Genomics | Bioinformatics | Protein Folding | Structural Genomics
On the complementarity of the consensus-based disorder prediction
Zhenling Peng, Lukasz Kurgan*
*Corresponding author: Lukasz Kurgan
Electrical & Computer Engineering Department, University of Alberta, Edmonton, Canada
F1000Posters 2012, 3: 40 (poster) [English]
Poster [12.65 MB]
Pacific Symposium on Biocomputing (PSB) 2012, 3 - 7 Jan 2012, 50
Recent research shows that consensus-based disorder predictors provide a viable way to improve the disorder prediction (1); however, the selection of the base predictors for a given consensus is usually performed in an ad-hock manner. Thus, we investigate the complementarity/dissimilarity among some recent disorder predictors, and propose a regression-based model to quantify the quality of the majority-vote consensus of a given triplet of predictors, based on their individual predictive performance and their complementarity measured at disorder residue and disorder segment levels.
- Improved predictive quality of a consensus is associated with the higher accuracy of base predictors, their higher similarity at the residue level, stronger complementarity (lower similarity) at the segment level, and stronger complementarity between the base and the consensus predictors.
- It is also beneficial to use different types of base predictors to build the consensus.
Our work provides useful clues to develop a new generation of accurate consensus-based disorder predictors.
No relevant competing interests disclosed.
Please note that most posters on this site present work that is preliminary in nature and has not been peer reviewed.
This poster is open access subject to the CC BY-NC Creative Commons 3.0 License