Noninvasive collection of samples from wild mammal populations presents an extraordinary opportunity to conduct genetic analysis without necessarily capturing or even directly observing the subjects of interest. Unfortunately, the hair or feces samples typically collected present formidable challenges in the laboratory. The investment of time and expense required makes careful experimental design a necessity. Researchers who want to examine the genetics of wild mammal populations have great interest in the use of noninvasively collected sample materials. Questions that can potentially be answered range from population-level inquiries into such topics as population size, sex composition, individual relationship, and dispersal, to comparisons between populations providing information on geographic differentiation, comparative levels of genetic diversity, and phylogenetic relationships.1 Effective investigation of these matters relies on an ability to examine genetic variation from the nuclear genome as well as the maternally transmitted mitochondrial DNA (mtDNA). A typical approach to examining the genetic structure of a population involves the assessment of variation at multiple nuclear microsatellite loci. Alleles of these microsatellite loci vary in the length of simple repeat elements, hence allowing for the potential assemblage, when enough loci are examined, of individually distinctive genotypes.2 I will therefore focus here on microsatellite analyses. Numerous studies have employed microsatellite analysis of noninvasive samples in the form of feces or shed hair.3–8 But, as Taberlet et al.9 recently noted, to date there have been relatively few “comprehensive” studies using nuclear DNA from noninvasive samples. This is surprising in view of the existence of the large body of literature on the topic, and the subjective impression that many field researchers are equipped with collection materials and a plan or at least a hope of arranging for later genetic analysis. The indication is thus that, years after it was heralded as representing a revolution in field studies of primates,10 noninvasive sampling for genetic analysis either has not yet reached its potential or has proved disappointing. Ironically, the simplicity and low costs of the polymerase chain reaction (PCR) that have led to its ubiquitous use in molecular analysis laboratories may have led to an underestimation by laboratory and field researchers alike of the difficulties inherent in the use of noninvasive samples. Such analyses in fact present considerable technical challenges that can only be overcome by investing a great deal of time, expense, and laboratory expertise. Most technical considerations involved in the production of accurate genotypes from noninvasive samples have been thoroughly addressed elsewhere.9,11 In this contribution, I will present some preliminary results from a study of habituated communities of chimpanzees (Pan troglodytes verus) in the Tai National Forest, Ivory Coast. The necessary investment, particularly in terms of time and expense, of generating reliable genotypes from a large number of individuals will be described.
Read full abstract