Deep learning in bioinformatics.

Area of Science:

• Bioinformatics
• Computational Biology
• Artificial Intelligence

Background:

• The transformation of biomedical big data into actionable knowledge presents a significant challenge in bioinformatics.
• Deep learning (DL) has achieved state-of-the-art performance across various fields since the early 2000s.
• The application of DL in bioinformatics is increasingly emphasized in both academic and industrial research.

Purpose of the Study:

• To provide a comprehensive review of deep learning applications in bioinformatics.
• To categorize current research by bioinformatics domain and DL architecture.
• To discuss theoretical and practical challenges and suggest future research directions.

Main Methods:

• Systematic review of deep learning research in bioinformatics.
• Categorization of studies by domain: omics, biomedical imaging, and biomedical signal processing.
• Categorization of studies by deep learning architecture: deep neural networks, convolutional neural networks, recurrent neural networks, and emergent architectures.

Main Results:

• Examples of current deep learning research in bioinformatics are presented.
• Research is categorized across diverse bioinformatics domains and various deep learning architectures.
• Theoretical and practical issues are discussed, alongside potential future research avenues.

Conclusions:

• Deep learning offers powerful approaches for gaining insights from complex biomedical big data.
• This review serves as a valuable resource for researchers applying deep learning in bioinformatics.
• Further exploration of deep learning is crucial for advancing bioinformatics and biomedical research.