Detecting signals of recent selection using ancient DNA data

Friday seminar by Mark Thomas from University College London


Almost all human populations have experienced fundamental and rapid shifts in ecology over the last 10,000 years, perhaps more than at any other point in human evolution. These changes were mostly driven by the Neolithic transition; a move from mobile hunting and gathering to sedentary food production, and occurred in the Holocene – a period of relative climate stability. Key changes include a reduction in dietary breadth and a shift to high carbohydrate staples, increased population densities, urbanization, more hierarchal societies, proximity to animals, and long-distance trade. These changes had significant effects on morbidity and mortality in the past – including increases in pathogen loads and a reduction in nutrient balance of most diets – and are implicated in a considerable proportion of global disease burden today. The detection of signatures of natural selection provides a means of identifying medically relevant genetic variation in past populations, since it identifies variants that affected survival, and the potential to determine the extent to which we have, or have not, adapted to these fundamental ecological shifts. A number of methods for detecting signatures of natural selection using genomic data have been developed in recent years. However, all of these methods are indirect and have poor sensitivity and temporal resolution, most are confounded by past demographic processes, and many are insensitive to selection acting on standing variation. Here I present an approach to detecting natural selection using ancient DNA data, and provide examples of its use to study the evolution of lactase persistence and skin, hair and eye pigmentation. This method is direct, considerably more sensitive than existing methods, and permits the temporal resolution of episodic selection.

Mark Thomas
Professor of Evolutionary Genetics
Research Department of Genetics, Evolution and Environment University College London Gower Street London WC1E 6BT



Mark Thomas has worked extensively on understanding how humans have evolved and migrated around the World. He has used genetic data – including ancient DNA – computer simulations and archaeological information to examined the origins and past migrations of a number of specific human populations including Jewish and Judaic groups, British populations and a number of enigmatic European and African peoples. In recent years he has worked on using 14C data as a proxy for past demography, on modelling cultural evolution to better understand the origins of modern human behaviour, and to examine ethnic structuring in past populations, on recent natural selection using genetic data – particularly in relation to diet and infectious disease – and on gene-culture co-evolution, particularly the origins of lactase persistence and dairying in Europe and Africa.

Published May 6, 2014 8:44 AM - Last modified May 6, 2014 11:01 AM