Four-dimensional chromosome reconstruction elucidates the spatiotemporal reorganization of the mammalian X chromosome.

Chromosomes are segmented into domains and compartments, but how these structures are spatially related in three dimensions (3D) is unclear. Here, we developed tools that directly extract 3D information from Hi-C experiments and integrate the data across time. With our "4DHiC" method, we use X chromosome inactivation (XCI) as a model to examine the time evolution of 3D chromosome architecture during large-scale changes in gene expression. Our modeling resulted in several insights. Both A/B and S1/S2 compartments divide the X chromosome into hemisphere-like structures suggestive of a spatial phase-separation. During the XCI, the X chromosome transits through A/B, S1/S2, and megadomain structures by undergoing only partial mixing to assume new structures. Interestingly, when an active X chromosome (Xa) is reorganized into an inactive X chromosome (Xi), original underlying compartment structures are not fully eliminated within the Xi superstructure. Our study affirms slow mixing dynamics in the inner chromosome core and faster dynamics near the surface where escapees reside. Once established, the Xa and Xi resemble glassy polymers where mixing no longer occurs. Finally, Xist RNA molecules initially reside within the A compartment but transition to the interface between the A and B hemispheres and then spread between hemispheres via both surface and core to establish the Xi.

Scaling molecular dynamics beyond 100,000 processor cores for large-scale biophysical simulations.

The growing interest in the complexity of biological interactions is continuously driving the need to increase system size in biophysical simulations, requiring not only powerful and advanced hardware but adaptable software that can accommodate a large number of atoms interacting through complex forcefields. To address this, we developed and implemented strategies in the GENESIS molecular dynamics package designed for large numbers of processors. Long-range electrostatic interactions were parallelized by minimizing the number of processes involved in communication. A novel algorithm was implemented for nonbonded interactions to increase single instruction multiple data (SIMD) performance, reducing memory usage for ultra large systems. Memory usage for neighbor searches in real-space nonbonded interactions was reduced by approximately 80%, leading to significant speedup. Using experimental data describing physical 3D chromatin interactions, we constructed the first atomistic model of an entire gene locus (GATA4). Taken together, these developments enabled the first billion-atom simulation of an intact biomolecular complex, achieving scaling to 65,000 processes (130,000 processor cores) with 1 ns/day performance. Published 2019. This article is a U.S. Government work and is in the public domain in the USA.

Polymer glass transition occurs at the marginal rigidity point with connectivity z* = 4.

We re-examine the physical origin of the polymer glass transition from the point of view of marginal rigidity, which is achieved at a certain average number of mechanically active intermolecular contacts per monomer. In the case of polymer chains in a melt/poor solvent, each monomer has two neighbors bound by covalent bonds and also a number of central-force contacts modelled by the Lennard-Jones (LJ) potential. We find that when the average number of contacts per monomer (covalent and non-covalent) exceeds the critical value z* ≈ 4, the system becomes solid and the dynamics arrested - a state that we declare the glass. Coarse-grained Brownian dynamics simulations show that at sufficient strength of LJ attraction (which effectively represents the depth of quenching, or the quality of solvent) the polymer globule indeed crosses the threshold of z*, and becomes a glass with a finite zero-frequency shear modulus, G∝ (z-z*). We verify this by showing the distinction between the 'liquid' polymer droplet at z < z*, which changes shape and adopts the spherical conformation in equilibrium, and the glassy 'solid' droplet at z > z*, which retains its shape frozen at the moment of z* crossover. These results provide a robust microscopic criterion to tell the liquid apart from the glass for the linear polymers.

The next revolution in computational simulations: Harnessing AI and quantum computing in molecular dynamics.

The integration of artificial intelligence, machine learning and quantum computing into molecular dynamics simulations is catalyzing a revolution in computational biology, improving the accuracy and efficiency of simulations. This review describes the advancements and applications of these technologies to process vast molecular dynamics simulation datasets, adapt parameters of simulations and gain insight into complex biological processes. These advances include the use of predictive force fields, adaptive algorithms and quantum-assisted methodologies. While the integration of artificial intelligence and quantum computing with MD simulations provides insightful and stimulating improvements to our understanding of molecular mechanisms, it could introduce new issues related to data quality, interpretability of models and computational complexity. Modern multidisciplinary approaches are needed to navigate these challenges and exploit the potential of these emerging technologies for MD simulations of biomolecular systems.

