Over the past decade the face of genetics has changed. The quantity of data that is being produced, coupled with an increase in computational power, means that information can be stored and analysed to an unprecedented degree. Fields as disparate as population genetics, systems biology and the mathematical modelling of development rely on quantitative descriptions, and statistical analysis — once a foreign concept to mainstream genetics — is pervading every aspect of the field.

The four Reviews in this Focus issue on Statistical Analysis aim to familiarize readers with the current and debated issues in this area of research. David Balding's article (page 781) is a primer on the statistical issues that concern the field of population association studies. This broad article touches on all aspects of study design, from the preliminary checks on data quality to the multiple-testing problem — the “bane of statistical genetics”. Large-scale molecular variation data demand powerful analysis methods, and Paul Marjoram and Simon Tavaré (page 759) describe the transition towards computational methodologies. Many statistical analysis programs and packages have emerged in various corners of the web, and Laurent Excoffier and Gerald Heckel (page 745) guide users to the most appropriate program for their study. Last, Bruce Weir and colleagues (page 771) discuss the advances in relatedness analysis, which has applications as far ranging as human linkage studies, forensics and animal and plant breeding.

The dialogue on the best solutions to statistical problems is ongoing, and all authors present their opinion on future trends.

This special issue is associated with an audio supplement, in which experts talk about the growing relevance of statistics for genetics (http://www.nature.com/nrg/focus/stats/audio). As usual, the issue is accompanied by a Web Focus (http://www.nature.com/nrg/focus/stats), an online collection of articles on Statistical Analysis that have been published by Nature Publishing Group journals.