Predicting membrane proteins and their types by extracting various sequence features into Chou's general PseAAC.
Mol Biol Rep
; 45(6): 2295-2306, 2018 Dec.
Article
em En
| MEDLINE
| ID: mdl-30238411
ABSTRACT
For many biological functions membrane proteins (MPs) are considered crucial. Due to this nature of MPs, many pharmaceutical agents have reflected them as attractive targets. It bears indispensable importance that MPs are predicted with accurate measures using effective and efficient computational models (CMs). Annotation of MPs using in vitro analytical techniques is time-consuming and expensive; and in some cases, it can prove to be intractable. Due to this scenario, automated prediction and annotation of MPs through CM based techniques have appeared to be useful. Based on the use of computational intelligence and statistical moments based feature set, an MP prediction framework is proposed. Furthermore, the previously used dataset has been enhanced by incorporating new MPs from the latest release of UniProtKB. Rigorous experimentation proves that the use of statistical moments with a multilayer neural network, trained using back-propagation based prediction techniques allows more thorough results.
Palavras-chave
Texto completo:
1
Base de dados:
MEDLINE
Assunto principal:
Análise de Sequência de Proteína
/
Proteínas de Membrana
Tipo de estudo:
Prognostic_studies
/
Risk_factors_studies
Limite:
Animals
/
Humans
Idioma:
En
Ano de publicação:
2018
Tipo de documento:
Article