Our group conducts research on forest genetics and tree breeding through understanding and dissecting genetic base of genetic variation for quantitative traits, including tree growth and form traits, wood quality traits, phenology traits, and biotic and abiotic resistance traits. The starting point is to investigate the genetic base of the phenotypic variation we observed between individual trees in forest stands. The first question we ask is how much of this variation between individual trees is due to genetic differences, and how much due to environmental factors. Then we need to find out the number of genes that are responsible for the genetic variation, and how these genes interact to influence the performance of trees in different environments. This basic knowledge can then be used in the design of breeding programmes for increased growth rate and quality of wood products in forest plantations.

Wu Harry 1150To design efficient breeding programmes and to increase genetic gain, we also need to identify the best native tree populations for selection and breeding for a specific forest region. This involves research on genotype by environment interaction and on response curves of existing populations and genotypes, so that we can delineate optimal breeding trees (population) that match the environment (breeding zone). After assembling a breeding population for further improvement, we need to design the best breeding strategy. We not only have to select the individuals that we want to use as a parent, to produce the best progeny, we also need to identify the ideal combinations of parents (mating design). Furthermore, we need to deal with inbreeding depression.

Another important question we need to address is how to deal with improvement of multiple traits that are adversely related, such as wood quantity vs. wood quality. First we need to find out the genetic causes of such correlation, using quantitative and molecular tools. Then we develop a gene model (locus and parametric model) and use simulations to identify the best selection and mating methods for dealing with such adversely correlated multiple traits. We also study the suitability of different mating and selection methods when it comes to avoiding or removing inbreeding depression in advanced breeding programmes.
Sampling for phenology and DNA in Scots pine Inspecting Eucalypt Plus Tree in China
Sampling for phenology observation and DNA in Scots pine  Plus tree in Eucalypts.

With the advance of gene sequencing technology, we start to study associations between individual genes or gene complexes with phenotype variation in trees. We study candidate genes or genome-wide approach to increase knowledge about the association between variations in DNA and observable phenotype variation. In addition, we are developing a quantitative genetics tool that integrates DNA sequence-based variation with phenotypic data, in order to improve the efficiency of genetic improvement in trees. This involves the development of advanced methods for breeding value prediction, such as genomic Best Linear Unbiased Prediction (G-BLUP) and Genome-wide Breeding Value (GBV).