An ontology driven architecture for derived representations of macromolecular structure

SUMMARY: An object metamodel based on a standard scientific ontology has been developed and used to generate a CORBA interface, an SQL schema and an XML representation for macromolecular structure (MMS) data. In addition to the interface and schema definitions, the metamodel was also used to generate the core elements of a CORBA reference server and a JDBC database loader. The Java source code which implements this metamodel, the CORBA server, database loader and XML converter along with detailed documentation and code examples are available as part of the OpenMMS toolkit. AVAILABILITY: http://openmms.sdsc.edu CONTACT: dsg@sdsc.edu     

Compressed representations of macromolecular structures and properties

We introduce a new and unified, compressed volumetric representation for macromolecular structures at varying feature resolutions, as well as for many computed associated properties. Important caveats of this compressed representation are fast random data access and decompression operations. Many computational tasks for manipulating large structures, including those requiring interactivity such as real-time visualization, are greatly enhanced by utilizing this compact representation. The compression scheme is obtained by using a custom designed hierarchical wavelet basis construction. Due to the continuity offered by these wavelets, we retain very good accuracy of molecular surfaces, at very high compression ratios, for macromolecular structures at multiple resolutions.     

New determinations and simplified representations of macromolecular surfaces

Several new methods or improvements of older algorithms determining the different pieces of molecular surface are presented. Their improvement in time and their complexity are discussed. Only the indexes of the atoms on which the pieces are relying are, in fact, determined, since their explicit representation from these numbers varies according to the 3D capabilities of the graphic workstation (dots, grid, etc.), and this generation is not C.P.U. consuming. To have a simplified representation of the surface of macromolecules, a polyhedron with planar triangular faces is then introduced: Each concave triangular surface piece is replaced with planar triangles relying on its three atomic centers, while saddle-shaped rectangles and convex pieces are wholly ignored. A minimal data structure of the polyhedron is then proposed, which contains only topological informations, since no coordinates have been generated. If the atomic radius is then considered to be constant (independent of atomic type), the surface of a set of N points is now defined by the choice of a subset with a topology. This choice is controlled by a parameter of rugosity (the atomic radius). Contrary to Voronoi polyhedrons partition, which gives a topology for a set of N points, our approach gives a topology only for the exterior points of this set. A few applications of this very simple definition of molecular surface are then discussed: the 3D interactive manipulation of macromolecules, the steric intermolecular recognition, and the determination of local and global properties of the surface.     

BioEditor-simplifying macromolecular structure annotation

SUMMARY: BioEditor is an application to enable scientists and educators to prepare and present structure annotations containing formatted text, graphics, sequence data, and interactive molecular views. It is intended to bridge the gap between printed journal articles and Internet presentation formats. BioEditor is relevant in the era of structural genomics, where annotation and publication could become the rate determining step in structure determination. AVAILABILITY: BioEditor is available at http://bioeditor.sdsc.edu. The Web site includes the latest version of the software for Microsoft Windows, including documentation, the opportunity to submit bug reports and suggestions, example documentaries prepared with BioEditor and a repository where users can submit documentaries for posting to the site.     

Consciousness and the structure of neuronal representations.

The hypothesis is defended that brains expressing phenomenal awareness are capable of generating metarepresentations of their cognitive processes, these metarepresentations resulting from an iteration of self-similar cortical operations. Search for the neuronal substrate of awareness therefore converges with the search for the nature of neuronal representations. It is proposed that evolved brains use two complementary representational strategies. One consists of the generation of neurons responding selectively to a particular constellation of features and is based on selective recombination of inputs in hierarchically structured feedforward architectures. The other relies on the dynamic association of feature-specific cells into functionally coherent cell assemblies that, as a whole, represent the constellation of features defining a particular perceptual object. Arguments are presented that favour the notion that the metarepresentations supporting awareness are established in accordance with the second strategy. Experimental data are reviewed that are compatible with the hypothesis that evolved brains use assembly codes for the representation of contents and that these assemblies become organized through transient synchronization of the discharges of associated neurons. It is argued that central states favouring the formation of assembly-based representations are similar to those favouring awareness.     

Confinement as a determinant of macromolecular structure and reactivity.

The confinement of macromolecules within enclosures or "pores" of comparable dimensions results in significant size- and shape-dependent alterations of macromolecular chemical potential and reactivity. Calculations of the magnitude of this effect for model particles of different shapes in model enclosures of different shapes were carried out using hard particle partition theory developed by Giddings et al. (J. Phys. Chem. 1968. 72:4397-4408). Results obtained indicate that the equilibrium constants of reactions, such as isomerization, self-association, and site binding, that result in significant change in macromolecular size, shape, and/or mobility may be altered within pores by as much as several orders of magnitude relative to the value in the unbounded or bulk phase. Confinement also produces a substantial size-dependent outward force on the walls of an enclosure. These results are likely to be important within the fluid phase of biological media, such as the cytoplasm of eukaryotic cells, containing significant volume fractions of large fibrous structures (e.g., the cytomatrix).     

