A Platform Based on the Multi-dimensional Data Model for Analysis of Bio-Molecular Structures

摘要

A platform called AnMol for supporting analytical applications over structural data of large biomolecules is described. The term "biomolecular structure" has various connotations and different representations. AnMol reduces these representations into graph structures. Each of these graphs are then stored as one or more vectors in a database. Vectors encapsulate structural features of these graphs. Structural queries like similarity and substructure are transformed into spatial constructs like distance and containment within regions. Query results are based on inexact matches. A refinement mechanism is supported for increasing accuracy of the results. Design and implementation issues of AnMol including schema structure and performance results are discussed in this paper.