Combinatorial optimisation models for analysing biological data sets — ASN Events
  1. Schedule
  2. Plenary 7
  3. Combinatorial optimisation models for analysing biological data sets

Combinatorial optimisation models for analysing biological data sets (#100)

Regina Berretta 1
  1. The University of Newcastle, Callaghan, NSW, Australia

This talk will present combinatorial optimisation models and algorithmic techniques that have been developed to analyse large datasets.

First, the presentation will focus on an approach, based on a combinatorial optimisation problem (called the (α,β)-k-Feature Set Problem) to deal with the problem of selecting groups of features, such as genes, that discriminate between different existing classes. We will illustrate the application of these models using different variations of the model in several datasets.

Next, the presentation will illustrate how a classical and well-known combinatorial optimisation problem; the Quadratic Assignment Problem (QAP), is employed as a mathematical model to produce a visualization of a data set, based on the relationships between the elements in the data set. The visualization method can also incorporate the results of a clustering algorithm to facilitate the process of data analysis.