Fer-COCL: A Novel Method Based on Multiple Deep Learning Algorithms for Identifying Fertility-Related Proteins

The survival of species depends on the fertility organisms. It is also worthwhile to study proteins that can regulate reproductive activity Since biological experiments are laborious confirm proteins, it has become a priority develop relevant computational models predict function fertility-related proteins. With development machine learning, pertinent various algorithms be key identifying In this work, we model Fer-COCL based deep learning. consists multiple features as well learning algorithms. First, extract using Amino acid composition (AAC), Dipeptide (DPC), CTD transition (CTDT) and deviation between dipeptide expected mean (DDE). After that, spliced fed into classifier. data processed jointly by convolutional neural network long short-term memory input fully connected layer for classification. evaluating 10-fold cross-validation, accuracy two sets reaches 97.1% 98.3%, respectively. results indicate efficient accurate, facilitating biologists' research fertility. addition, free online tool predicting available at http://fercocl.zhanglab.site/.