Homology modelling and structural analysis of Phytases

Homology modeling is a powerful tool in predicting the three dimensional (3D) structure of a protein using a solved structure within the same family as template. Compared to other methods such as X-ray crystallography and NMR, homology modelling has the advantage of being a fast yet reliable techniq...

Full description

Bibliographic Details
Main Authors: Noorbatcha, Ibrahim Ali, Mohd. Salleh, Hamzah, Amid, Azura
Format: Proceeding Paper
Language:English
Published: 2010
Subjects:
Online Access:http://irep.iium.edu.my/4071/
http://irep.iium.edu.my/4071/4/IRIIE2010_Homology__Poster.pdf
_version_ 1848776266278240256
author Noorbatcha, Ibrahim Ali
Mohd. Salleh, Hamzah
Amid, Azura
author_facet Noorbatcha, Ibrahim Ali
Mohd. Salleh, Hamzah
Amid, Azura
author_sort Noorbatcha, Ibrahim Ali
building IIUM Repository
collection Online Access
description Homology modeling is a powerful tool in predicting the three dimensional (3D) structure of a protein using a solved structure within the same family as template. Compared to other methods such as X-ray crystallography and NMR, homology modelling has the advantage of being a fast yet reliable technique in solving proteins’ 3D structures, starting from their amino acid sequences. In this study, homology modelling was used to obtain the 3D structure of two different phytases using only their amino acid sequence; phyFAUIA1_H which belongs to the histidine acid phosphatases, and Bacillus subtilis SAUIA243 which belongs to the β-propeller phytases. E. coli phytase (PDB code 1dkm) was used as a template for phyFAUIA1_H model with 99.512% sequence identity. On the other hand, Bacillus amyloliquefaciens (PDB code 2poo) was used as a template for B. subtilis SAUIA243 model with 71.268% sequence identity. The BLAST (Basic Local Alignment Search Tool) algorithm was used to find the highest scoring ungapped local alignment between the query and database sequences. This particular score uses the BLOSUM substitution matrix, which is a block substitution matrix that uses segments of blocks corresponding to the most highly conserved regions of proteins. Both models were evaluated and found to be quite satisfactory without being manually modified. Phytases are important constituents in Animal feed industry. These models can now be used to design new phytases, with improve properties such as enhanced enzyme activity, pH tolerance, and thermostability.
first_indexed 2025-11-14T14:27:21Z
format Proceeding Paper
id iium-4071
institution International Islamic University Malaysia
institution_category Local University
language English
last_indexed 2025-11-14T14:27:21Z
publishDate 2010
recordtype eprints
repository_type Digital Repository
spelling iium-40712021-06-23T03:32:25Z http://irep.iium.edu.my/4071/ Homology modelling and structural analysis of Phytases Noorbatcha, Ibrahim Ali Mohd. Salleh, Hamzah Amid, Azura TP248.13 Biotechnology Homology modeling is a powerful tool in predicting the three dimensional (3D) structure of a protein using a solved structure within the same family as template. Compared to other methods such as X-ray crystallography and NMR, homology modelling has the advantage of being a fast yet reliable technique in solving proteins’ 3D structures, starting from their amino acid sequences. In this study, homology modelling was used to obtain the 3D structure of two different phytases using only their amino acid sequence; phyFAUIA1_H which belongs to the histidine acid phosphatases, and Bacillus subtilis SAUIA243 which belongs to the β-propeller phytases. E. coli phytase (PDB code 1dkm) was used as a template for phyFAUIA1_H model with 99.512% sequence identity. On the other hand, Bacillus amyloliquefaciens (PDB code 2poo) was used as a template for B. subtilis SAUIA243 model with 71.268% sequence identity. The BLAST (Basic Local Alignment Search Tool) algorithm was used to find the highest scoring ungapped local alignment between the query and database sequences. This particular score uses the BLOSUM substitution matrix, which is a block substitution matrix that uses segments of blocks corresponding to the most highly conserved regions of proteins. Both models were evaluated and found to be quite satisfactory without being manually modified. Phytases are important constituents in Animal feed industry. These models can now be used to design new phytases, with improve properties such as enhanced enzyme activity, pH tolerance, and thermostability. 2010-01-26 Proceeding Paper NonPeerReviewed application/pdf en http://irep.iium.edu.my/4071/4/IRIIE2010_Homology__Poster.pdf Noorbatcha, Ibrahim Ali and Mohd. Salleh, Hamzah and Amid, Azura (2010) Homology modelling and structural analysis of Phytases. In: IIUM Research, Innovation & Invention Exhibition (IRIIE 2010), 26 - 27 January 2010, Kuala Lumpur. (Unpublished) http://www.iium.edu.my/irie/10/sub10/author/list_p.php
spellingShingle TP248.13 Biotechnology
Noorbatcha, Ibrahim Ali
Mohd. Salleh, Hamzah
Amid, Azura
Homology modelling and structural analysis of Phytases
title Homology modelling and structural analysis of Phytases
title_full Homology modelling and structural analysis of Phytases
title_fullStr Homology modelling and structural analysis of Phytases
title_full_unstemmed Homology modelling and structural analysis of Phytases
title_short Homology modelling and structural analysis of Phytases
title_sort homology modelling and structural analysis of phytases
topic TP248.13 Biotechnology
url http://irep.iium.edu.my/4071/
http://irep.iium.edu.my/4071/
http://irep.iium.edu.my/4071/4/IRIIE2010_Homology__Poster.pdf