Machine Learning "Toolbox" for Protein Analysis

A novel machine learning "toolbox" that can read and analyse the sequences of proteins has been described in the open-access journal eLife. The study demonstrates that, when trained to read sequence data, artificial neural networks called Restricted Boltzmann Machines (RBM) can provide a wealth of information on protein structure, function and evolutionary features. It is believed to be the first method that can extract this level of detail from sequence data alone. Proteins are formed of sequences of molecules called amino acids, which determine a given protein's structural and functional properties. But understanding which parts of the sequences are responsible for which properties is challenging.