Mining a Sub-Matrix of Maximal Sum

TL;DR

This paper introduces new algorithms for efficiently finding a sub-matrix with the maximum sum in large data matrices, improving upon previous methods and demonstrating their effectiveness on gene expression data.

Contribution

It proposes two novel algorithms, a CP approach with a global constraint and an MILP formulation, for solving the NP-hard max-sum sub-matrix problem more efficiently.

Findings

CPGC is the fastest method for large problems

MILP is easier to formulate and competitive

Both methods outperform the previous CP-LNS approach

Abstract

Biclustering techniques have been widely used to identify homogeneous subgroups within large data matrices, such as subsets of genes similarly expressed across subsets of patients. Mining a max-sum sub-matrix is a related but distinct problem for which one looks for a (non-necessarily contiguous) rectangular sub-matrix with a maximal sum of its entries. Le Van et al. (Ranked Tiling, 2014) already illustrated its applicability to gene expression analysis and addressed it with a constraint programming (CP) approach combined with large neighborhood search (CP-LNS). In this work, we exhibit some key properties of this NP-hard problem and define a bounding function such that larger problems can be solved in reasonable time. Two different algorithms are proposed in order to exploit the highlighted characteristics of the problem: a CP approach with a global constraint (CPGC) and mixed integer linear programming (MILP). Practical experiments conducted both on synthetic and real gene expression data exhibit the characteristics of these approaches and their relative benefits over the original CP-LNS method. Overall, the CPGC approach tends to be the fastest to produce a good solution. Yet, the MILP formulation is arguably the easiest to formulate and can also be competitive.