ncPro-ML: An integrated computational tool for identifying non-coding RNA promoters in multiple species

Area of Science:

• Genomics
• Computational Biology
• Molecular Genetics

Background:

• Promoter identification is crucial for understanding gene regulation mechanisms.
• Existing computational methods struggle to identify promoters for non-coding RNAs (ncRNAs).
• Experimental methods for promoter identification are often costly and inefficient.

Purpose of the Study:

• To develop an accurate computational tool, ncPro-ML, for identifying ncRNA promoters.
• To evaluate the performance of ncPro-ML in human (Homo sapiens) and mouse (Mus musculus) species.
• To provide a freely accessible web-server for the ncPro-ML tool.

Main Methods:

• Developed ncPro-ML, a machine learning-based predictor for ncRNA promoter identification.
• Utilized various sequence encoding schemes to convert DNA sequences into feature vectors.
• Tested the impact of sequence length on predictor performance using datasets of varying lengths.
• Evaluated performance using independent datasets and cross-species testing.

Main Results:

• ncPro-ML demonstrated optimal performance with a sequence length of 221 nucleotides for both human and mouse.
• The predictor achieved satisfying performance in independent dataset tests.
• Cross-species testing confirmed the robustness and generalizability of ncPro-ML.
• A web-server for ncPro-ML was developed and made publicly available.

Conclusions:

• ncPro-ML is a powerful and accurate tool for the discovery of ncRNA promoters.
• The developed method overcomes limitations of existing tools in identifying ncRNA promoters.
• The publicly accessible web-server facilitates broader research in ncRNA gene regulation.