Department of Computer Science (PG)

Permanent URI for this communityhttps://dspace.psgrkcw.com/handle/123456789/146

Browse

Author: search.filters.author.Pream Sudha VHas files: Yes

Search Results

Now showing 1 - 8 of 8
  • No Thumbnail Available
    Item
    GATED RECURRENT NEURAL NETWORK FOR AUTISM SPECTRUM DISORDER GENE PREDICTION
    (Research Trend , International Journal on Emerging Technologies, 2020) Pream Sudha V; Vijaya M.S
    Autism Spectrum Disorder (ASD) is the fastest-growing complex disorder and the genetic ground of this comprehensive developmental disability is very difficult to research. Autism diagnosis for an average child is not done till the age of four, though it can be given at the age of 18 months to two years. Hence a computational model that enables the early diagnosis and personalized treatment is the need of the hour. In this research work, a deep learning based approach is proposed for Autism Spectrum Disorder (ASD) gene prediction. There are various contributors for Autism including genes, mutations, chromosomal settings influence of the environment, prenatal influences, family factors and problems during birth. Recurrent Neural Network (RNN) based Gated Recurrent Units (GRU) are implemented to develop a model that predicts ASD genes, mutations and gene susceptibility. GRUs with their internal memory capability are valuable to store and filter information using the update and reset gates. Also GRU offers a powerful tool to handle sequence data. The model is trained using three simulated datasets with features representing genes, mutations and gene susceptibility to ASD. Besides, the proposed approach is compared to original RNN and Long Short Term Memory Units (LSTM) for ASD prediction. The experimental results confirm that the proposed approach is promising with 82.5% accuracy and hence GRU RNN is found to be effective for ASD gene prediction
  • No Thumbnail Available
    Item
    PREDICTION OF GENE SUSCEPTIBILITY TO AUTISM SPECTRUM DISORDER USING DEEP ARCHITECTURES
    (International Journal of Scientific and Technology Research, 2020) Pream Sudha V; Vijaya M.S
    A genetic predisposition or susceptibility to Autism Spectrum Disorder (ASD) is an increased likelihood of developing it based on the genetic makeup of a person. The multiple variants found in each gene have their own probability of associated risk and so the major problem lies in the systematic evaluation of their functional significance to ASD. Hence it is essential to develop methods for quantitative evaluation of ASD candidate genes with co-occurring mutations which will provide a clear understanding of their relevance to ASD. This research work deals with the development of a discriminative model for prioritization of candidate genes considering mutations in them and to classify them based on their predisposition to the disorder. The model for gene susceptibility prediction is built by integrating the combined potential of substantiation for each ASD linked gene and the related mutations. In this research work gene susceptibility prediction is modelled as a pattern classification problem and deep learning techniques are employed to build the models. The performance evaluation of these models proves that Long Short Term Memory (LSTM) based gene susceptibility prediction model has shown better performance.
  • No Thumbnail Available
    Item
    RECURRENT NEURAL NETWORK BASE MODEL TO PREDICT AUTISM SPECTRUM DISORDER CAUSATIVE GENES
    (Science and Engineering Research Support Society(SERSC) International journal of Advanced science and technology, 2019) Pream Sudha V; Vijaya M S
    Recognizing genes causing Autism Spectrum Disorder (ASD) is still a complex task. The role played by domain experts is crucial in identifying relevant contributive features and as recognizing hand-crafted attributes occupies a great deal of time, a varying successful solution is necessary. The swift advancements in the design of deep architecture models have shown substantial accomplishment in sequential data processing tasks. Deep learning models examine the data to discover associations among the features and enable faster learning without being explicitly programmed to do so. Hence the principal goal of this work is to categorize the ASD genes by applying deep learning based models without feature engineering. One hot encoding method is used to encode the gene sequences as vector of numerical values and to further simplify the input representation to aid the prediction of ASD gene sequences. Recurrent Neural Network (RNN) models like Bidirectional Recurrent Neural Network (BRNN), Long Short Term Memory (LSTM) and Gated Recurrent Units (GRU) are employed to build the prediction models using user defined and self learned features. The performances of the models evaluated using cross validation with various metrics like precision, recall, accuracy and F-measure confirm that GRU model shows promising results using one hot encoding technique.
  • No Thumbnail Available
    Item
    DEEP LEARNING BASED PREDICTION OF AUTISM SPECTRUM DISORDER USING CODON ENCODING OF GENE SEQUENCES
    (International Journal of Engineering and Advanced Technology, 2019) Pream Sudha V; Vijaya M S
    The development of computational tools to recognize Autism Spectrum Disorder (ASD) originated by genetic mutations is vital to the development of disease-specific targeted therapies. Identifying genes causing the genetically transmitted ASD is still a challenging task. As genomics data is dependent on domain specific experts for identifying efficient features and extracting hand-crafted attributes involves much time, an alternate effective solution is the need of the hour. The rapid developments in the design of deep architecture models have led to the broad application of these models in a variety of research areas and they have shown considerable success in sequential data processing tasks. The primary goal of this work is to classify the ASD gene sequences by employing a Deep Neural Network based model. This in turn will enable effective genetic diagnoses of this disease and facilitate the targeted genetic testing of individuals. This work utilizes codon encoding and one hot encoding technique to transform the mutated gene sequences which are exploited for self learning the features by deep network. Experiments showed that the performance of the proposed model was better than that of the conventional Multilayer Perceptron with promising accuracy of 77.8%, 80.1% and 81.2% for three different datasets.
