35th NPS abstract book

P11
Gene expression and natural selection shape the evolution of protein-coding
genes in Picea
A. R. DE LA TORRE 1, 2 , Y-C. LIN 3 , .Y. VAN DE PEER 3, 4 and P. INGVARSSON 1, 2
1 Department of Ecology and Environmental Science, Umeå University, Linneaus väg 6, SE-901 87
Umeå, Sweden; 2 Umeå Plant Science Centre, Umeå, Sweden; 3 Department of Plant Systems Biology,
VIB and Department of Plant Biotechnology and Bioinformatics, Ghent University, Technologiepark
927, 9052 Ghent, Belgium; 4 Genomics Research Institute, University of Pretoria, Hatfield Campus,
Pretoria 0028, South Africa
The role of natural selection on the evolution of gene expression levels remains largely unknown in
non-model species. In this study, we use the first two fully sequenced representatives of the
gymnosperm plant clade (Picea abies and Picea glauca) to test for the evidence of expressionmediated
selection. We used whole-genome gene expression data (>50,000 genes) to study the
relationship between gene expression, codon bias, rates of sequence divergence, protein length,
pathway position and gene duplication. We found that gene expression correlates with rates of
sequence divergence and codon bias for translational efficiency. A strong correlation between gene
expression and gene duplication was found, with genes in large multi-copy gene families having, on
average, a lower expression level and breadth, lower codon bias, and higher rates of sequence
divergence than single-copy gene families. A correlation between pathway position and gene
duplication was also found. Single-copy genes encoded essential biological functions and were under
strong selective pressures to maintain their copy number in Picea. In contrast, large paralogous gene
families had great expression divergence and higher levels of tissue-specific genes. Our study
highlights the importance of gene expression and natural selection in shaping the evolution of
protein-coding genes in Picea species.
P12
Forest tree improvement: the shift from quantitative genetics to quantitative
genomics
Y. A. EL-KASSABY and J. KLÁPŠTĚ
Department of Forest and Conservation Sciences, Faculty of Forestry, The University of British
Columbia, 2424 Main Mall, Vancouver, BC, V6T 1Z4, Canada
Traditional and Next Generation Sequencing technologies increased the availability of sequence data
for model and non-model forest tree species. This in turn transformed quantitative genetics from
the pedigree-based founded on the utilization of Sewell Wright’s coefficient of relationship between
individuals (Wright S. 1922; Amr Nat 56:330-338) to a genomic-based realized kinship method for
estimating individuals’ breeding values and traits’ heritability and genetic correlation. This
substantial shift literally changed quantitative genetics to quantitative genomics and led to the
development of innovative methods such as the pedigree-free and the unified single-step (a
combination of pedigree and genomic realized kinship) evaluation approaches where the classical
Best Linear Unbiased Predictor (BLUP) is replaced by the Genomic Best Linear Unbiased Predictor
(GBLUP). The advantages of quantitative genomics offered opportunities for obtaining more precise
genetic parameters, better partition of the genetic variance, and even breeding without structured
pedigree. Examples from unstructured black cottonwood and white spruce open-pollinated
populations will be presented to demonstrate the unsurpassed potential of incorporating sequence
data in classical breeding.
36