Nature: Tuberculosis kills more than 1 million people globally each year, and many strains of TB are becoming drug resistant. A synthetic compound that works differently from current treatments may provide an answer. Kevin Pethe of the Pasteur Institute Korea and his colleagues spent five years investigating some 100 000 compounds for their ability to inhibit the growth of TB in mice. They found just a single compound suitable for extended testing. It works by inhibiting the TB bacteria from synthesizing adenosine triphosphate, which powers the bacteria’s growth. Although the compound is now entering phase 1 clinical trials, only 5% of all drugs that reach that stage ever reach the market.