Background: The ligand-receptor interaction plays an important role in signal transduction required for cellular differentiation, proliferation, and immune response process. The analysis of ligand-receptor interactions is helpful to provide a deeper understanding of cellular proliferation/ differentiation and other cell processes.

Methods: The computational technique would be used to promote ligand-receptor interactions research in future proteomics research. In this paper, we propose a novel computational method to predict ligand-receptor interactions from amino acid sequences by a machine learning approach. We extract features from ligand and receptor sequences by Histogram of Oriented Gradient (HOG) and Discrete Cosine Transform (DCT). Then, these features are fed into the Fuzzy C-Means (FCM) clustering algorithm for clustering, and also we get multiple training subsets to generate the same number of sub-classifiers. We choose an optimal sub-classifier for predicting ligand-receptor interactions according to the similarity from one sample to training subsets.

Observations: In order to verify the performance, we perform five-fold cross-validation experiments on a ligand-receptor interactions dataset and achieve 80.08% accuracy, 82.98% sensitivity and 80.02% specificity. Then, we test our extracted feature method on two Protein-Protein Interactions (PPIs) datasets, and achieve accuracies of 93.79% and 87.46%, respectively.

Conclusion: Our proposed method can be a useful tool for identifying of ligand-receptor interactions. Related data sets and source code are available at https://github.com/guofei-tju/ligand-receptorinteractions. git.

背景：配体-受体相互作用在细胞分化，增殖和免疫反应过程所需的信号转导中起重要作用。配体-受体相互作用的分析有助于深入了解细胞增殖/分化和其他细胞过程。

方法：该计算技术将用于促进未来蛋白质组学研究中的配体-受体相互作用研究。在本文中，我们提出了一种新的计算方法，可以通过机器学习方法来预测氨基酸序列中的配体-受体相互作用。我们通过定向梯度直方图（HOG）和离散余弦变换（DCT）从配体和受体序列中提取特征。然后，将这些特征输入到模糊C均值（FCM）聚类算法中进行聚类，并且我们得到多个训练子集以生成相同数量的子分类器。根据从一个样本到训练子集的相似性，我们选择一个最佳的子分类器来预测配体-受体的相互作用。

观察结果：为了验证性能，我们对配体-受体相互作用数据集进行了五次交叉验证实验，获得了80.08％的准确性，82.98％的敏感性和80.02％的特异性。然后，我们在两个蛋白质-蛋白质相互作用（PPI）数据集上测试提取的特征方法，并分别达到93.79％和87.46％的准确性。

结论：我们提出的方法可以作为鉴定配体-受体相互作用的有用工具。相关数据集和源代码可从https://github.com/guofei-tju/ligand-receptorinteractions获得。git。