Machine learning for plant microRNA prediction: A systematic review
It provides a comprehensive overview for researchers in bioinformatics to identify gaps and improve miRNA prediction in plants, but it is incremental as it reviews existing studies.
This systematic review examines machine learning methods for predicting plant microRNAs, addressing the high cost and time of experimental determination, and concludes that plant-specific computational approaches are needed.
MicroRNAs (miRNAs) are endogenous small non-coding RNAs that play an important role in post-transcriptional gene regulation. However, the experimental determination of miRNA sequence and structure is both expensive and time-consuming. Therefore, computational and machine learning-based approaches have been adopted to predict novel microRNAs. With the involvement of data science and machine learning in biology, multiple research studies have been conducted to find microRNAs with different computational methods and different miRNA features. Multiple approaches are discussed in detail considering the learning algorithm/s used, features considered, dataset/s used and the criteria used in evaluations. This systematic review focuses on the machine learning methods developed for miRNA identification in plants. This will help researchers to gain a detailed idea about past studies and identify novel paths that solve drawbacks occurred in past studies. Our findings highlight the need for plant-specific computational methods for miRNA identification.