id: 06095266
dt: a
an: 06095266
au: Vanneschi, Leonardo; Mondini, Matteo; Bertoni, Martino; Ronchi, Alberto;
    Stefano, Mattia
ti: GeNet: a graph-based genetic programming framework for the reverse
    engineering of gene regulatory networks.
so: Giacobini, Mario (ed.) et al., Evolutionary computation, machine learning
    and data mining in bioinformatics. 10th European conference, EvoBIO
    2012, Málaga, Spain, April 11‒13, 2012. Proceedings. Berlin:
    Springer (ISBN 978-3-642-29065-7/pbk). Lecture Notes in Computer
    Science 7246, 97-109 (2012).
py: 2012
pu: Berlin: Springer
la: EN
cc: 
ut: 
ci: 
li: doi:10.1007/978-3-642-29066-4_9
ab: Summary: A standard tree-based genetic programming system, called GRNGen,
    for the reverse engineering of gene regulatory networks starting from
    time series datasets, was proposed in EvoBIO 2011. Despite the
    interesting results obtained on the simple IRMA network, GRNGen has
    some important limitations. For instance, in order to reconstruct a
    network with GRNGen, one single regression problem has to be solved by
    GP for each gene. This entails a clear limitation on the size of the
    networks that it can reconstruct, and this limitation is crucial, given
    that real genetic networks generally contain large numbers of genes. In
    this paper we present a new system, called GeNet, which aims at
    overcoming the main limitations of GRNGen, by directly evolving entire
    networks using graph-based genetic programming. We show that GeNet
    finds results that are comparable, and in some cases even better, than
    GRNGen on the small IRMA network, but, even more importantly (and
    contrarily to GRNGen), it can be applied also to larger networks. Last
    but not least, we show that the time series datasets found in
    literature do not contain a sufficient amount of information to
    describe the IRMA network in detail.
rv: