Skip to main content
Top

2015 | Book

Computational Intelligence Techniques for Comparative Genomics

Dedicated to Prof. Allam Appa Rao on the Occasion of His 65th Birthday

Editors: Naresh Babu Muppalaneni, Vinit Kumar Gunjan

Publisher: Springer Singapore

Book Series : SpringerBriefs in Applied Sciences and Technology

insite
SEARCH

About this book

This Brief highlights Informatics and related techniques to Computer Science Professionals, Engineers, Medical Doctors, Bioinformatics researchers and other interdisciplinary researchers. Chapters include the Bioinformatics of Diabetes and several computational algorithms and statistical analysis approach to effectively study the disorders and possible causes along with medical applications.

Table of Contents

Frontmatter
Diversified Insulin-Associated Beta-Behavioral and Endogenously Triggered Exposed Symptoms (DIABETES) Model of Diabetes in India
Abstract
The objective of the paper was to propose a diversified insulin-associated beta-behavioral and endogenously triggered exposed symptoms (DIABETES) model due to multiple factors so as to suggest medical remedies to improve the avoidance of diabetes disease. The various causes and their chances toward the most common diabetes effects including deaths in the nation are modeled as an diversified insulin association. The regional, social, biological and cultural aspects of the Indian community are considered to model the diabetes that of different categories. The performance and the correctness of the model are determined by considering the heterogeneity due to different factors that are specified as proposed IAB-ETES process algebra. The model-driven approach needs restricted operations on the variables to supplement any health care information system. The individual human responsibility and societal awareness along with the health regulation acts can minimize the vulnerability of the disease if the information technology for biological system complies with the enforcement acts of the developing nation.
P. Raja Rajeswari, Chandrasekaran Subramaniam, Allam Appa Rao
Automatic Teaching–Learning-Based Optimization: A Novel Clustering Method for Gene Functional Enrichments
Abstract
Multi-objective optimization emerged as a significant research area in engineering studies because most of the real-world problems require optimization with a group of objectives. The most recently developed meta-heuristics called the teaching–learning-based optimization (TLBO) and its variant algorithms belongs to this category. This paper provokes the importance of hybrid methodology by illuminating this meta-heuristic over microarray datasets to attain functional enrichments of genes in the biological process. This paper persuades a novel automatic clustering algorithm (AutoTLBO) with a credible prospect by coalescing automatic assignment of k value in partitioned clustering algorithms and cluster validations into TLBO. The objectives of the algorithm were thoroughly tested over microarray datasets. The investigation results that endorse AutoTLBO were impeccable in obtaining optimal number of clusters, co-expressed cluster profiles, and gene patterns. The work was further extended by inputting the AutoTLBO algorithm outcomes into benchmarked bioinformatics tools to attain optimal gene functional enrichment scores. The concessions from these tools indicate excellent implications and significant results, justifying that the outcomes of AutoTLBO were incredible. Thus, both these rendezvous investigations give a lasting impression that AutoTLBO arises as an impending colonizer in this hybrid approach.
Ramachandra Rao Kurada, K. Karteeka Pavan, Allam Appa Rao
A Comparative Study of Methodologies of Protein Secondary Structure
Abstract
All living organisms are made up of cells and each cell in its turn consists of certain protein consequences which exercise an important role in catalyzing the chemical reactions. So, a study of a protein structure becomes a search lamp in the diagnosis of a disease. When the percent identity between two protein sequences falls below 33 %, it necessities to carry out the analysis of protein secondary structure. Of the several methodologies developed to analyze the protein secondary structure, two methods proved to be sound-dictionary of secondary structure of proteins (DSSP) and Garnier, Osguthrope and Robson (GOR), even though the prediction accuracy of GOR V is 73.5 % due to hazards in its implementation, GOR IV is generally used in spite of its accuracy being only to 64.4 %.
M. Rithvik, G. Nageswara Rao
A Sparse-modeled ROI for GLAM Construction in Image Classification Problems—A Case Study of Breast Cancer
Abstract
