Let's hope your couch is shaped like an old-fashioned telephone receiver. A graphical representation of what a real Gerver's sofa might look like. Breakthroughs, discoveries, and DIY tips sent every weekday. If you've ever struggled to squeeze a couch around a tight corner while moving into a new apartment, you'll probably find that the pure mathematics problem known as the "sofa problem" is incredibly relatable. The question seeks to find a maximum value for the area of a sofa that can slide around a 90-degree corner in a corridor of a given width.

The structure of human interleukin-12 protein bound to its receptor, as predicted by machine-learning software.Credit: Ian Haydon, UW Medicine Institute for Protein Design Software that accurately determines the 3D shape of proteins is set to become widely available to scientists. On 15 July, the London-based company DeepMind released an open-source version of its deep-learning neural network AlphaFold 2 and described its approach in a paper in Nature1. The network dominated a protein-structure prediction competition last year. Meanwhile, an academic team has developed its own protein-prediction tool inspired by AlphaFold 2, which is already gaining popularity with scientists. That system, called RoseTTaFold, performs nearly as well as AlphaFold 2, and is described in a Science paper also published on 15 July2.