School of Computer Science and Technology, Hainan University, 58 Renmin Avenue, Meilan District, Haidian Campus, Haikou 570228, China.
International Business School, Hainan University, 58 Renmin Avenue, Meilan District, Haidian Campus, Haikou 570228, China.
Brief Bioinform. 2024 Sep 23;25(6). doi: 10.1093/bib/bbae505.
Inflammatory responses may lead to tissue or organ damage, and proinflammatory peptides (PIPs) are signaling peptides that can induce such responses. Many diseases have been redefined as inflammatory diseases. To identify PIPs more efficiently, we expanded the dataset and designed an ensemble learning model with manually encoded features. Specifically, we adopted a more comprehensive feature encoding method and considered the actual impact of certain features to filter them. Identification and prediction of PIPs were performed using an ensemble learning model based on five different classifiers. The results show that the model's sensitivity, specificity, accuracy, and Matthews correlation coefficient are all higher than those of the state-of-the-art models. We named this model MultiFeatVotPIP, and both the model and the data can be accessed publicly at https://github.com/ChaoruiYan019/MultiFeatVotPIP. Additionally, we have developed a user-friendly web interface for users, which can be accessed at http://www.bioai-lab.com/MultiFeatVotPIP.
炎症反应可能导致组织或器官损伤，而促炎肽（PIPs）是可以诱导这种反应的信号肽。许多疾病已被重新定义为炎症性疾病。为了更有效地识别 PIPs，我们扩展了数据集并设计了一个具有手动编码特征的集成学习模型。具体来说，我们采用了更全面的特征编码方法，并考虑了某些特征的实际影响来对其进行筛选。使用基于五个不同分类器的集成学习模型进行 PIPs 的识别和预测。结果表明，该模型的灵敏度、特异性、准确性和马修斯相关系数均高于现有模型。我们将这个模型命名为 MultiFeatVotPIP，模型和数据都可以在 https://github.com/ChaoruiYan019/MultiFeatVotPIP 上公开访问。此外，我们还为用户开发了一个用户友好的网络界面，可以在 http://www.bioai-lab.com/MultiFeatVotPIP 上访问。
