DeepMind presented remarkably accurate predictions at the recent CASP14 protein structure prediction assessment conference. We explored network architectures incorporating related ideas and obtained the best performance with a three-track network in which information at the 1D sequence level, the 2D distance map level, and the 3D coordinate level is successively transformed and integrated. The three-track network produces structure predictions with accuracies approaching those of DeepMind in CASP14, enables the rapid solution of challenging X-ray crystallography and cryo-EM structure modeling problems, and provides insights into the functions of proteins of currently unknown structure. The network also enables rapid generation of accurate protein-protein complex models from sequence information alone, short circuiting traditional approaches which require modeling of individual subunits followed by docking. We make the method available to the scientific community to speed biological research.

Understanding the evolution of a protein, including both close and distant relationships, often reveals insight into its structure and function. Fast and easy access to such up-to-date information facilitates research. We have developed a hierarchical evolutionary classification of all proteins with experimentally determined spatial structures, and presented it as an interactive and updatable online database. ECOD (Evolutionary Classification of protein Domains) is distinct from other structural classifications in that it groups domains primarily by evolutionary relationships (homology), rather than topology (or “fold”). This distinction highlights cases of homology between domains of differing topology to aid in understanding of protein structure evolution. ECOD uniquely emphasizes distantly related homologs that are difficult to detect, and thus catalogs the largest number of evolutionary links among structural domain classifications. Placing distant homologs together underscores the ancestral similarities of these proteins and draws attention to the most important regions of sequence and structure, as well as conserved functional sites. ECOD also recognizes closer sequence-based relationships between protein domains. Currently, approximately 100,000 protein structures are classified in ECOD into 9,000 sequence families clustered into close to 2,000 evolutionary groups. The classification is assisted by an automated pipeline that quickly and consistently classifies weekly releases of PDB structures and allows for continual updates. This synchronization with PDB uniquely distinguishes ECOD among all protein classifications. Finally, we present several case studies of homologous proteins not recorded in other classifications, illustrating the potential of how ECOD can be used to further biological and evolutionary studies.

The goal of Dynameomics is to perform atomistic molecular dynamics (MD) simulations of representative proteins from all known folds in explicit water in their native state and along their thermal unfolding pathways. Here we present 188-fold representatives and their native state simulations and analyses. These 188 targets represent 67% of all the structures in the Protein Data Bank. The behavior of several specific targets is highlighted to illustrate general properties in the full dataset and to demonstrate the role of MD in understanding protein function and stability. As an example of what can be learned from mining the Dynameomics database, we identified a protein fold with heightened localized dynamics. In one member of this fold family, the motion affects the exposure of its phosphorylation site and acts as an entropy sink to offset another portion of the protein that is relatively immobile in order to present a consistent interface for protein docking. In another member of this family, a polymorphism in the highly mobile region leads to a host of disease phenotypes. We have constructed a web site to provide access to a novel hybrid relational/multidimensional database (described in the succeeding two papers) to view and interrogate simulations of the top 30 targets: http://www.dynameomics.org. The Dynameomics database, currently the largest collection of protein simulations and protein structures in the world, should also be useful for determining the rules governing protein folding and kinetic stability, which should aid in deciphering genomic information and for protein engineering and design.

Summary The dynamic behavior of proteins is important for an understanding of their function and folding. We have performed molecular dynamics simulations of the native state and unfolding pathways of over 1000 proteins, representing the majority of folds in globular proteins. These data are stored and organized using an innovative database approach, which can be mined to obtain both general and specific information about the dynamics and folding/unfolding of proteins, relevant subsets thereof, and individual proteins. Here we describe the project in general terms and the type of information contained in the database. Then we provide examples of mining the database for information relevant to protein folding, structure building, the effect of single-nucleotide polymorphisms, and drug design. The native state simulation data and corresponding analyses for the 100 most populated metafolds, together with related resources, are publicly accessible through www.dynameomics.org.