Comparative study of ammonium transporters in different organisms by study of a large number of structural protein features via data mining algorithms

Abstract

Ammonium is an excellent nitrogen source, and ammonium transfer is a fundamental process in most organisms. Membrane transport of ammonium is the key component of nitrogen metabolism mediated by Ammonium Transporter/Methylamine Permease/Rhesus (AMT/MEP/Rh) protein family. Ammonium transporters play different physiological roles in various organisms. Here, we looked at the protein characteristics of ammonium transporters in different organisms to create a link between protein characteristics and the organism. In order to increase the accuracy and precision of the employed models, for the first time, an attempt was made to cover all structural aspects of ammonium transporters in animals, bacteria, fungi, plants, and human by extracting and calculating 874 protein attributes of primary, secondary, and tertiary structures for each ammonium transporter. Then, various weighting and modeling algorithms were applied to determine how structural protein features change between organisms. Considering a large number of protein attributes made it possible to detect key protein characteristics in the structure of ammonium transporters. The results, for the first time, indicated that His-based features including count/frequency of His and frequency/count of Ile-His were the most significant features generating different types of ammonium transporters within organisms. Within different tested models, the C5.0 model was the most efficient and precise model for discrimination of organism type, based on ammonium transporter sequence, with the precision of 94.85%. The determination of protein characteristics of ammonium transporters in different organisms provides a new vista for understanding the evolution of transporters based on the modulation of protein characteristics and facilitates engineering of new transporters. In our point of view, dissecting a large number of structural protein characteristics through data mining algorithms provides a novel functional strategy for studying evolution and phylogeny. This research will serve as a basis for future studies on engineering novel ammonium transporters.

Understanding the Underlying Mechanism of HA-Subtyping in the Level of Physic-Chemical Characteristics of Protein

“…Sequence-alignment based methods (BLAST and phylogenic trees), drawn by nucleic acid or amino acid sequence alignments, have been extensively employed as the basis for evolutionary studies. However, homology-based methods does not consider the structural and functional features of proteins during evolution [24] . The presented approch in this study based on the machine learning algorithms running on structural protein features provides a new evolutionary pathway separation of HA a subtype which takes into account the structural reasons of this diversity.…”

“…Recently, we demonstrated that, instead of raw sequence analysis, extracting a large number of amino acid attributes and utilizing adequate data mining models can result in efficient and precise models in predicting the behaviour of malignant and benign breast cancer proteins [68] , thermostable proteins [23] , halostable proteins [69] , ammonium transporters [70] , and protein pumps [71] . A large scale analysis of amino acid structural attributes of influenza surfaces proteins rather than raw sequence allignment, may provide a clearer image of underlying molecular mechanisms of host range increase by detecting the key structural protein characteristics which govern HA subtyping.…”

DrugMiner: comparative analysis of machine learning algorithms for prediction of potential druggable proteins

“…It is well established that different attributes of proteins have a significant role in the interaction between proteins or between one or more drugs and a protein. These attributes vary from sequence-based features to physicochemical features [19,21,22,[24][25][26]33,34]. In this study, attributes were grouped into three categories: (i) Group 1: physicochemical properties of protein sequences, based on amino acid values in terms of their physicochemical parameters, such as length, weight, hydrophobicity, alpha helix, and so on; (ii) Group 2: amino acid composition, which was calculated based on the frequency of the 20 amino acid residues in the protein sequences; and (iii) Group 3: dipeptide composition, which was established based on the frequency of amino acid dimers in the protein sequences (Table 1).…”

“…In these approaches, simple sequence properties, such as amino acid and di-peptide content and/or frequency, are used to predict potential targets [16][17][18]. Computationally calculated structural amino acid and/or protein features are useful because they can be easily calculated based on sequence and frequently predict protein function accurately [19][20][21][22][23][24][25][26].…”

Data Mining for Identification of Forkhead Box O (FOXO3a) in Different Organisms Using Nucleotide and Tandem Repeat Sequences

“…count of hydrogen and CG) is associated with treatment outcome [12]. Tahrokh et al (2011) by study of a large number of structural protein specification, showed that data mining algorithms is a novel functional strategy for studying the evolution [13]. Two different databases were used in our study to find a method for examining the structural differences in the Foxo3a gene of different organism; one database was based on nucleotide features and one based on the tandem repeat sequences of the gene.…”