TY - JOUR
T1 - Approximating the Pathway Axis and the Persistence Diagrams for a Collection of Balls in 3-Space
AU - Yaffe, Eitan
AU - Halperin, Dan
N1 - Funding Information:
This work has been supported in part by the Israel Science Foundation (grant No. 236/06), by the German–Israeli Foundation (grant No. 969/07), by the IST Programme of the EU as Shared-cost RTD (FET Open) Project under Contract No. IST-006413 (ACS—Algorithms for Complex Shapes), and by the Hermann Minkowski Minerva Center for Geometry at Tel Aviv University.A preliminary version of this paper appeared in Proceedings of the 24th ACM Symposium on Computational Geometry (SoCG), pp. 260–269, 2008.
PY - 2010
Y1 - 2010
N2 - Given a collection ℬ of balls in a three-dimensional space, we wish to explore the cavities, voids, and tunnels in the complement space of ∪ℬ. We introduce the pathway axis of ℬ as a useful subset of the medial axis of the complement of ∪ℬ and prove that it satisfies several desirable geometric properties. We present an algorithm that constructs the pathway graph of ∪ℬ, a piecewise-linear approximation of the pathway axis. At the heart of our approach is an approximation scheme that constructs a collection K of same-size balls that approximate ℬ so that the Hausdorff distance between ∪ℬ and ∪ K is bounded by a prescribed parameter. We prove a bound on the ratio between the number of balls in K and the number of balls in ℬ. We employ this bound and the approximation scheme to show how to approximate the persistence diagrams for ∪ℬ, which can be used to extract major topological features such as the large voids and tunnels in the complement of ∪ℬ. We show that our approach is superior in terms of complexity to the standard point-sample approaches for the two problems that we address in this paper: approximating the pathway axis of ℬ and approximating the persistence diagrams for ∪ℬ. In a companion paper we introduce MolAxis, a tool for the identification of channels in macromolecules that demonstrates how the pathway graph and the persistence diagrams are used to identify plausible pathways in the complement of molecules.
AB - Given a collection ℬ of balls in a three-dimensional space, we wish to explore the cavities, voids, and tunnels in the complement space of ∪ℬ. We introduce the pathway axis of ℬ as a useful subset of the medial axis of the complement of ∪ℬ and prove that it satisfies several desirable geometric properties. We present an algorithm that constructs the pathway graph of ∪ℬ, a piecewise-linear approximation of the pathway axis. At the heart of our approach is an approximation scheme that constructs a collection K of same-size balls that approximate ℬ so that the Hausdorff distance between ∪ℬ and ∪ K is bounded by a prescribed parameter. We prove a bound on the ratio between the number of balls in K and the number of balls in ℬ. We employ this bound and the approximation scheme to show how to approximate the persistence diagrams for ∪ℬ, which can be used to extract major topological features such as the large voids and tunnels in the complement of ∪ℬ. We show that our approach is superior in terms of complexity to the standard point-sample approaches for the two problems that we address in this paper: approximating the pathway axis of ℬ and approximating the persistence diagrams for ∪ℬ. In a companion paper we introduce MolAxis, a tool for the identification of channels in macromolecules that demonstrates how the pathway graph and the persistence diagrams are used to identify plausible pathways in the complement of molecules.
UR - http://www.scopus.com/inward/record.url?scp=77955773185&partnerID=8YFLogxK
U2 - 10.1007/s00454-009-9240-9
DO - 10.1007/s00454-009-9240-9
M3 - ???researchoutput.researchoutputtypes.contributiontojournal.article???
AN - SCOPUS:77955773185
VL - 44
SP - 660
EP - 685
JO - Discrete and Computational Geometry
JF - Discrete and Computational Geometry
SN - 0179-5376
IS - 3
ER -