Wardah Wafaa, Dehzangi Abdollah, Taherzadeh Ghazaleh, Rashid Mahmood A, Khan M G M, Tsunoda Tatsuhiko, Sharma Alok
School of Computing, Information and Mathematical Sciences, Faculty of Science, Technology and Environment, The University of the South Pacific, Suva, Fiji.
Department of Computer Science, Morgan State University, Baltimore, USA.
J Theor Biol. 2020 Jul 7;496:110278. doi: 10.1016/j.jtbi.2020.110278. Epub 2020 Apr 13.
Interactions between proteins and peptides influence biological functions. Predicting such bio-molecular interactions can lead to faster disease prevention and help in drug discovery. Experimental methods for determining protein-peptide binding sites are costly and time-consuming. Therefore, computational methods have become prevalent. However, existing models show extremely low detection rates of actual peptide binding sites in proteins. To address this problem, we employed a two-stage technique - first, we extracted the relevant features from protein sequences and transformed them into images applying a novel method and then, we applied a convolutional neural network to identify the peptide binding sites in proteins.
We found that our approach achieves 67% sensitivity or recall (true positive rate) surpassing existing methods by over 35%.
蛋白质与肽之间的相互作用会影响生物学功能。预测此类生物分子相互作用有助于更快地预防疾病并助力药物研发。用于确定蛋白质 - 肽结合位点的实验方法成本高昂且耗时。因此，计算方法变得十分普遍。然而，现有模型对蛋白质中实际肽结合位点的检测率极低。为解决这一问题，我们采用了一种两阶段技术——首先，我们从蛋白质序列中提取相关特征，并运用一种新颖的方法将其转化为图像，然后，我们应用卷积神经网络来识别蛋白质中的肽结合位点。
我们发现我们的方法实现了67%的灵敏度或召回率（真阳性率），比现有方法高出35%以上。