Image segmentation is a process to determine regions of interest (ROI) in mammograms. Mammograms can be classified by extracting textural features of ROI using Gray Level Aura Matrices (GLAM). Scientists are selecting a fixed window size for all ROIs to find respective GLAM, though the masses will not occur in regular two-dimensional geometries. This paper makes an attempt to replicate the problem but by choosing arbitrary shape of masses as they occur. It is found that this kind of natural selection of the arbitrary shape yielded drastic reduction in time complexity by adopting the method suggested by us.
K. Karteeka Pavan, Ch. Srinivasa Rao
A Survey on Identification of Protein Complexes in Protein–protein Interaction Data: Methods and Evaluation
Abstract
Since identification of protein complexes from protein–protein interaction (PPI) networks plays an important role in the computational biology, in this paper, we discuss different types of protein complex identification algorithms such as Markov Clustering algorithm, ClusterBFS, Connected Affinity Clique Extension, PE-weighted Clustering algorithm, Detection of Protein Complex Core and Attachment Algorithm and Dynamic Protein Complex Algorithm. Thereafter, we focus on computational analysis of protein complexes through various measures and various protein interaction databases, with which we can detect protein complexes effectively and efficiently.
Praveen Tumuluru, Bhramaramba Ravi, Sujatha Ch
Modeling Artificial Life: A Cellular Automata Approach
Abstract
The key feature of artificial life is the idea of emergence, where new patterns or behaviors emerge from complex computational processes that cannot be predicted. Emergence initiates the formation of higher-order properties via the interaction of lower-level properties. Biological networks contain many theory models of evolution. Similarities between the theoretically estimated networks and empirically modeled counterpart networks are considered as evidence of the theoretic and predictive biological evolution. However, the methods by which these theoretical models are parameterized and modeled might lead to inference validity questions. Opting for randomized parametric values is a probabilistic concern that a model produces. There persists a wide range of probable parameter values which allow a model to produce varying statistic results according to the parameters selected. While using the phenomenon of cellular automata, we tried to model life on a grid of squares. Each square in the grid is taken as a biological cell; we have framed rules such that the process of cell division and pattern formation in terms of biological theoretic perspective is studied. Relatively complex behaviors of the cell patterns which vary from generation to generation are visually analyzed. Three algorithms—game of life, Langton’s ant, and hodgepodge—have been implemented whose technical implementation will provide an inspiration and foundation to build simulators that exhibit characteristics and behaviors of biological systems of reproduction.
Kunjam Nageswara Rao, Madugula Divya, M. Pallavi, B. Naga Priyanka
Identification of Deleterious SNPs in TACR1 Gene Using Genetic Algorithm
Abstract
Bioinformatics is a specific research and development area. The purpose of bioinformatics mainly deals with data mining and the relationships and patterns in large databases to provide useful information analysis and diagnosis. Single nucleotide polymorphisms (SNP) are one of the major causes of genetic diseases. Identification of disease-causing SNPs can identify better disease diagnosis. Hence, the present study aims at the identification of deleterious SNPs in TACR1 gene. Developing an algorithm plays a vital role in computational intelligence techniques. In this paper, a genetic algorithm (GA) approach is to develop rules and it is presented. The importance of the accuracy, sensitivity, specificity, and comprehensibility of the rules is simplified for the implementation of a GA. The outline of encoding and genetic operators and fitness function of GA are discussed. GA is using to identify deleterious or damaged SNPs.
Dharmaiah Devarapalli, Ch. Anusha, Panigrahi Srikanth
Identification of AIDS Disease Severity Using Genetic Algorithm
Abstract
