Department of Electronics and Information Engineering, Jeonbuk National University, Jeonju 54896, Republic of Korea.
School of Pharmacy, Jeonbuk National University, Jeonju 54896, Republic of Korea.
Cells. 2023 Mar 7;12(6):829. doi: 10.3390/cells12060829.
The sigma (σ) factor of RNA holoenzymes is essential for identifying and binding to promoter regions during gene transcription in prokaryotes. σ54 promoters carried out various ancillary methods and environmentally responsive procedures; therefore, it is crucial to accurately identify σ54 promoter sequences to comprehend the underlying process of gene regulation. Herein, we come up with a convolutional neural network (CNN) based prediction tool named "iProm-Sigma54" for the prediction of σ54 promoters. The CNN consists of two one-dimensional convolutional layers, which are followed by max pooling layers and dropout layers. A one-hot encoding scheme was used to extract the input matrix. To determine the prediction performance of iProm-Sigma54, we employed four assessment metrics and five-fold cross-validation; performance was measured using a benchmark and test dataset. According to the findings of this comparison, iProm-Sigma54 outperformed existing methodologies for identifying σ54 promoters. Additionally, a publicly accessible web server was constructed.
RNA 全酶的 σ 因子对于原核生物基因转录过程中识别和结合启动子区域至关重要。σ54 启动子执行各种辅助方法和环境响应程序；因此，准确识别 σ54 启动子序列对于理解基因调控的基本过程至关重要。在这里，我们提出了一种基于卷积神经网络（CNN）的预测工具，名为“iProm-Sigma54”，用于预测 σ54 启动子。CNN 由两个一维卷积层组成，后面是最大池化层和 dropout 层。采用独热编码方案提取输入矩阵。为了确定 iProm-Sigma54 的预测性能，我们采用了四种评估指标和五折交叉验证；使用基准数据集和测试数据集进行性能衡量。根据这一比较的结果，iProm-Sigma54 在识别 σ54 启动子方面优于现有的方法。此外，还构建了一个公共访问的网络服务器。
