Salient Features, Data and Algorithms for microRNA Screening from Plants: A Review on the Gains and Pitfalls of Machine Learning Techniques | Bentham Science

摘要

The era of big data and high-throughput genomic technology has enabled scientists tohave a clear view of plant genomic profiles. However, it has also led to a massive need forcomputational tools and strategies to interpret this data. In this scenario of huge data inflow,machine learning (ML) approaches are emerging to be the most promising for analysingheterogeneous and unstructured biological datasets. Extending its application to healthcare andagriculture, ML approaches are being useful for microRNA (miRNA) screening as well.Identification of miRNAs is a crucial step towards understanding post-transcriptional generegulation and miRNA-related pathology. The use of ML tools is becoming indispensable inanalysing such data and identifying species-specific, non-conserved miRNA. However, thesetechniques have their own benefits and lacunas. In this review, we will discuss the current scenarioand pitfalls of ML-based tools for plant miRNA identification and provide some insights into theimportant features, the need for deep learning models and direction in which studies are needed.