Bioinformatics is a data intentionally the field in Research and Development. The purpose of bioinformatics data mining (DM) is to observe the relationships and patterns in large databases to provide useful data analysis and results. Evolutionary algorithms play a main role in computational intelligence techniques. An developing situation was created throughout the world regarding the human immunodeficiency virus infection/acquired immunodeficiency syndrome (HIV/AIDS) disease that is mainly stigma. Every country is facing this problem. According to present survey is World health organization (WHO), AIDS disease has its complexity are health disease going present century. A best way to early examine of AIDS may improve the lives of all people affected by AIDS and people may lead healthy life. In this part, we have present an evolutionary algorithm known as Genetic Algorithm (GA) for better results of AIDS disease using association rule mining. In this computational intelligence technique, we tested the performance of the method using AIDS dataset. We presented a better fitness function using coverage, comprehensibility, and rule length. This fitness function we achieved is promising accuracy for model.
Dharmaiah Devarapalli, Panigrahi Srikanth
A Novel Clustering Approach Using Hadoop Distributed Environment
Abstract
Nowadays, information retrieval plays a vital role by allowing users to retrieve documents of their interest based on relevance score. Such systems can be implemented either in distributed systems or parallel systems to achieve high throughput. If such kind of framework is deployed in a cloud, grouping of relevant documents is essential to retrieve documents of interest. Hence, an efficient and scalable clustering is required to process huge volume of documents. To handle huge documents and to provide scalability while processing Apache Hadoop is efficient with its powerful feature map reduce. Hence, in this paper, a novel approach is proposed that is capable of clustering bulk data with high throughput. This paper also demonstrates the need of parallel caching approach for obtaining effective results.
Nagesh Vadaparthi, P. Srinivas Rao, Y. Srinivas, M. Athmaja
Framework for Evaluation of Programming Language Examinations
Abstract
Recent advancements in the field of e-learning and virtual learning have changed the face of education. An important part of learning process is evaluation of student learning through examinations. This paper suggests a framework for evaluation of computer science practical examinations. The framework is implemented using Java programming language and many open source tools and libraries. The developed framework performs evaluation in four steps. The first step is compiler output interpretation in which the false errors generated by compiler are eliminated and only actual errors are reported. In the second step, unit testing of compiled programs is done. In third step, software metrics like lines of code, lines of comment, McCabe’s cyclomatic complexity, and number of modules are calculated for the programs. Finally, the semantic similarity of student programs is checked against the model program. The implemented framework is tested on student programs, and the accuracy of results is satisfactory. This framework will be helpful in efficiently evaluating student programs in practical examinations. It works for C, C++, and Java programming languages.
Himani Mittal, Syamala Devi Mandalika
An Efficient Data Integration Framework in Cloud Using MapReduce
Abstract
In Bigdata applications, providing security to massive data is an important challenge because working with such data requires large scale resources that must be provided by cloud service provider. Here, this paper demonstrates a cloud implementation and technologies using big data and discusses how to protect such data using hashing and how users can be authenticated. In particular, technologies using big data such as the Hadoop project of Apache are discussed, which provides parallelized and distributed data analyzing and processing of petabyte of data, along with a summarized view of monitoring and usage of Hadoop cluster. In this paper, an algorithm called FNV hashing is introduced to provide integrity of the data that has been outsourced to cloud by the user. The data within Hadoop cluster can be accessed and verified using hashing. This approach brings out to enable many new security challenges over the cloud environment using Hadoop distributed file system. The performance of the cluster can be monitored by using ganglia monitoring tool. This paper designs an evaluation cloud model which will provide quantity related results for regularly checking accuracy and cost. From the results of the experiment found out that this model is more accurate, cheaper and can respond in real time.
P. Srinivasa Rao, M. H. M. Krishna Prasad, K. Thammi Reddy
Metadata
Title
Computational Intelligence Techniques for Comparative Genomics
Editors
Naresh Babu Muppalaneni
Vinit Kumar Gunjan
Copyright Year
2015
Publisher
Springer Singapore
Electronic ISBN
978-981-287-338-5
Print ISBN
978-981-287-337-8
DOI
https://doi.org/10.1007/978-981-287-338-5

Premium Partner