Protein model quality assessment using rotation-equivariant transformations on point clouds.

Machine learning research concerning protein structure has seen a surge in popularity over the last years with promising advances for basic science and drug discovery. Working with macromolecular structure in a machine learning context requires an adequate numerical representation, and researchers have extensively studied representations such as graphs, discretized 3D grids, and distance maps. As part of CASP14, we explored a new and conceptually simple representation in a blind experiment: atoms as points in 3D, each with associated features. These features-initially just the basic element type of each atom-are updated through a series of neural network layers featuring rotation-equivariant convolutions. Starting from all atoms, we further aggregate information at the level of alpha carbons before making a prediction at the level of the entire protein structure. We find that this approach yields competitive results in protein model quality assessment despite its simplicity and despite the fact that it incorporates minimal prior information and is trained on relatively little data. Its performance and generality are particularly noteworthy in an era where highly complex, customized machine learning methods such as AlphaFold 2 have come to dominate protein structure prediction.

Geometric deep learning of RNA structure.

RNA molecules adopt three-dimensional structures that are critical to their function and of interest in drug discovery. Few RNA structures are known, however, and predicting them computationally has proven challenging. We introduce a machine learning approach that enables identification of accurate structural models without assumptions about their defining characteristics, despite being trained with only 18 known RNA structures. The resulting scoring function, the Atomic Rotationally Equivariant Scorer (ARES), substantially outperforms previous methods and consistently produces the best results in community-wide blind RNA structure prediction challenges. By learning effectively even from a small amount of data, our approach overcomes a major limitation of standard deep neural networks. Because it uses only atomic coordinates as inputs and incorporates no RNA-specific information, this approach is applicable to diverse problems in structural biology, chemistry, materials science, and beyond.

Gold nanoparticles and tilt pairs to assess protein flexibility by cryo-electron microscopy.

A computational method was developed to recover the three-dimensional coordinates of gold nanoparticles specifically attached to a protein complex from tilt-pair images collected by electron microscopy. The program was tested on a simulated dataset and applied to a real dataset comprising tilt-pair images recorded by cryo electron microscopy of RNA polymerase II in a complex with four gold-labeled single-chain antibody fragments. The positions of the gold nanoparticles were determined, and comparison of the coordinates among the tetrameric particles revealed the range of motion within the protein complexes.

Hierarchical, rotation-equivariant neural networks to select structural models of protein complexes.

Predicting the structure of multi-protein complexes is a grand challenge in biochemistry, with major implications for basic science and drug discovery. Computational structure prediction methods generally leverage predefined structural features to distinguish accurate structural models from less accurate ones. This raises the question of whether it is possible to learn characteristics of accurate models directly from atomic coordinates of protein complexes, with no prior assumptions. Here we introduce a machine learning method that learns directly from the 3D positions of all atoms to identify accurate models of protein complexes, without using any precomputed physics-inspired or statistical terms. Our neural network architecture combines multiple ingredients that together enable end-to-end learning from molecular structures containing tens of thousands of atoms: a point-based representation of atoms, equivariance with respect to rotation and translation, local convolutions, and hierarchical subsampling operations. When used in combination with previously developed scoring functions, our network substantially improves the identification of accurate structural models among a large set of possible models. Our network can also be used to predict the accuracy of a given structural model in absolute terms. The architecture we present is readily applicable to other tasks involving learning on 3D structures of large atomic systems.

How GPCR Phosphorylation Patterns Orchestrate Arrestin-Mediated Signaling.

Binding of arrestin to phosphorylated G-protein-coupled receptors (GPCRs) controls many aspects of cell signaling. The number and arrangement of phosphates may vary substantially for a given GPCR, and different phosphorylation patterns trigger different arrestin-mediated effects. Here, we determine how GPCR phosphorylation influences arrestin behavior by using atomic-level simulations and site-directed spectroscopy to reveal the effects of phosphorylation patterns on arrestin binding and conformation. We find that patterns favoring binding differ from those favoring activation-associated conformational change. Both binding and conformation depend more on arrangement of phosphates than on their total number, with phosphorylation at different positions sometimes exerting opposite effects. Phosphorylation patterns selectively favor a wide variety of arrestin conformations, differently affecting arrestin sites implicated in scaffolding distinct signaling proteins. We also reveal molecular mechanisms of these phenomena. Our work reveals the structural basis for the long-standing "barcode" hypothesis and has important implications for design of functionally selective GPCR-targeted drugs.

Molecular mechanism of GPCR-mediated arrestin activation.

Despite intense interest in discovering drugs that cause G-protein-coupled receptors (GPCRs) to selectively stimulate or block arrestin signalling, the structural mechanism of receptor-mediated arrestin activation remains unclear1,2. Here we reveal this mechanism through extensive atomic-level simulations of arrestin. We find that the receptor's transmembrane core and cytoplasmic tail-which bind distinct surfaces on arrestin-can each independently stimulate arrestin activation. We confirm this unanticipated role of the receptor core, and the allosteric coupling between these distant surfaces of arrestin, using site-directed fluorescence spectroscopy. The effect of the receptor core on arrestin conformation is mediated primarily by interactions of the intracellular loops of the receptor with the arrestin body, rather than the marked finger-loop rearrangement that is observed upon receptor binding. In the absence of a receptor, arrestin frequently adopts active conformations when its own C-terminal tail is disengaged, which may explain why certain arrestins remain active long after receptor dissociation. Our results, which suggest that diverse receptor binding modes can activate arrestin, provide a structural foundation for the design of functionally selective ('biased') GPCR-targeted ligands with desired effects on arrestin signalling.