principle that provides a quantitative readout of cell surface crowding. Our measurements reveal that surface crowding decreases IgG antibody binding by 2 to 20 fold in live cells compared to a bare membrane surface. Our sensors show that sialic acid, a negatively charged monosaccharide, contributes disproportionately to red blood cell surface crowding via electrostatic repulsion, despite occupying only ~1% of the total cell membrane by mass. We also observe significant differences in surface crowding for different cell types and find that expression of single oncogenes can both increase and decrease crowding, suggesting that surface crowding may be an indicator of both cell type and state. Our high-throughput, single-cell measurement of cell surface crowding may be combined with functional assays to enable further biophysical dissection of the cell surfaceome. 

We introduce a simple but general online learning framework in which a learner plays against an adversary in a vector-valued game that changes every round. Even though the learner's objective is not convex-concave (and so the minimax theorem does not apply), we give a simple algorithm that can compete with the setting in which the adversary must announce their action first, with optimally diminishing regret. We demonstrate the power of our framework by using it to (re)derive optimal bounds and efficient algorithms across a variety of domains, ranging from multicalibration to a large set of no-regret algorithms, to a variant of Blackwell's approachability theorem for polytopes with fast convergence rates. As a new application, we show how to (multi)calibeat'' an arbitrary collection of forecasters --- achieving an exponentially improved dependence on the number of models we are competing against, compared to prior work. 

Phase separation of biomolecules into condensates has emerged as a mechanism for intracellular organization and affects many intracellular processes, including reaction pathways through the clustering of enzymes and pathway intermediates. Precise and rapid spatiotemporal control of reactions by condensates requires tuning of their sizes. However, the physical processes that govern the distribution of condensate sizes remain unclear. Here we show that both native and synthetic condensates display an exponential size distribution, which is captured by Monte Carlo simulations of fast nucleation followed by coalescence. In contrast, pathological aggregates exhibit a power-law size distribution. These distinct behaviours reflect the relative importance of nucleation and coalescence kinetics. We demonstrate this by utilizing a combination of synthetic and native condensates to probe the underlying physical mechanisms determining condensate size. The appearance of exponential distributions for abrupt nucleation versus power-law distributions under continuous nucleation may reflect a general principle that determines condensate size distributions.

 

Abstract A major informatic challenge in single cell RNA-sequencing analysis is the precise annotation of datasets where cells exhibit complex multilayered identities or transitory states. Here, we present devCellPy a highly accurate and precise machine learning-enabled tool that enables automated prediction of cell types across complex annotation hierarchies. To demonstrate the power of devCellPy , we construct a murine cardiac developmental atlas from published datasets encompassing 104,199 cells from E6.5-E16.5 and train devCellPy to generate a cardiac prediction algorithm. Using this algorithm, we observe a high prediction accuracy (>90%) across multiple layers of annotation and across de novo murine developmental data. Furthermore, we conduct a cross-species prediction of cardiomyocyte subtypes from in vitro - derived human induced pluripotent stem cells and unexpectedly uncover a predominance of left ventricular (LV) identity that we confirmed by an LV-specific TBX5 lineage tracing system. Together, our results show devCellPy to be a useful tool for automated cell prediction across complex cellular hierarchies, species, and experimental systems. 

The cell nucleus can be thought of as a complex, dynamic, living material, which functions to organize and protect the genome and coordinate gene expression. These functions are achieved via intricate mechanical and biochemical interactions among its myriad components, including the nuclear lamina, nuclear bodies, and the chromatin itself. While the biophysical organization of the nuclear lamina and chromatin have been thoroughly studied, the concept that liquid–liquid phase separation and related phase transitions play a role in establishing nuclear structure has emerged only recently. Phase transitions are likely to be intimately coupled to the mechanobiology of structural elements in the nucleus, but their interplay with one another is still not understood. Here, we review recent developments on the role of phase separation and mechanics in nuclear organization and discuss the functional implications in cell physiology and disease states. 

We introduce a simple but general online learning framework in which a learner plays against an adversary in a vector-valued game that changes every round. Even though the learner's objective is not convex-concave (and so the minimax theorem does not apply), we give a simple algorithm that can compete with the setting in which the adversary must announce their action first, with optimally diminishing regret. We demonstrate the power of our framework by using it to (re)derive optimal bounds and efficient algorithms across a variety of domains, ranging from multicalibration to a large set of no-regret algorithms, to a variant of Blackwell's approachability theorem for polytopes with fast convergence rates. As a new application, we show how to (multi)calibeat'' an arbitrary collection of forecasters --- achieving an exponentially improved dependence on the number of models we are competing against, compared to prior work. 

Despite the great success of inferring and programming with universal types, their dual—existential types—are much harder to work with. Existential types are useful in building abstract types, working with indexed types, and providing first-class support for refinement types. This paper, set in the context of Haskell, presents a bidirectional type-inference algorithm that infers where to introduce and eliminate existentials without any annotations in terms, along with an explicitly typed, type-safe core language usable as a compilation target. This approach is backward compatible. The key ingredient is to use strong existentials, which support (lazily) projecting out the encapsulated data, not weak existentials accessible only by pattern-matching. 

Abstract Sudden cardiac death from arrhythmia is a major cause of mortality worldwide. In this study, we developed a novel deep learning (DL) approach that blends neural networks and survival analysis to predict patient-specific survival curves from contrast-enhanced cardiac magnetic resonance images and clinical covariates for patients with ischemic heart disease. The DL-predicted survival curves offer accurate predictions at times up to 10 years and allow for estimation of uncertainty in predictions. The performance of this learning architecture was evaluated on multi-center internal validation data and tested on an independent test set, achieving concordance indexes of 0.83 and 0.74 and 10-year integrated Brier scores of 0.12 and 0.14. We demonstrate that our DL approach, with only raw cardiac images as input, outperforms standard survival models constructed using clinical covariates. This technology has the potential to transform clinical decision-making by offering accurate and generalizable predictions of patient-specific survival probabilities of arrhythmic death over time.