PS 54-89
Landscape genomic analysis of single-nucleotide polymorphisms (SNPs) associated with geographic and climate variables in Hawaiian koa (Acacia koa)

Thursday, August 14, 2014
Exhibit Hall, Sacramento Convention Center
Jessica W. Wright, Pacific Southwest Research Station, USDA-Forest Service, Conservation of Biodiversity, Davis, CA
Christina T. Liang, USDA Forest Service, Pacific Southwest Research Station, Hilo, HI
Paul F. Gugger, Ecology and Evolutionary Biology, University of California, Los Angeles, Los Angeles, CA
Nicklos Dudley, Hawaii Agriculture Research Center, Kunia, HI
Tyler Jones, Hawaii Agriculture Research Center, Kunia, HI
Aileen Yeh, Hawaii Agriculture Research Center, Kunia, HI
Hardeep Rai, Dept of Wildland Resources, Utah State University, Logan, UT
Paul Hodgskiss, Pacific Southwest Research Station, USDA-Forest Service, Conservation of Biodiversity, Davis, CA
Victoria L. Sork, Ecology and Evolutionary Biology; Institute of the Environment and Sustainability, University of California, Los Angeles, Los Angeles, CA
Background/Question/Methods

Acacia koa is a keystone tree species native to the Hawaiian Islands. It provides food and habitat for a range of native birds and insects. Climate change is anticipated to impact the Hawaiian Islands. We focused on the landscape genomics of A. koa in Hawaiʻi to identify geographic patterns of genetic differentiation and genes associated with climate variation.  The goal of this study is to identify adaptive genetic variation that can inform future management projects to optimize the ability of koa populations to respond to climate change. We have sampled 220 trees across the geographic, elevational, and climate range of A. koa on the islands of Kauaʻi, Oʻahu, Maui, and Hawaiʻi.  To generate single nucleotide polymorphisms (SNPs) data for a large number of DNA fragments, we randomly subsampled the genome using genotyping-by-sequencing.  We then used the PRISM climate model to generate climate data for each of our sampled trees. We quantified the spatial, climatic, and geological forces that might structure SNP variation using a multivariate model.  We then used EMMAx, a linear mixed model, to identify individual SNPs extremely associated with climate variables, while controlling for the kinship structure among the trees.

Results/Conclusions

We found over 38,000 SNPs with representation in at least 90% of the 220 individuals.  Populations are genetically differentiated across islands, which span a wide range of climate environments. On Hawai’i (n=126), preliminary analyses show overall SNP variation is strongly associated with mean annual precipitation, temperature seasonality, the age of the volcanic rock substrate, altitude, as well as other climate and spatial variables.  A number of SNPs were extremely associated with climate and geographic variables: 2 with altitude, 21 with mean precipitation, 5 with precipitation seasonality, 2 with mean temperature, and 2 with temperature seasonality.  These SNPs may play a role in local adaptation to climate and our preliminary analyses implicate precipitation-related stresses as a major selective force on the island of Hawai’i.  Geographic patterns of adaptive genetic variation within and among islands can be incorporated into land management plans for forest trees under threat of changing climate.