Scholarlink

<p>Splice site prediction refers to the computational identification of exon–intron boundaries within DNA sequences. It is essential for understanding the transcription and translation processes involved in gene expression. This paper presents the Randomized Exponential <i>Kbest</i> Gravitational Search Algorithm (REKGSA), a new variant of GSA that efficiently balances the explorative and exploitative behavior of the search process. For this purpose, a new <i>Kbest</i> strategy has been proposed with a random number multiplied by an exponential factor that varies with iteration. Apart from this, it has employed a new position-updating strategy incorporating a new scaling constant. REKGSA has been tested on the benchmark problems taken from the CEC 2014 benchmark instances. In most cases, it has performed better than the basic GSA and its three other versions. 3 classifiers, namely, Support Vector Machine (SVM), k Nearest Neighbors (kNN), and Decision Tree (DT), have been incorporated with REKGSA to produce three different models (REKGSA + SVM, REKGSA + kNN, and REKGSA + DT) to identify the Exon-Intron (EI), Intron-Exon (IE) junction, or Neither (N) existing in a DNA sequence. The results obtained have been compared with other state-of-the-art methods, where REKGSA + SVM achieved the highest accuracy among all.</p>

错误：搜索内容不能为空，请输入英文关键词
错误：关键词超出字数限制，请精简
高级检索

Randomized exponential Kbest gravitational search algorithm (REKGSA): a new variant of GSA to predict splicing sites in DNA transcription process

  • 期刊论文

  • 发表日期：

    2025年12月27日

查看全文
  • Joy Adhikary,
  • Sriyankar Acharyya

摘要

Splice site prediction refers to the computational identification of exon–intron boundaries within DNA sequences. It is essential for understanding the transcription and translation processes involved in gene expression. This paper presents the Randomized Exponential Kbest Gravitational Search Algorithm (REKGSA), a new variant of GSA that efficiently balances the explorative and exploitative behavior of the search process. For this purpose, a new Kbest strategy has been proposed with a random number multiplied by an exponential factor that varies with iteration. Apart from this, it has employed a new position-updating strategy incorporating a new scaling constant. REKGSA has been tested on the benchmark problems taken from the CEC 2014 benchmark instances. In most cases, it has performed better than the basic GSA and its three other versions. 3 classifiers, namely, Support Vector Machine (SVM), k Nearest Neighbors (kNN), and Decision Tree (DT), have been incorporated with REKGSA to produce three different models (REKGSA + SVM, REKGSA + kNN, and REKGSA + DT) to identify the Exon-Intron (EI), Intron-Exon (IE) junction, or Neither (N) existing in a DNA sequence. The results obtained have been compared with other state-of-the-art methods, where REKGSA + SVM achieved the highest accuracy among all.