Search

SARS-CoV-2 has intricate mechanisms for initiating infection, immune evasion/suppression and replication that depend on the structure and dynamics of its constituent proteins. Many protein structures have been solved, but far less is known about their relevant conformational changes. To address this challenge, over a million citizen scientists banded together through the Folding@home distributed computing project to create the first exascale computer and simulate 0.1 seconds of the viral proteome. Our adaptive sampling simulations predict dramatic opening of the apo spike complex, far beyond that seen experimentally, explaining and predicting the existence of 'cryptic' epitopes. Different spike variants modulate the probabilities of open versus closed structures, balancing receptor binding and immune evasion. We also discover dramatic conformational changes across the proteome, which reveal over 50 'cryptic' pockets that expand targeting options for the design of antivirals. All data and models are freely available online, providing a quantitative structural atlas.

Understanding the structural determinants of a protein’s biochemical properties, such as activity and stability, is a major challenge in biology and medicine. Comparing computer simulations of protein variants with different biochemical properties is an increasingly powerful means to drive progress. However, success often hinges on dimensionality reduction algorithms for simplifying the complex ensemble of structures each variant adopts. Unfortunately, common algorithms rely on potentially misleading assumptions about what structural features are important, such as emphasizing larger geometric changes over smaller ones. Here we present DiffNets, self-supervised autoencoders that avoid such assumptions, and automatically identify the relevant features, by requiring that the low-dimensional representations they learn are sufficient to predict the biochemical differences between protein variants. For example, DiffNets automatically identify subtle structural signatures that predict the relative stabilities of β-lactamase variants and duty ratios of myosin isoforms. DiffNets should also be applicable to understanding other perturbations, such as ligand binding., Comparing and contrasting structural ensembles of different protein variants helps connect specific structural features to a protein’s biochemical properties. Here, the authors propose DiffNets, a self-supervised, deep learning method that streamlines this process.

Simulations can provide tremendous insight into the atomistic details of biological mechanisms, but micro- to millisecond timescales are historically only accessible on dedicated supercomputers. We demonstrate that cloud computing is a viable alternative that brings long-timescale processes within reach of a broader community. We used Google's Exacycle cloud-computing platform to simulate two milliseconds of dynamics of a major drug target, the G-protein-coupled receptor β2AR. Markov state models aggregate independent simulations into a single statistical model that is validated by previous computational and experimental results. Moreover, our models provide an atomistic description of the activation of a G-protein-coupled receptor and reveal multiple activation pathways. Agonists and inverse agonists interact differentially with these pathways, with profound implications for drug design.

TEM β-lactamase confers bacteria with resistance to many antibiotics and rapidly evolves activity against new drugs. However, functional changes are not easily explained by differences in crystal structures. We employ Markov state models to identify hidden conformations and explore their role in determining TEM’s specificity. We integrate these models with existing drug-design tools to create a new technique, called Boltzmann docking, which better predicts TEM specificity by accounting for conformational heterogeneity. Using our MSMs, we identify hidden states whose populations correlate with activity against cefotaxime. To experimentally detect our predicted hidden states, we use rapid mass spectrometric footprinting and confirm our models’ prediction that increased cefotaxime activity correlates with reduced Ω-loop flexibility. Finally, we design novel variants to stabilize the hidden cefotaximase states, and find their populations predict activity against cefotaxime in vitro and in vivo. Therefore, we expect this framework to have numerous applications in drug and protein design., Expression of TEM β-lactamase is a predominant mechanism underlying antibiotic resistance in pathogenic Gram-negative bacteria. Here, the authors use Markov state models to reveal and experimentally confirm hidden conformations that determine TEM substrate specificity.

Interleukin 15 (IL-15) and IL-2 have distinct immunological functions even though both signal through the receptor subunit IL-2Rβ and the common γ-chain (γ(c)). Here we found that in the structure of the IL-15-IL-15Rα-IL-2Rβ-γ(c) quaternary complex, IL-15 binds to IL-2Rβ and γ(c) in a heterodimer nearly indistinguishable from that of the IL-2-IL-2Rα-IL-2Rβ-γ(c) complex, despite their different receptor-binding chemistries. IL-15Rα substantially increased the affinity of IL-15 for IL-2Rβ, and this allostery was required for IL-15 trans signaling. Consistent with their identical IL-2Rβ-γ(c) dimer geometries, IL-2 and IL-15 showed similar signaling properties in lymphocytes, with any differences resulting from disparate receptor affinities. Thus, IL-15 and IL-2 induced similar signals, and the cytokine specificity of IL-2Rα versus IL-15Rα determined cellular responsiveness. Our results provide new insights for the development of specific immunotherapeutics based on IL-15 or IL-2.

Nature Chemistry 6, 15–21 (2014); published online 15 December 2013; corrected after print 24 July 2015. In the version of this Article originally published, Figure 4 displayed incorrectly drawn chemical structures for five of the ligands. The correct structures were, however, used in the calculations.

We used custom-designed resequencing arrays to generate 3.1 Mb of genomic sequence from a panel of 56 Bacillus anthracis strains. Sequence quality was shown to be very high by replication (discrepancy rate of 7.4 × 10-7) and by comparison to independently generated shotgun sequence (discrepancy rate < 2.5 × 10-6). Population genomics studies of microbial pathogens using rapid resequencing technologies such as resequencing arrays are critical for recognizing newly emerging or genetically engineered strains.