©2021-2032  All Rights Reserved. Online Journal of Bioinformatics.  You may not store these pages in any form except for your own personal use. All other usage or distribution is illegal under international copyright treaties. Permission to use any of these pages in any other way besides  the before mentioned must be gained in writing from the publisher. This article is exclusively copyrighted in its entirety to onlinejournals@gmail.com publications. This article may be copied once but may not be, reproduced or  re-transmitted without the express permission of the editors. Linking: To link to this page or any pages linking to this page you must link directly to this page only here rather than put up your own page.

OJBTM

Online Journal of Bioinformatics©

Established 1995

ISSN  1443-2250

 

Volume 23 (2):211-217, 2022.

Machine learning protein thermostability

 

E. Ebrahimie1 and M. Ebrahimi2

 

1Department of Crop Production & Plant Breeding, College of Agriculture, Shiraz University, 2Department of Biology & Bioinformatics Research Group, University of Qom, Qom, Iran

 

ABSTRACT

 

Ebrahimie E, Ebrahimi M., Machine learning protein thermostability, Onl J Bioinform., 23 (2):218-247, 2022. Paper, biofuel and detergent industries use toxic chlorines. Enzymes function below 60C and denature at higher temperatures but some withstand higher temperatures by adaptation. We report error-prone PCR mutagenic thermostable Bacillus halodurans by supervised and unsupervised machine learning algorithms with attribute weighting for amino acids that contribute to enzyme thermostability. We compared meso and thermostable enzymes by attribute weighting supervised and unsupervised clustering algorithm prediction model protein thermos-stability from amino acid properties. Mining protein sequences by machine learning algorithms of >900 amino acid attributes, increased accuracy. Models predicted thermostability from primary structure of protein maximization clustering  with uncertainly and correlation attributes, weighting algorithms classified thermo and meso-stable proteins 100%. Seventy per cent selected Gln content and frequency of hydrophilic residues as most important attributes. For dipeptides, frequency of Asn-Glu distinguished meso-stable from thermo-stable enzymes but we found no difference (p < 0.05) in performance of forward backward propagation neural networks. Thermo-stability irrespective of sequence similarity may reveal thermostable enzymes.

 

Keywords: Amino Acid, Bioinformatics, Modeling.

MAIN

 

FULL-TEXT (SUBSCRIBE OR PURCHASE TITLE)