Analysis of macromolecular structure and dynamics by electron cryo-microscopy.

Electron cryo-microscopy has yielded a wealth of detailed new information on structures of biological macromolecules ranging from alphabeta-tubulin at 3.7 A resolution to hepatitis B virus at 7.4 A resolution, as well as a number of membrane proteins at 6-8 A resolution. Much of this progress was made possible by recent advances in instrumentation and image processing techniques.     

Pulsed electron paramagnetic resonance methods for macromolecular structure determination.

Pulsed electron paramagnetic resonance (EPR) distance measurement techniques target macromolecular structure elucidation at both the local and global level. Recent developments in pulse microwave technology and high-field EPR have led to the development of a variety of pulsed EPR distance measurement techniques. These methods have emerged as powerful tools for the determination of structure/function relationships in macromolecular systems. In this review article, we discuss recent applications of long-range and short-range EPR distance measurements.     

Single particle macromolecular structure determination via electron microscopy

Three-dimensional structure determination of macromolecules and macromolecular complexes is an integral part of understanding biological functions. For large protein and macromolecular complexes structure determination is often performed using electron cryomicroscopy where projection images of individual macromolecular complexes are combined to produce a three-dimensional reconstruction. Single particle methods have been devised to perform this structure determination for macromolecular complexes with little or no underlying symmetry. These computational methods generally involve an iterative process of aligning unique views of the macromolecular images followed by determination of the angular components that define those views. In this review, this structure determination process is described with the aim of clarifying a seemingly complex structural method.     

Tissue structure and macromolecular diffusion in umbilical cord. Immobilization of endogenous hyaluronic acid

Diffusion of endogenous hyaluronic acid and 125I-labelled albumin, monitored by desorption from umbilical cord (Wharton's jelly) slices, was studied in relation to tissue structure. Diffusion of hyaluronic acid was Fickian and some two orders of magnitude slower than that in free solution. After treatment of tissue with trypsin which removes proteoglycan(s) and degrades glycoprotein microfibrils, hyaluronic acid mobility through the collagen fibril network that remains is increased by an order of magnitude. These findings indicate that the mobility of hyaluronic acid in tissue is reduced both by the collagen network and by the presence of proteoglycan(s) and/or microfibrils. Estimates of the reduction in mobility due to physical entanglements with the fibrillar networks show that these play a major role. The mobility of hyaluronic acid found for intact tissue is sufficient for it to permeate the extracellular space within its metabolic turnover time. Labelled albumin diffusion is intact tissue, on the other hand, is reduced by only some 30% relative to free solution. This is consistent with the approximate 10% reduction found for the polysaccharide-free tissue (given by the excluded volume fraction) and the approximate 20% reduction expected for the polysaccharides in the interstitial fluid. Similar effects appear to be involved in the mobility of endogenous diffusible proteins in tissue.     

Teaching macromolecular modeling.

Training newcomers to the field of macromolecular modeling is as difficult as is training beginners in x-ray crystallography, nuclear magnetic resonance, or other methods in structural biology. In one or two lectures, the most that can be conveyed is a general sense of the relationship between modeling and other structural methods. If a full semester is available, then students can be taught how molecular structures are built, manipulated, refined, and analyzed on a computer. Here we describe a one-semester modeling course that combines lectures, discussions, and a laboratory using a commercial modeling package. In the laboratory, students carry out prescribed exercises that are coordinated to the lectures, and they complete a term project on a modeling problem of their choice. The goal is to give students an understanding of what kinds of problems can be attacked by molecular modeling methods and which problems are beyond the current capabilities of those methods.     

Comparing graph representations of protein structure for mining family-specific residue-based packing motifs.

We find recurring amino-acid residue packing patterns, or spatial motifs, that are characteristic of protein structural families, by applying a novel frequent subgraph mining algorithm to graph representations of protein three-dimensional structure. Graph nodes represent amino acids, and edges are chosen in one of three ways: first, using a threshold for contact distance between residues; second, using Delaunay tessellation; and third, using the recently developed almost-Delaunay edges. For a set of graphs representing a protein family from the Structural Classification of Proteins (SCOP) database, subgraph mining typically identifies several hundred common subgraphs corresponding to spatial motifs that are frequently found in proteins in the family but rarely found outside of it. We find that some of the large motifs map onto known functional regions in two protein families explored in this study, i.e., serine proteases and kinases. We find that graphs based on almost-Delaunay edges significantly reduce the number of edges in the graph representation and hence present computational advantage, yet the patterns extracted from such graphs have a biological interpretation approximately equivalent to that of those extracted from distance based graphs.