| Prediction on DNA Binding Sequence in Deep Learning Approach |
| Pages: 77-85 (9) | [Full Text] PDF (856 KB) |
| T Akter, L Pinky, MM Islam, MM Rahman |
| Department of Computer Science and Engineering (CSE), Mawlana Bhashani Science and Technology University (MBSTU), Santosh, Tangail-1902, Bangladesh |
|
Abstract - Prediction of DNA-binding proteins from sequence information is the most challenging tasks in genome annotation. The principal study of our research is to determine DNA-binding proteins from primary protein sequences comparing the accuracy of the model in a deep learning based procedure. We also put the protein sequence in DNA binder tool to predict whether the sequence is DNA-Binding or non-DNA-Binding protein. We have encoded the datasets on hot encoded process and two stages of convolutional neutral network to detect the function domains of protein sequences. Our proposed method is being trained and tested with protein sequences of human and mouse, while it predicts 37% of training, 13% of testing and 50 % of the original sequences that finally count 74% of them for training and 26% of them for testing datasets of DNA-Binding proteins. The proposed method achieves a prediction accuracy of 70.50 %, sensitivity of 62.32%, specificity of 78.27% and the Matthews correlation coefficient at 1. Support vector machine and logistic regression classifier are used upon our method to increase the accuracy. The accuracy has increased using both classifier and it raises 12% and 9% respectively. We have also count prediction analysis of precision, recall, f-measure and false discovery rate of the protein sequence. |
| Index Terms - DNA-binding and non DNA-binding, DNA sequence classi?cation, Convolutional Neutral Network (CNN),Long Short Term Memory (LSTM), Support Vector Machine (SVM) |
|
C itation - T Akter, L Pinky, MM Islam, MM Rahman. "Prediction on DNA Binding Sequence in Deep Learning Approach." International Journal of Computer Science and Software Engineering 8, no. 4 (2019): 77-85. |
| A Methodology for Requirement Volatility and Traceability |
| Pages: 86-98 (13) | [Full Text] PDF (p2-V8I4.pdf) |
| A Babiker, A SAHRAOUI |
| Sudan University of Science and technology, Khartum, SudanLAAS du CNRS and Toulouse University, UT2J7, Avenue Colonel Roche 31077 Toulouse, France |
|
Abstract - This paper describes an approach for requirements volatility in systems development. The methodology is based on an original traceability model. The impact of requirements on safety is outlined through backward and forward traceability. |
| Index Terms - Requirements Traceability, Evolution Safety, Requirements Volatility, Volere |
|
C itation - A Babiker, A SAHRAOUI. "A Methodology for Requirement Volatility and Traceability." International Journal of Computer Science and Software Engineering 8, no. 4 (2019): 86-98. |
| Algorithms for Automating Artifact Attribute Classification of Software Systems |
| Pages: 99-103 (5) | [Full Text] PDF (244 KB) |
| VK Reddy, AA Rao |
| Department of Computer Science & Engineering, JNTUA College of Engineering, Anantapur, India |
|
Abstract - Software systems generally involve a number of phases and tend to evolve over a period of time. Several revisions of individual artifacts which make up the system take place during the evolution process. The revisions and refinements are captured and maintained as different versions using configuration/version managem-ent tools. A key issue in the version management of object oriented software system is classification of attributes of an artifact. Evolution needs to be captured, for capturing the evolution we need to maintain the various artifacts of software systems. For this purpose the behavior of the artifacts to be understood properly. This is possible through various attributes of artifacts. Presently attributes of an artifact are categorized into two different types called versioning and non-versioning. Versioning attributes govern the behavior of an artifact and non versioning attribute do not. This paper proposes an algorithms for automating the process of determining the versioning and non-versioning functionalities of the above classification. |
| Index Terms - Change Propagation, Equivalent Change, Version Change, and Version Management |
|
C itation - VK Reddy, AA Rao. "Algorithms for Automating Artifact Attribute Classification of Software Systems." International Journal of Computer Science and Software Engineering 8, no. 4 (2019): 99-103. |