9718

• UCREFRP Catalog Number: 9718
• Author: Strand, A.
• Title: Analysis of a preliminary microsatellite dataset for wild and captive populations of humpback chub, Gila cypha.
• USFW Year: 2006.
• USFW - Doc Type: Grice Marine Laboratory, College of Charleston
• Copyright Material: NO

A
DRAFT
4 Results and Conclusions
Theoretical results indicate that in those populations (with the exception of Boulders) that
were sampled, it is likely that the majority of genetic variation was obtained. As a result,
it seems reasonable to compare these wild-sampled populations to the captive population to
construct a test of whether the captive population represents wild stock.
Initial examination of the raw microsatellite data indicated that they are well suited to
estimate neutral genetic variation in the captive and wild populations. This examination was
followed by binning of alleles to facilitate the rest of the analyses presented in this report.
Examination of Hardy-Weinberg equilibrium revealed that the loci devised for bonytail
do not result in null alleles in Humpback chub. Linkage disequilibrium analyses also indicate
that loci can be treated as genetically independent markers.
To address the question of whether the captive population represents the wild sampled
fish, I ran a genotypic and allelic comparison. The genotype frequencies did differ among
populations, but it is unclear whether this difference would persist if subsequent generations
from each population were analyzed. In other words, this difference may be due to the
particulars of segregation in the generations sampled. The pairwise allelic analysis indicates
that there is very little difference between the wild-sampled and captive population. The one
exception is Boulders, and this population is not well sampled. My overall interpretation of
these data is that the genetic markers in the captive population adequately, but not perfectly,
represent the genetic variation present in wild populations of Humpback chub. As a result,
I believe that the Willow Beach broodstock is an important source of genetic material, and
maintaining this broodstock should be a management priority.
4.1 Recommendations for future work
To develop a stronger case for the conclusions above, I would recommend several things.
First, it is important to sample all wild populations sufficiently to provide a comparison to
the captive population. In particular, Boulders could be better represented in this dataset.
Further field sampling efforts should also target Salt Camp and Coyote Camp populations,
if possible, to capture changes that may have occurred among demographic cohorts. Second,
additional loci would provide better information on which to base inference. Eleven loci
were developed for bonytail and have been shown to amplify IN- umpback chub genomic DNA
(Keeler-Foster et al.; 2004). Therefore, it may not be too difficult to add to the loci already
characterized in this study. Finally, for the remaining loci, especially those shown to have
large amounts of missing data such as locus 226, it would be advisable to increase the number
of individuals with resolved alleles.
Finally, I believe that it is important to consider quantitative variation in the captive
and wild populations. If possible, it would be advisable to develop and analyze a dataset
that included both the marker genetics and phenotypes of both wild and captive fish. Such
phenotypes could include standard lengths, mass, fin ray counts, and other relatively easy
to assess morphological characteristics.
8