Biology & Philosophy Гуманитарные и социальные науки
Measuring ancestry
Аннотация
In the past decades, there has been a significant shift in the labels used to describe human populations in large-scale genomic studies. Such labels are known as population descriptors. The once prevalent descriptor “race” has progressively given way to the concept of “ancestry”. Ancestry is measured through genetic similarity: individuals who are genetically similar are considered to have shared ancestry. The use of rigorous techniques—including measurement procedures—to assess “ancestry” give it a cloak of objectivity, which in turn justifies the widespread adoption of the concept. However, this objectivity might only be apparent once we understand the challenges to measuring ancestry. This paper contributes to the literature by bridging the previously disconnected fields of philosophy of measurement and human population genetics. To do so, I draw on model-based accounts of measurement to describe practices of measuring ancestry. I then outline two measurement challenges in clustering humans based on genetic ancestry. The first is the challenge of coordination which describes the circularity between theoretical assumptions and measurement outcomes, also known as the theory-ladenness of observation. The second is the issue of biases, which include the problems of sampling biases and the issue of proxy variables. Finally, I outline potential solutions to address these challenges.