MLCDForest: multi-label classification with deep forest in disease prediction for long non-coding RNAs.

Area of Science:

• Genomics
• Bioinformatics
• Computational Biology

Background:

• Long non-coding RNAs (lncRNAs) are increasingly linked to human diseases.
• Accurate prediction of disease and tissue based on lncRNA data is crucial for diagnosis and therapy.
• Existing machine learning models require enhanced accuracy and robustness.

Purpose of the Study:

• To develop an advanced deep learning model for multi-label classification of lncRNA tissue prediction.
• To implement a novel deep forest approach for lncRNA-based disease and tissue prediction.
• To improve the accuracy and robustness of predictive models in lncRNA research.

Main Methods:

• Proposed a deep learning model named Multi-Label Classifications with Deep Forest (MLCDForest).
• MLCDForest employs a sequential multi-label-grained scanning method.
• The model incorporates label correlation during sequential training for multi-label classification.

Main Results:

• MLCDForest demonstrated superior performance compared to state-of-the-art methods in disease prediction using lncRNA-disease association datasets.
• The model achieved high accuracy in tissue prediction for given lncRNAs.
• Systematic comparisons confirmed the effectiveness of the proposed sequential scanning and label correlation approach.

Conclusions:

• MLCDForest offers a powerful new tool for multi-label classification and tissue prediction based on lncRNA data.
• The model's ability to consider label correlation enhances predictive capabilities.
• This approach advances the application of artificial intelligence in understanding lncRNA functions and disease associations.