Computational prediction of disease related lncRNAs using machine learning

Area of Science:

• Genomics
• Bioinformatics
• Computational Biology

Background:

• Long non-coding RNAs (lncRNAs) are crucial regulators of biological processes.
• Dysregulation and mutations in lncRNAs are implicated in complex diseases.
• Accurate identification of lncRNA-disease associations is vital for therapeutic development.

Purpose of the Study:

• To develop an effective machine learning model for predicting disease-related lncRNAs.
• To integrate sequence and structure-based features for enhanced prediction accuracy.
• To address limitations in existing methods, such as redundancy and class imbalance.

Main Methods:

• Developed a machine learning model combining sequence and structure-based features of lncRNAs.
• Utilized Support Vector Machine (SVM) and Random Forest classifiers for training.
• Implemented redundancy checking and oversampling techniques for data balancing.

Main Results:

• Achieved the highest F1 score of 76% using the SVM classifier.
• Demonstrated improved performance compared to state-of-the-art methods for lncRNA-disease association prediction.
• Highlighted the significant contribution of lncRNA sequence mutations to prediction accuracy.

Conclusions:

• The developed machine learning model effectively predicts disease-related lncRNAs.
• Combining diverse features, particularly sequence mutations, enhances prediction capabilities.
• The study provides a more robust approach for identifying lncRNA-disease associations, aiding future research and treatment strategies.