Nature
-
How can a machine learn from experience? Probabilistic modelling provides a framework for understanding what learning is, and has therefore emerged as one of the principal theoretical and practical approaches for designing machines that learn from data acquired through experience. The probabilistic framework, which describes how to represent and manipulate uncertainty about models and predictions, has a central role in scientific data analysis, machine learning, robotics, cognitive science and artificial intelligence. This Review provides an introduction to this framework, and discusses some of the state-of-the-art advances in the field, namely, probabilistic programming, Bayesian optimization, data compression and automatic model discovery.
-
Cancer cells have been at the centre of cell metabolism research, but the metabolism of stromal and immune cells has received less attention. Nonetheless, these cells influence the progression of malignant, inflammatory and metabolic disorders. Here we discuss the metabolic adaptations of stromal and immune cells in health and disease, and highlight how metabolism determines their differentiation and function.
-
Sphingolipids are ubiquitous building blocks of eukaryotic cell membranes. Progress in our understanding of sphingolipid metabolism, state-of-the-art sphingolipidomic approaches and animal models have generated a large body of evidence demonstrating that sphingolipid metabolites, particularly ceramide and sphingosine-1-phosphate, are signalling molecules that regulate a diverse range of cellular processes that are important in immunity, inflammation and inflammatory disorders. Recent insights into the molecular mechanisms of action of sphingolipid metabolites and new perspectives on their roles in regulating chronic inflammation have been reported. The knowledge gained in this emerging field will aid in the development of new therapeutic options for inflammatory disorders.
-
Advances in our understanding of the mechanisms that bring about the resolution of acute inflammation have uncovered a new genus of pro-resolving lipid mediators that include the lipoxin, resolvin, protectin and maresin families, collectively called specialized pro-resolving mediators. Synthetic versions of these mediators have potent bioactions when administered in vivo. ⋯ Although they have been identified in inflammation resolution, specialized pro-resolving mediators are conserved structures that also function in host defence, pain, organ protection and tissue remodelling. This Review covers the mechanisms of specialized pro-resolving mediators and omega-3 essential fatty acid pathways that could help us to understand their physiological functions.
-
Recent reports have described an intricate interplay among diverse RNA species, including protein-coding messenger RNAs and non-coding RNAs such as long non-coding RNAs, pseudogenes and circular RNAs. These RNA transcripts act as competing endogenous RNAs (ceRNAs) or natural microRNA sponges - they communicate with and co-regulate each other by competing for binding to shared microRNAs, a family of small non-coding RNAs that are important post-transcriptional regulators of gene expression. Understanding this novel RNA crosstalk will lead to significant insight into gene regulatory networks and have implications in human development and disease.