Human Coronavirus Infection Reorganizes Spatial Genomic Architecture in Permissive Lung Cells.

Chromatin conformation capture followed by next-generation sequencing in combination with large-scale polymer simulations (4DHiC) produces detailed information on genomic loci interactions, allowing for the interrogation of 3D spatial genomic structures. Here, Hi-C data was acquired from the infection of fetal lung fibroblast (MRC5) cells with α-coronavirus 229E (CoV229E). Experimental Hi-C contact maps were used to determine viral-induced changes in genomic architecture over a 48-hour time period following viral infection, revealing substantial alterations in contacts within chromosomes and in contacts between different chromosomes. To gain further structural insight and quantify the underlying changes, we applied the 4DHiC polymer simulation method to reconstruct the 3D genomic structures and dynamics corresponding to the Hi-C maps. The models successfully reproduced experimental Hi-C data, including the changes in contacts induced by viral infection. Our 3D spatial simulations uncovered widespread chromatin restructuring, including increased chromosome compactness and A-B compartment mixing arising from infection. Our model also suggests increased spatial accessibility to regions containing interferon-stimulated genes upon infection with CoV229E, followed by chromatin restructuring at later time points, potentially inducing the migration of chromatin into more compact regions. This is consistent with previously observed suppression of gene expression. Our spatial genomics study provides a mechanistic structural basis for changes in chromosome architecture induced by coronavirus infection in lung cells.

Unveiling RCOR1 as a rheostat at transcriptionally permissive chromatin.

RCOR1 is a known transcription repressor that recruits and positions LSD1 and HDAC1/2 on chromatin to erase histone methylation and acetylation. However, there is currently an incomplete understanding of RCOR1's range of localization and function. Here, we probe RCOR1's distribution on a genome-wide scale and unexpectedly find that RCOR1 is predominantly associated with transcriptionally active genes. Biochemical analysis reveals that RCOR1 associates with RNA Polymerase II (POL-II) during transcription and deacetylates its carboxy-terminal domain (CTD) at lysine 7. We provide evidence that this non-canonical RCOR1 activity is linked to dampening of POL-II productive elongation at actively transcribing genes. Thus, RCOR1 represses transcription in two ways-first, via a canonical mechanism by erasing transcriptionally permissive histone modifications through associating with HDACs and, second, via a non-canonical mechanism that deacetylates RNA POL-II's CTD to inhibit productive elongation. We conclude that RCOR1 is a transcription rheostat.

Revealing RCOR2 as a regulatory component of nuclear speckles.

BACKGROUND: Nuclear processes such as transcription and RNA maturation can be impacted by subnuclear compartmentalization in condensates and nuclear bodies. Here, we characterize the nature of nuclear granules formed by REST corepressor 2 (RCOR2), a nuclear protein essential for pluripotency maintenance and central nervous system development. RESULTS: Using biochemical approaches and high-resolution microscopy, we reveal that RCOR2 is localized in nuclear speckles across multiple cell types, including neurons in the brain. RCOR2 forms complexes with nuclear speckle components such as SON, SRSF7, and SRRM2. When cells are exposed to chemical stress, RCOR2 behaves as a core component of the nuclear speckle and is stabilized by RNA. In turn, nuclear speckle morphology appears to depend on RCOR2. Specifically, RCOR2 knockdown results larger nuclear speckles, whereas overexpressing RCOR2 leads to smaller and rounder nuclear speckles. CONCLUSION: Our study suggests that RCOR2 is a regulatory component of the nuclear speckle bodies, setting this co-repressor protein as a factor that controls nuclear speckles behavior.

Modeling the Influenza A NP-vRNA-Polymerase Complex in Atomic Detail.

