The Computational Complexity of Orientation Search Problems in Cryo-Electron Microscopy

Abstract

In this report we study the problem of determining three-dimensional orientations for noisy projections of randomly oriented identical particles. The problem is of central importance in the tomographic reconstruction of the density map of macromolecular complexes from electron microscope images and it has been studied intensively for more than 30 years. We analyze the computational complexity of the orientation problem and show that while several variants of the problem are $NP$-hard, inapproximable and fixed-parameter intractable, some restrictions are polynomial-time approximable within a constant factor or even solvable in logarithmic space. The orientation search problem is formalized as a constrained line arrangement problem that is of independent interest. The negative complexity results give a partial justification for the heuristic methods used in orientation search, and the positive complexity results on the orientation search have some positive implications also to the problem of finding functionally analogous genes. A preliminary version ``The Computational Complexity of Orientation Search in Cryo-Electron Microscopy'' appeared in Proc. ICCS 2004, LNCS 3036, pp. 231--238. Springer-Verlag 2004.