Perspectives in 2018

This paper presents standards and best practices for reporting genome sequences of uncultivated viruses.

Brendan Frey and colleagues provide a personal overview of the machine learning field and in particular deep learning. They outline the technical challenges in applying machine learning to different types of biological and biomedical data and go on to discuss the challenges in implementing these approaches in the clinical realm, in drug discovery programs and within regulatory agencies.

Machine learning and natural language processing methods are applied to learn from PubMed searches and improve user experience.

Metabolomics activity screening can identify small-molecule metabolites that are readily available and can be used to alter phenotypes, including stem cell differentiation, T-cell survival, oligodendrocyte maturation, and insulin signaling.