Skip to Main content Skip to Navigation
Conference papers

"Hybrid Protein Model": a new clustering approach for 3D local structures: Evolution of Hybrid Protein Model

Abstract : The "Hybrid Protein Model" (HPM) is a fuzzy model for compacting biological data series such as 3D structures into a limited number of overlapping cluster. The 3D structures of a non-redundant structural databank is encoded in a previously defined structural alphabet composed of 16 protein blocks (PBs) [1]. The hybrid protein is composed of a series of distributions of the probability of observing the PBs. The training is an iterative unsupervised process that for every fold to be learn consists of looking for the most similar pattern present in the hybrid protein and modifying it slightly. Finally each position of the hybrid protein corresponds to a set of similar local structures. Compared with conventional clustering and the definition of a partition into independent subsets, the hybrid protein characterizes a series of structurally dependent subsets: that is, it maintains the sequentiality of the local structures. In a second step, to define the optimal length of the hybrid protein, we have defined an new approach based on the deletion of the redundant parts of the hybrid protein.
Complete list of metadatas

https://www.hal.inserm.fr/inserm-00134559
Contributor : Alexandre G. de Brevern <>
Submitted on : Friday, March 2, 2007 - 3:47:16 PM
Last modification on : Friday, March 27, 2020 - 2:58:21 AM
Long-term archiving on: : Wednesday, April 7, 2010 - 1:18:51 AM

Identifiers

  • HAL Id : inserm-00134559, version 1

Collections

Citation

Cristina Benros, Serge Hazout, Alexandre de Brevern. "Hybrid Protein Model": a new clustering approach for 3D local structures: Evolution of Hybrid Protein Model. 2002, pp.36-45. ⟨inserm-00134559⟩

Share

Metrics

Record views

253

Files downloads

90