  • No Thumbnail Available
    Item
    IDENTIFICATION OF AUTISM SPECTRUM DISORDER USING A MULTI-LABEL APPROACH
    (Journal of Advanced Research in Dynamical & Control Systems, 2019) Pream Sudha V; Vijaya M S
    There has been an increase in the application of Multi-dimensional models in domains like bioinformatics, image processing, video and audio processing and text categorization. Multi-dimensional approach has its advantages with regard to predictive accuracy and time taken to build the model. The search for candidate genes of Autism spectrum disorder (ASD) is complicated as it involves significant interactions among mutations in several genes. This work investigates multi- dimensional learning which builds a model to predict ASD related multiple variables simultaneously using varied features. The study explored the different methods of multi- dimensional learning. ASD related gene sequences were analysed using different characteristics and each sequence was represented by a profile of 58 features from three different categories specific to gene, substitution matrix and amino acid. By combining three different problem transformation approaches with three base classifiers and using two algorithm adaptation methods a total of 15 different configurations were constructed. The different configurations were evaluated with multiple measurements including Hamming Loss, Hamming score, Zero One loss, Exact match, Accuracy, training and testing time. The results showed that the problem transformation algorithm Nearest Set replacement together with Naïve Bayes classifier outperformed the other configurations with 93.4 % accuracy and Hamming loss of 0.06, 0.12 Zero one loss.
  • No Thumbnail Available
    Item
    IDENTIFICATION AND CLASSIFICATION OF LEAF DISEASES IN TURMERIC PLANTS
    (International Journal of Engineering Research and Applications, 2016) Nandhini M; Pream Sudha V; Vijaya M S
    Plant disease identification is the most important sector in agriculture. Turmeric is one of the important rhizomatous crops grown in India. The turmeric leaf is highly exposed to diseases like rhizome rot, leaf spot, and leaf blotch. The identification of plant diseases requires close monitoring and hence this paper adopts technologies to manage turmeric plant diseases caused by fungi to enable production of high quality crop yields. Various image processing and machine learning techniques are used to identify and classify the diseases in turmeric leaf. The dataset with 800 leaf images of different categories were pre-processed and segmented to promote efficient feature extraction. Machine learning algorithms like support vector machine, decision tree and naïve bayes were applied to train the model. The performance of the model was evaluated using 10 fold cross validation and the results are reported
  • No Thumbnail Available
    Item
    MACHINE LEARNING-BASED MODEL FOR IDENTIFICATION OF SYNDROMIC AUTISM SPECTRUM DISORDER
    (Springer Studies in Computational Intelligence, 2018-09-15) Pream Sudha V; Vijaya M S
    Autism spectrum disorder (ASD) is characterized by a set of developmental disorders with a strong genetic origin. The genetic cause of ASD is difficult to track, as it includes a wide range of developmental disorders, a spectrum of symptoms and varied levels of disability. Mutations are key molecular players in the cause of ASD, and it is essential to develop effective therapeutic strategies that target these mutations. The development of computational tools to identify ASD originated by genetic mutations is vital to aid the development of disease-specific targeted therapies. This chapter employs supervised machine learning techniques to construct a model to identify syndromic ASD by classifying mutations that underlie these phenotypes, and supervised learning algorithms, namely support vector machines, decision trees and multilayer perceptron, are used to explore the results. It has been observed that the decision tree classifier performs better compared to other learning algorithms, with an accuracy of 94%. This model will provide accurate predictions in new cases with similar genetic background and enable the pathogenesis of ASD.
  • No Thumbnail Available
    Item
    DECISION TREE BASED MODEL FOR THE CLASSIFICATION OF PATHOGENIC GENE SEQUENCES CAUSING ASD
    (Springer Communications in Computer and Information Science, 2018) Pream Sudha V; Vijaya M S
    Pathogenic gene identification is an important research problem in biomedical domain. The genetic cause of ASD, which is a multifaceted developmental disability is hard to research. Hence, there is a critical need for inventive approaches to further portray the genetic basis of ASD which will enable better filtering and specific therapies. This paper adopts machine learning techniques to classify gene sequences which are the significant drivers of syndromic and asyndromic ASD. The synthetic dataset with 150 sequences of six different categories of genes were prepared and coding measures of gene sequences were taken as attributes for gene identification. Pattern learning algorithms like support vector machine, decision tree and Multiplayer perceptron were used to train the model. The model was evaluated using 10 fold cross validation and the results are reported. The study reveals that Decision trees outperform other classifiers with an accuracy of 97.33%