IJSTR

International Journal of Scientific & Technology Research

Home About Us Scope Editorial Board Blog/Latest News Contact Us
0.2
2019CiteScore
 
10th percentile
Powered by  Scopus
Scopus coverage:
Nov 2018 to May 2020

CALL FOR PAPERS
AUTHORS
DOWNLOADS
CONTACT

IJSTR >> Volume 9 - Issue 6, June 2020 Edition



International Journal of Scientific & Technology Research  
International Journal of Scientific & Technology Research

Website: http://www.ijstr.org

ISSN 2277-8616



Binding Affinity Prediction Of SCA Using PLD And PMLD With Functional DNN And Its Variants

[Full Text]

 

AUTHOR(S)

P. R. Asha, M. S. Vijaya

 

KEYWORDS

Binding affinity; Deep Neural Network; Docking; Functional Deep Neural Network; Layer Creation; Optimizers; Prediction; PLD Dataset; PMLD Dataset; Protein Structure; Repeat Mutation.

 

ABSTRACT

Binding affinity prediction of hereditary anarchy like spinocerebellar ataxia (SCA) is imperative in medical field. Existing affinity prediction models built through machine learning demonstrates less prediction rate due to issues like selection of features, architecture of learning algorithm, hyper parameters used in learning. Hence a new affinity prediction model is proposed to meet the above challenges and to increase the performance of prediction. In this work, binding affinity prediction is implemented with customized layers in deep neural network by training weights and sharing features. Pre-trained weights with customized layers in DNN is experimented with two approaches namely protein-ligand docking (PLD) and protein-mutated-ligand docking (PMLD) to facilitate accurate prediction. Two datasets are created using two docking approaches. The first dataset is created by docking 17 molecular structures of six types of spinocerebellar ataxia with 18 ligands. Features like energy calculations are extracted from the docked complex to predict binding affinity and the dataset is termed as PLD. The second dataset is shaped by mutating the protein of spinocerebellar ataxia by repeat mutation and docked with ligand to produce the complexes. Features like scoring functions, energy calculations and descriptors are extracted from the complex to model the affinity binding and it is phrased as PMLD. Customized layers in deep neural network are defined with three optimizers namely adam, rmsprop and nadam optimizer and the experiments have been carried out using two datasets. The results are compared with performance results of the model through functional deep neural network. Results demonstrates that the model built with customized layers in deep neural network for PLD dataset attains the highest prediction rate where the self learnt features from the hand crafted features found to be more precise in prediction.

 

REFERENCES

[1] George E. Dahl, “Multi-task Neural Networks for QSAR Predictions”, datascience association, 2014
[2] Hakime Öztürk, Arzucan Özgür, Elif Ozkirimli, “DeepDTA: deep drug–target binding affinity prediction”, Bioinformatics, vol 34, issue 17, Sep 2018
[3] Dong-Sheng Cao, Liu-Xia Zhang, Gui-Shan Tan, Zheng Xiang, Wen-Bin Zeng, Qing-Song Xu and Alex F Chen, “Computational Prediction of Drug-Target Interactions Using Chemical, Biological, and Network Features”, Molecular Informatics, vol 33, issue 10, Oct 2014
[4] Dong-Sheng Cao, ShaoLiu, Qing-songXu, Hong-MeiLu, Jian-HuaHuang, Qian-NanHu, Yi-ZengLiang “Large-scale prediction of drug–target interactions using protein sequences and drug topological structures”, Analytica chimica acta, vol 752, issue 8, Nov 2012
[5] Asha P R, Vijaya MS, “Deep Neural Networks for Affinity Prediction of Spinocerebellar Ataxia using Protein Structures” Journal of Advanced Research and Control Dynamical Systems, Vol 11, no 4.
[6] Asha P R, Vijaya M S, “Affinity Prediction of Spinocerebellar Ataxia Using Protein-Ligand and Protein-Protein Interactions with Functional Deep Learning”, International journal of Engineering and Advanced Technology, vol 8, no 5, July 2017
[7] Alexander Aliper, Sergey Plis, Artem Artemov, Alvaro Ulloa, Polina Mamoshina, and Alex Zhavoronkov, Deep learning applications for predicting pharmacological properties of drugs and drug repurposing using transcriptomic data, Molecular Pharmaceutics, 2016 13(7), 2524-2530
[8] Thomas Unterthiner , Andreas Mayr , Gunter KlambauerJesse, Marvin Steijaert, Jorg K. Wegner, Hugo Ceulemans, Sepp Hochreiter, Deep Learning for Drug Target Prediction, Semantic scholar
[9] Rhys Heffernan, Kuldip Paliwal, James Lyons, Abdollah Dehzangi, Alok Sharma, Jihua Wang, Abdul Sattar, YuedongYang & Yaoqi Zhou, Improving prediction of secondary structure, local backbone angles, and solvent accessible surface area of proteins by iterative deep learning, Scientific Reports, june 22, 2015
[10] Preamsudha V, Vijaya MS, “Identification of Autism Spectrum Disorder using a Multi-Label Approach”, Journal of Advanced Research in Dynamical and Control Systems, Vol 11, no 4, pp 134-141, June 2019
[11] Sathyavikasini K, Vijaya MS, “Muscular Dystrophy Disease Classification Using Relative Synonymous Codon Usage”, International Journal of Machine Learning and Computing, Vol 6, pp 139-144, 2016
[12] Asha P R, Vijaya M S, “Affinity Prediction of Spinocerebellar Ataxia using ProteinProtein Interactions and Deep Neural Network with User-Defined Layer”, International Journal of Advanced science and Technology, vol 28, pp 20-37, 2019