Scholarlink

Multiple sequence alignment plays a vital role in the computational analysis of biological data. Various programs have been developed to analyze sequence similarity. The alignment accuracy of these programs is difficult to analyze and compare, given the noisy effect of several parameters and their interaction on the nonlinear evolution of this performance. This work provides a mathematical modeling analysis and comparison study of many popular sequence alignment programs of DNA sequences. Several tools were also used to generate, simulate sequences and evaluate alignments. Several parameters are considered, namely, number of sequences, sequence lengths, insertion rate and deletion rate. Quadratic mathematical models have been generated and analyzed and can be used for personalized analysis purposes like prediction and optimization of the accuracy of alignment tools. Furthermore, the adequacy of the obtained regression models was tested and analyzed. For most models the coefficient values: \(R^2\) , adjusted \(R^2\) , and predicted \(R^2\) were found to be good. These models were then employed for performance analysis and comparison of the investigated MSA tools, where thirty 3D surfaces were presented. The obtained models and graphical results provide a clear and easy way to analyze, interpret, optimize and compare MSA tools globally or under personalized configurations.

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

Quantitative Modeling and Evaluation of DNA Multiple Sequence Alignment Tools Through Response Surface Methodology

  • 会议论文

  • 发表日期：

    2025年

查看全文
  • Mohamed Skander Daas,
  • Safia Rouabah,
  • Khaled Boulahrouf,
  • Salim Chikhi,
  • El-Bay Bourennane

摘要

Multiple sequence alignment plays a vital role in the computational analysis of biological data. Various programs have been developed to analyze sequence similarity. The alignment accuracy of these programs is difficult to analyze and compare, given the noisy effect of several parameters and their interaction on the nonlinear evolution of this performance. This work provides a mathematical modeling analysis and comparison study of many popular sequence alignment programs of DNA sequences. Several tools were also used to generate, simulate sequences and evaluate alignments. Several parameters are considered, namely, number of sequences, sequence lengths, insertion rate and deletion rate. Quadratic mathematical models have been generated and analyzed and can be used for personalized analysis purposes like prediction and optimization of the accuracy of alignment tools. Furthermore, the adequacy of the obtained regression models was tested and analyzed. For most models the coefficient values: \(R^2\) , adjusted \(R^2\) , and predicted \(R^2\) were found to be good. These models were then employed for performance analysis and comparison of the investigated MSA tools, where thirty 3D surfaces were presented. The obtained models and graphical results provide a clear and easy way to analyze, interpret, optimize and compare MSA tools globally or under personalized configurations.