It is widely acknowledged that encoding methods play a fundamental role in the field of protein secondary structure prediction, their task being to transform the available biological information in a form directly usable by the under- lying predictor. This transformation is particularly critical, as the relationship between primary and secondary structure is very subtle and difficult to capture. In this paper we compare three different encoding methods and introduce a new encoding which we show to be superior. Experiments have been per- formed with a software architecture devised to guarantee the statistical significance of the results. The current release of the predictor is freely usable through the online web interface at http://iasc2.diee.unica.it/ssp. The corresponding stand alone application (together with the data sets used for benchmarking purposes), as well as the source listing (in Java) of the GAME generic architecture used to implement the predictor, can be downloaded from the main page of the web interface. The soft- ware provided in source format can be freely distributed un- der the GPL license. The proposed method has been compared with other state-of-the-art encoding methods, and experimental results confirm its superiority. In particular, we obtained an improvement that ranges from 0.5 to 1.5%, measured both by Q3 and SOV performance indexes.

Sum-Linear Blosum: A Novel Protein - Encoding Method for Secondary Structure Prediction

ARMANO, GIULIANO;VARGIU, ELOISA
2009-01-01

Abstract

It is widely acknowledged that encoding methods play a fundamental role in the field of protein secondary structure prediction, their task being to transform the available biological information in a form directly usable by the under- lying predictor. This transformation is particularly critical, as the relationship between primary and secondary structure is very subtle and difficult to capture. In this paper we compare three different encoding methods and introduce a new encoding which we show to be superior. Experiments have been per- formed with a software architecture devised to guarantee the statistical significance of the results. The current release of the predictor is freely usable through the online web interface at http://iasc2.diee.unica.it/ssp. The corresponding stand alone application (together with the data sets used for benchmarking purposes), as well as the source listing (in Java) of the GAME generic architecture used to implement the predictor, can be downloaded from the main page of the web interface. The soft- ware provided in source format can be freely distributed un- der the GPL license. The proposed method has been compared with other state-of-the-art encoding methods, and experimental results confirm its superiority. In particular, we obtained an improvement that ranges from 0.5 to 1.5%, measured both by Q3 and SOV performance indexes.
File in questo prodotto:
Non ci sono file associati a questo prodotto.

I documenti in IRIS sono protetti da copyright e tutti i diritti sono riservati, salvo diversa indicazione.

Utilizza questo identificativo per citare o creare un link a questo documento: https://hdl.handle.net/11584/101766
 Attenzione

Attenzione! I dati visualizzati non sono stati sottoposti a validazione da parte dell'ateneo

Citazioni
  • ???jsp.display-item.citation.pmc??? ND
  • Scopus ND
  • ???jsp.display-item.citation.isi??? ND
social impact