Machine learning in biological research: key algorithms, applications, and future directions

Machine learning is a robust framework to analyze questions using complex data in a variety of fields. We present definitions and recent applications of four key machine learning methods and discuss their advantages and challenges in biological research. Through a set of systematically selected case studies, we highlight how machine learning models have been used in a range of applications, including phylogenomics, disease prediction, and host taxonomy prediction. We identify additional potential areas of integration of machine learning into questions with biological relevance. This intersection can be further enhanced through collaboration and innovation on parallelization, interpretability, and preprocessing.