作者: Alfredo Vellido Alcacena , Jesús Giraldo , Raúl Cruz Barbosa
DOI:
关键词:
摘要: G Protein-coupled receptors (GPCRs) are integral cell mem- brane proteins of great relevance for pharmacology due to their role in transducing extracellular signals. The 3-D structure is unknown most them, and the investigation structure-function relationships usually relies on construction receptor models from amino acid sequence alignment onto those known structure. Se- quence risks loss relevant information. Different ap- proaches have attempted analysis alignment-free sequences basis physicochemical properties. In this paper, we use Auto-Cross Covariance method compare it an compo- sition representation. Novel semi-supervised manifold learning methods then used classify several members class C GPCRs transformed data. This approach because pro- tein not always labeled that provide robust classification a limited amount labels required.