We investigate the Minimum Evolution Problem (MEP), an NP-hard network design problem arising from computational biology. The MEP consists in finding a weighted unrooted binary tree having n leaves, minimal length, and such that the sum of the edge weights belonging to the unique path between each pair of leaves is greater than or equal to a prescribed value. We study the polyhedral combinatorics of the MEP and investigate its relationships with the Balanced Minimum Evolution Problem. We develop an exact solution approach for the MEP based on a nontrivial combination of a parallel branch-price-and-cut scheme and a non-isomorphic enumeration of all possible solutions to the problem. Computational experiments show that the new solution approach outperforms the best mixed integer linear programming formulation for the MEP currently described in the literature. Our results give a perspective on the combinatorics of the MEP and suggest new directions for the development of future exact solution approaches that may turn out useful in practical applications. We also show that the MEP is statistically consistent.
Abstract We investigate the Minimum Evolution Problem (MEP), an NP-hard network design problem arising from computational biology. The MEP consists in finding a weighted unrooted binary tree having n leaves, minimal length, and such that the sum of the edge weights belonging to the unique path between each pair of leaves is greater than or equal to a prescribed value. We study the polyhedral combinatorics of the MEP and investigate its relationships with the Balanced Minimum Evolution Problem. We develop an exact solution approach for the MEP based on a nontrivial combination of a parallel branch-price-and-cut scheme and a non-isomorphic enumeration of all possible solutions to the problem. Computational experiments show that the new solution approach outperforms the best mixed integer linear programming formulation for the MEP currently described in the literature. Our results give a perspective on the combinatorics of the MEP and suggest new directions for the development of future exact solution approaches that may turn out useful in practical applications. We also show that the MEP is statistically consistent.
A branch-price-and-cut algorithm for the minimum evolution problem
PESENTI, Raffaele
2015-01-01
Abstract
Abstract We investigate the Minimum Evolution Problem (MEP), an NP-hard network design problem arising from computational biology. The MEP consists in finding a weighted unrooted binary tree having n leaves, minimal length, and such that the sum of the edge weights belonging to the unique path between each pair of leaves is greater than or equal to a prescribed value. We study the polyhedral combinatorics of the MEP and investigate its relationships with the Balanced Minimum Evolution Problem. We develop an exact solution approach for the MEP based on a nontrivial combination of a parallel branch-price-and-cut scheme and a non-isomorphic enumeration of all possible solutions to the problem. Computational experiments show that the new solution approach outperforms the best mixed integer linear programming formulation for the MEP currently described in the literature. Our results give a perspective on the combinatorics of the MEP and suggest new directions for the development of future exact solution approaches that may turn out useful in practical applications. We also show that the MEP is statistically consistent.File | Dimensione | Formato | |
---|---|---|---|
manuscript.pdf
non disponibili
Tipologia:
Documento in Pre-print
Licenza:
Accesso chiuso-personale
Dimensione
2.37 MB
Formato
Adobe PDF
|
2.37 MB | Adobe PDF | Visualizza/Apri |
I documenti in ARCA sono protetti da copyright e tutti i diritti sono riservati, salvo diversa indicazione.