Transcript 1420-1440 Butcher NZIF Conference ppt 888 KB

What gains can we expect from Genetics?
John A Butcher
Radiata Pine Breeding Company
What are the past contributions of Genetics?
Breed
Seedlot
Volume
Gain (%)
Acceptability
(%)
GF Rating
0
45
1
5-10
50
7
850 orchard
13-18
65
14
268 orchard
15-20
70
16
Top 16 268
19-23
70
19
CP 268
27-32
80
23
Unimproved
Climbing
Select
Increased genetic gain in predictive models
300 Index
35
On average, an increase in GFPlus growth
rating of 1 unit corresponds to a 1%
increase in stem volume growth
30
25
300 Index = 22 + 0.26 GFPlus
20
5
10
15
20
GFPlus Growth rating
25
30
On average, an increase in GFPlus density
rating of 1 unit corresponds to an increase
of 1.85 kg/m3 in wood density
Density Index (kg/m3)
550
Density Index = 398 + 1.85 GFPlus
500
450
400
350
-20
0
20
GF Plus rating
40
60
Long term Gain Trials
Yield gain achieved in a sawlog regime. Genetic Gain Trial kaingaroa 1210
100000
1300
95000
1200
y = 1204.3x + 70705
1100
R 2 = 0.8914
85000
1000
80000
900
y = 11.21x + 743.94
75000
800
R2 = 0.8491
70000
700
65000
600
Total Value
60000
Total Recoverable
Volume
55000
50000
500
400
300
0
5
10
15
GF Rating
20
25
Total Recoverable Volume M3
Total value $/hectare
90000
Forest Industry Drivers for Tree Breeding
Greater productivity
Less variability across and between stems
Enhanced wood properties
Improved disease resistance
Increased profitability
How can Tree Breeding and Genetics respond?
High Impact Technology Developments
Genomic Selection
Selection
Genomic
Genotype to Site Matching
Genomic Selection
A step-change in animal and plant breeding
Increasingly used in animal breeding:chickens, sheep, dairy cattle
Increasingly used in plant breeding:horticulture, crops, pasture
Now being introduced to forest trees:eucalypts, poplar, spruce, pines
Genomic Selection is not a speculative technology
The Genomic Selection Programme
MBIE Partnership Programme with RPBC
$5 million for 5 years
Scion our major science provider
Dr John Hay Project Manager
Oversight by RPBC Advisory Board of international experts
Well linked nationally and internationally (Australia, USA,
Canada, UK, France, Scandinavia, Chile)
Genomic Selection
Uses genomic information to estimate breeding values
(GeBVs) rather than just phenotypic information (BVs)
Eliminates the need for progeny testing and thus reduces the
breeding and deployment cycle
Halves the breeding and development cycle
Doubles genetic gain per unit time
More rapid turnover of generations
Some additional benefits of genomic selection
Better ability to identify genotypes with traits of interest
Ability to assess all traits in year 1
Ability to look at multiple traits simultaneously and understand interactions in
multiple trait selection
Overcomes issues with costly or “difficult to measure” phenotypic traits
Allows screening for resistance to exotic pests and diseases by simple
screening of DNA extracts from susceptible and resistant genotypes
Provides a greater opportunity to respond to individual grower needs
How does Genomic Selection Work?
Genes for Trait 1
SNP Markers
Saturate the Genome with Markers
Make a SNP Panel
Genes for Trait 2
How the SNP Panel is trained
Training Set
1000 clones
with
phenotype
and
genotype
Training
Population
800 clones
with
phenotype
and
genotype
Prediction
model
200 clones
with
genotype
alone
Validation Set
Predict
phenotypes
based on
genotypes alone
Initial radiata training population of ca. 1000 clones
Target training population of 2000-4000 clones
How does the SNP Panel Work?
Genes for Trait 1
Genes for Trait 2
SNP Markers
DNA from genotype of
unknown phenotype
Compare marker pattern (Genotype) and trait pattern (Phenotype)
Use patterns to predict phenotype
Produce Genomic Breeding Values
New crosses
0
Breeding Test
5
Progeny Test
10
15
Bulk
Seed
20
Current Situation Backward Selection
New crosses
0
Breeding Test
5
Bulk
10
15
Seed
20
Impact of forward selection
New crosses
0
GS
Regional clonal tests
5
10
Cuttings
15
GS plus regional clonal tests
New GS
crosses
0
cuttings
5
GS alone
Future possibility
Current R&D Target
30
Impact on Breeding and Deployment Timelines
Slash Pine in Florida
Matias Kirst
Top
grafting
SE & rooted
cuttings
Genomics
What it means to optimising returns from tree
breeding
Base Case: Genetic Gains achieved
33 years to first deployment of
new germplasm
2010
2030
10% volume increase
10 kg/m³ density increase
26 years to first harvest
of new germplasm
2050
2070
$168 million additional gross revenue (mill door/wharf gate)
Each year for 20 years from 2069
PV of Benefit = $ 19 million
What it means to optimising returns from tree
breeding
Genetic Gains achieved
Breeding/Deployment cycle reduced by 17 years
15 years to first deployment 26 years to first harvest
of new germplasm
of new germplasm
2010
2030
2050
2070
$168 million additional gross revenue (mill door/wharf gate)
Each year for 20 years from 2051
PV of Benefit = $ 76 million
Greater gain with Genomic election
GENOMIC SELECTION with
forward selection
Percentage Gain
50
First harvest from
Genomic Selections
40
30
Implementation of
Genomic Selection
Additional gain from
genomic selection
20
10
Gain from existing
genetic improvement
2010
2030
CONVENTIONAL BREEDING
with backward selection
2050
Time
2070
Expectations from the Genomic Selection Programme
Improved Dothistroma resistance available from 2018
Top performing selections of 1997/99 elites available from 2018
Top performing selections of 2013/14 elites available from 2022
New selections form 2018/19 elites available 2028
Initial focus (2018 deliverables) on Volume and Dothistroma
Anticipated gains at harvest in ca. 2050
10-15% gain already in developing commercial estate
15% additional from genomics
then an expectation of an additional 10% gain every 7-8 Years
The Future
Genomic Selection will become mainstream in tree breeding
Timelines of the breeding and deployment cycle will be further
reduced
Cost reductions in genotyping will make genomic selection a tool
that could be used for individual companies to drive customised
breeding initiatives
Genomic Selection will be a key component in the implementation
of genotype to site matching and the exploitation of GxE