Protein structure prediction with visuospatial analogy
We show that visuospatial representations and reasoning techniques can be used as a similarity metric for analogical protein structure prediction. Our system retrieves pairs of α-helices based on contact map similarity, then transfers and adapts the structure information to an unknown helix pair, showing that similar protein contact maps predict similar 3D protein structure. The success of this method provides support for the notion that changing representations can enable similarity metrics in analogy.
|Series||Lecture Notes in Computer Science|
Davies, J, Glasgow, J. (Janice), & Kuo, T. (Tony). (2007). Protein structure prediction with visuospatial analogy. In Lecture Notes in Computer Science.