Chemical Graohs. 4. Enumeration and drocarbons. I. The Formal Graph-Theoretical Description. MATCH 1976, 2, 63-90. Entriger, R. C.; Jackson, D. E.; Snyder, D. A. Distance in Graphs, Czech. Marh. J. 1976, 26, 283-296. Skorobogatov, V. A.; Khvorostov, P. V. Analiz Metricheskikh svoistv grafov. Vychisl. Sist. 1981, 91, 1-20. Balaban, A. T.; Mekenyan, 0.; Bonchev, D. Unique Description of Chemical Structures Based on Hierarchical Ordered Extended Connectivities (HOC Procedures). I. Algorithms for Finding Graph Orbits and Canonical Numbering of Atoms. J. Compur. Chem. 1985, 6, 538-551. Skorobogatov, V. A.; Dobrynin, A. A. Metric Analysis of Graphs. MATCH 1988, 23, 105-151. Polanskv. 0. E.: Bonchev. D. The Wiener Number of Grauhs. I. Generai Theory 'and Changes Due to Graph Operations. niATCH 1987, 21, 133-186. . , Proposed Nomenclaturk'of Benzenoid Caticondensed Polycyclic Aromatic Hydrocarbons. Tetrahedron 1968,24,2505-25 16. Polansky, 0. E.; Rouvray, D. H. Graph-Theoretical Treatment of Aromatic Hy-(30) Bonchev, D.; Mekenyan, 0.; Karabunarliev, S.The chemical notation language SMILES is designed for the conversion of an arbitrarily chosen description of a chemical structure to one unique notation. This is accomplished in a two-stage algorithm, CANGEN. The first stage involves CANonicalization of structure, whereby the molecule is treated as a graph with nodes (atoms) and edges (bonds). Each atom is canonically ordered and labeled. In the second stage, starting with the lowest labeled atom, a molecular graph is GENerated, which is the unique SMILES structure.

No abstract

The ability to identify the functional correlates of structural and sequence variation in proteins is a critical capability. We related structures of influenza A N10 and N11 proteins that have no established function to structures of proteins with known function by identifying spatially conserved atoms. We identified atoms with common distributed spatial occupancy in PDB structures of N10 protein, N11 protein, an influenza A neuraminidase, an influenza B neuraminidase, and a bacterial neuraminidase. By superposing these spatially conserved atoms, we aligned the structures and associated molecules. We report spatially and sequence invariant residues in the aligned structures. Spatially invariant residues in the N6 and influenza B neuraminidase active sites were found in previously unidentified spatially equivalent sites in the N10 and N11 proteins. We found the corresponding secondary and tertiary structures of the aligned proteins to be largely identical despite significant sequence divergence. We found structural precedent in known non-neuraminidase structures for residues exhibiting structural and sequence divergence in the aligned structures. In N10 protein, we identified staphylococcal enterotoxin I-like domains. In N11 protein, we identified hepatitis E E2S-like domains, SARS spike protein-like domains, and toxin components shared by alpha-bungarotoxin, staphylococcal enterotoxin I, anthrax lethal factor, clostridium botulinum neurotoxin, and clostridium tetanus toxin. The presence of active site components common to the N6, influenza B, and S. pneumoniae neuraminidases in the N10 and N11 proteins, combined with the absence of apparent neuraminidase function, suggests that the role of neuraminidases in H17N10 and H18N11 emerging influenza A viruses may have changed. The presentation of E2S-like, SARS spike protein-like, or toxin-like domains by the N10 and N11 proteins in these emerging viruses may indicate that H17N10 and H18N11 sialidase-facilitated cell entry has been supplemented or replaced by sialidase-independent receptor binding to an expanded cell population that may include neurons and T-cells.

No abstract