Seasonal flu is an acute respiratory disease that exacts a massive toll on human populations, healthcare systems and economies. The disease is caused by an enveloped Influenza virus containing eight ribonucleoprotein (RNP) complexes. Each RNP incorporates multiple copies of nucleoprotein (NP), a fragment of the viral genome (vRNA), and a viral RNA-dependent RNA polymerase (POL), and is responsible for packaging the viral genome and performing critical functions including replication and transcription. A complete model of an Influenza RNP in atomic detail can elucidate the structural basis for viral genome functions, and identify potential targets for viral therapeutics. In this work we construct a model of a complete Influenza A RNP complex in atomic detail using multiple sources of structural and sequence information and a series of homology-modeling techniques, including a motif-matching fragment assembly method. Our final model provides a rationale for experimentally-observed changes to viral polymerase activity in numerous mutational assays. Further, our model reveals specific interactions between the three primary structural components of the RNP, including potential targets for blocking POL-binding to the NP-vRNA complex. The methods developed in this work open the possibility of elucidating other functionally-relevant atomic-scale interactions in additional RNP structures and other biomolecular complexes.

Connectivity and free-surface effects in polymer glasses.

The glass transition is one of the few unsolved problems in condensed matter physics: agreement on the cause of the slowing down of structural relaxation in glass-forming liquids is lacking. Glasses are amorphous solids, which do not possess the long-range crystalline order, yet display arrested dynamics and the shear elastic modulus characteristic of equilibrium elasticity. It has been suggested that due to the influence of intramolecular interactions and chain connectivity, the nature of the glass transition in polymers and in standard glass-formers is fundamentally different. Here, we discuss the role of connectivity in polymer glasses, demonstrating that although covalent bonding promotes glass formation, bonding sequentiality that defines a polymer chain is not critical in the bulk: glassy dynamics is purely a result of the number of connections per particle, independently of how these connections are formed, agreeing with the classical Phillips-Thorpe topological constraint theory. We show that bonding sequentiality does play an important role in the surface effects of the glass, highlighting a major difference between polymeric and colloidal glasses. Further, we identify the heterogenous dynamics of model coarse-grained polymer chains both in 'bulk' and near the free surface, and demonstrate characteristic domain patterns in local displacement and connectivity.

Structural Transition and Antibody Binding of EBOV GP and ZIKV E Proteins from Pre-Fusion to Fusion-Initiation State.

Membrane fusion proteins are responsible for viral entry into host cells—a crucial first step in viral infection. These proteins undergo large conformational changes from pre-fusion to fusion-initiation structures, and, despite differences in viral genomes and disease etiology, many fusion proteins are arranged as trimers. Structural information for both pre-fusion and fusion-initiation states is critical for understanding virus neutralization by the host immune system. In the case of Ebola virus glycoprotein (EBOV GP) and Zika virus envelope protein (ZIKV E), pre-fusion state structures have been identified experimentally, but only partial structures of fusion-initiation states have been described. While the fusion-initiation structure is in an energetically unfavorable state that is difficult to solve experimentally, the existing structural information combined with computational approaches enabled the modeling of fusion-initiation state structures of both proteins. These structural models provide an improved understanding of four different neutralizing antibodies in the prevention of viral host entry.

Ratcheted diffusion transport through crowded nanochannels.

The problem of transport through nanochannels is one of the major questions in cell biology, with a wide range of applications. In this paper we discuss the process of spontaneous translocation of molecules (Brownian particles) by ratcheted diffusion: a problem relevant for protein translocation along bacterial flagella or injectosome complex, or DNA translocation by bacteriophages. We use molecular dynamics simulations and statistical theory to identify two regimes of transport: at low rate of particle injection into the channel the process is controlled by the individual diffusion towards the open end (the first passage problem), while at a higher rate of injection the crowded regime sets in. In this regime the particle density in the channel reaches a constant saturation level and the resistance force increases substantially, due to the osmotic pressure build-up. To achieve a steady-state transport, the apparatus that injects new particles into a crowded channel has to operate with an increasing power consumption, proportional to the length of the channel and the required rate of transport. The analysis of resistance force, and accordingly--the power required to inject the particles into a crowded channel to overcome its clogging, is also relevant for many microfluidics applications.