Top-Down Hierarchical Ensembles of Classifiers for Predicting G-Protein-Coupled-Receptor Functions

Abstract

Abstract. Despite the recent advances in Molecular Biology, the function of a large amount of proteins is still unknown. An approach that can be used in the prediction of a protein function consists of searching against secondary databases, also known as signature databases. Different strategies can be applied to use protein signatures in the prediction of function of proteins. A sophisticated approach consists of inducing a classification model for this prediction. This paper applies five hierarchical classification methods based on the standard Top-Down approach and one hierarchical classification method based on a new approach named Top-Down Ensembles -based on the hierarchical combination of classifiers -to three different protein functional classification datasets that employ protein signatures. The algorithm based on the Top-Down Ensembles approach presented slightly better results than the other algorithms, indicating that combinations of classifiers can improve the performance of hierarchical classification models.

A survey of hierarchical classification across different application domains

“…the MIPS FunCat (Ruepp et al 2004). These different hierarchies have been used by different authors (Clare and King 2003;Kriegel et al 2004;Wu et al 2005;Blockeel et al 2006;Rousu et al 2006;Alves et al 2008;Guan et al 2008;Costa et al 2008;Vens et al 2008;Otero et al 2009). Burred and Lerch (2003) 8.3 Music genre classification…”

“…, Sun and Lim (2001), Mladenic and Grobelnik (2003), Sun et al (2003Sun et al ( , 2004, Wu et al (2005), Cesa-Bianchi et al (2006a,b), Esuli et al (2008), Jin et al (2008), Punera and Ghosh (2008), Xue et al (2008) and Bennett and Nguyen (2009) LCPN Koller and Sahami (1997), Chakrabarti et al (1998), McCallum et al (1998, Weigend et al (1999), D'Alessio et al (2000), Dumais and Chen (2000), Ruiz and Srinivasan (2002), Tikk and Biró (2003), Tikk et al ( , 2007, Kriegel et al (2004) and Gauch et al (2009) GC Labrou and Finin (1999), Wang et al (1999Wang et al ( , 2001, Dekel et al (2004a), Hofmann (2004, 2007), Rousu et al (2005Rousu et al ( , 2006, Kiritchenko et al (2005Kiritchenko et al ( , 2006, Peng and Choi (2005) Holden and Freitas (2005, 2009), Secker et al (2007, 2010, Costa et al (2008) and Kriegel et al (2004) Acknowledgements The first author is financially supported by CAPES-a Brazilian research-support agency (process number 4871-06-5). We also thank the anonymous reviewers for their insightful feedback on the earlier version of this manuscript.…”

Neural Networks for Hierarchical Classification of G-Protein Coupled Receptors

“…In [1], the authors proposed an ensemble of classifiers using the LCPN strategy. The idea was to have a hierarchy made of different flat classifiers in order to improve the final performance.…”

“…Proteins are present in the physiological metabolic pathways of cells, and perform almost all functions related to cell activity, such as biochemical reactions, cell signaling, structural, and mechanical functions [1].…”

Hierarchical classification of Gene Ontology-based protein functions with neural networks

“…Proteins are related to biochemical reac tions, cell signaling, structural, and mechanical functions [1], to name a few. Given that protein functions are hierarchically related, PFP can be considered a typical HMC problem.…”