eprintid: 17891
rev_number: 2
eprint_status: archive
userid: 1
dir: disk0/00/01/78/91
datestamp: 2023-12-19 03:24:12
lastmod: 2023-12-19 03:24:12
status_changed: 2023-12-19 03:08:52
type: article
metadata_visibility: show
creators_name: Naseer, S.
creators_name: Ali, R.F.
creators_name: Khan, Y.D.
creators_name: Dominic, P.D.D.
title: iGluK-Deep: computational identification of lysine glutarylation sites using deep neural networks with general pseudo amino acid compositions
ispublished: pub
keywords: lysine; amino acid, accuracy; amino acid sequence; Article; artificial intelligence; artificial neural network; chromatin structure; deep neural network; DNA damage; DNA repair; feature extraction; gray matter; human; k nearest neighbor; nerve cell network; prediction; protein secondary structure; proteomics; sensitivity and specificity; signal noise ratio; telomere; algorithm; biology; chemistry; procedures; protein processing, Algorithms; Amino Acids; Computational Biology; Lysine; Neural Networks, Computer; Protein Processing, Post-Translational
note: cited By 27
abstract: Lysine glutarylation is a post-translation modification which plays an important regulatory role in a variety of physiological and enzymatic processes including mitochondrial functions and metabolic processes both in eukaryotic and prokaryotic cells. This post-translational modification influences chromatin structure and thereby results in global regulation of transcription, defects in cell-cycle progression, DNA damage repair, and telomere silencing. To better understand the mechanism of lysine glutarylation, its identification in a protein is necessary, however, experimental methods are time-consuming and labor-intensive. Herein, we propose a new computational prediction approach to supplement experimental methods for identification of lysine glutarylation site prediction by deep neural networks and Chou�s Pseudo Amino Acid Composition (PseAAC). We employed well-known deep neural networks for feature representation learning and classification of peptide sequences. Our approach opts raw pseudo amino acid compositions and obsoletes the need to separately perform costly and cumbersome feature extraction and selection. Among the developed deep learning-based predictors, the standard neural network-based predictor demonstrated highest scores in terms of accuracy and all other performance evaluation measures and outperforms majority of previously reported predictors without requiring expensive feature extraction process. iGluK-Deep:Computational Identification of lysine glutarylationsites using deep neural networks with general Pseudo Amino Acid Compositions Sheraz Naseer, Rao Faizan Ali, Yaser Daanial Khan, P.D.D Dominic Communicated by Ramaswamy H. Sarma. © 2021 Informa UK Limited, trading as Taylor & Francis Group.
date: 2022
publisher: Taylor and Francis Ltd.
official_url: https://www.scopus.com/inward/record.uri?eid=2-s2.0-85112579260&doi=10.1080%2f07391102.2021.1962738&partnerID=40&md5=fd9d654d50c6ac4589b6da50581c4ed5
id_number: 10.1080/07391102.2021.1962738
full_text_status: none
publication: Journal of Biomolecular Structure and Dynamics
volume: 40
number: 22
pagerange: 11691-11704
refereed: TRUE
issn: 07391102
citation:   Naseer, S. and Ali, R.F. and Khan, Y.D. and Dominic, P.D.D.  (2022) iGluK-Deep: computational identification of lysine glutarylation sites using deep neural networks with general pseudo amino acid compositions.  Journal of Biomolecular Structure and Dynamics, 40 (22).  pp. 11691-11704.  ISSN 07391102