Planning rice breeding programs for impact

1
Planning rice breeding programs
for impact

• Choosing parents and
• managing a pedigree breeding program

2
Learning objectives

• Strategies for choosing parents
• Types of crosses
• F2 population size
• Describing lines and crosses
• Pedigree versus bulk selection
• Traits for early-generation selection
• How many generations?
• Field design for pedigree nurseries

3
Question What is important when choosing
parents?

• At least 1 adapted parent should be used
• Improved donors should be used to avoid linkage
  drag
• Pre-breeding of improved donors is important
• MAS allows major gene to be moved with less
  linkage drag
• At least 1 high-quality parent should be used
• Backcross to high-quality parent may be necessary

4
Types of crosses

• 2-way, 3-way (topcrosses), and double crosses
  can be used if lines are of similar performance
  or quality, but break up adapted linkage blocks
  and epistatic combinations

5
Types of crosses

• Backcross populations
• When use them?
• ?if the objective is to improve one adapted
  parent with genes from a donor
• (BC1 and BC2 populations have proven very useful
  in developing lines with improved drought stress
  tolerance in IR64 background at IRRI)

6
Backcrossing with selection to improve IR64 for
drought
Cross
recurrent parent
Plants crossed
Selection
IR64 x donor
F1 x IR64
50
1-5 crossed
None
20 crossed
None
BC1F1 x IR64
75
20 crossed
None
BC2F1 x IR64
87.5
X
20 selfed
None
87.5
BC2F2
X
BC2F3
Screened under severe stress
87.5
2000 selfed
7
Question How many crosses should one make?

• Most programs make too many crosses, examine
  populations that are too small
• Choosing parents carefully is more important than
  making many crosses.
• Some programs make very few crosses with diverse
  parents
• E.g. Witcombe and collaborators in India and
  Nepal
• Crossed Kalinga III/IR64
• Farmers selected both upland- and lowland-adapted
  cultivars from the cross

8
Describing crosses

• Parents are separate by slashes (/)
• e.g. IR 64/Swarna
• In complex pedigrees, additional slashes are used
  to describe more recent crosses.
• e.g. IR64/Swarna//PSBRC 80 (PSBRC 80 was crossed
  to plants derived from IR64/Swarna)
• Backcrosses are denoted by a number indicating
  number of doses of recurrent parent, followed by
  an asterisk or x.
• e.g. if Kalinga III is crossed to IR64 and then
  the F1 is backcrossed to IR 64, the cross is
  described as
• Kalinga III/2IR 64

9
Describing lines

• Each cross should receive a unique, consecutive
  number
• e.g. The IRRI cross IR55419-04/Way Rarem is
• IR74371
• (This number can be thought of as referring to
  the F1.)
• Individual plants selected from segregating
  generations (F2 onward for a single cross) are
  given a unique plant number. Lines derived from
  these plants are identified by the plant number
• e.g. The line derived from the 24th F2 plant
• selected from IR74371 is IR74371-24.

10
Describing lines

• If a generation is harvested in bulk, without
  single plant selection, the resulting population
  is denoted B.
• e.g. If the F2 of IR74371 is harvested in bulk,
  without selection, it is denoted IR74371-B
• IR74371-B-7 denotes the 7th F3 plant selected
  from a bulk F2
• Exercise
• Describe the selection history of
• line IR74371-B-5-6-B-B-43

11
Homozygosity of a line

• For a line in F generation n, the proportion of
  loci that are homozygous, relative to the number
  of heterozygous loci in the F1, is
• Exercise
• What proportion of loci that were heterozygous in
  the F1 are homozygous in the F5 ?

1 (0.5)n-1
?6.25
12
Homogeneity of a line

• Homogeneity means sameness or uniformity
• Lines derived from homozygous plants are
  homogeneous
• Lines derived from heterozygous plants are
  segregating and non-uniform
• The degree of genetic uniformity of a line is
  determined by the level of homozygosity of the
  plant from which it was derived
• e.g. A single plant selected from an F3 bulk has
  a homozygosity level of 1-(.5)2 0.75, relative
  to the F1. A line derived from this plant by
  selfing is fixed at 75 of the loci that were
  heterozygous in the F1.

13
Homogeneity of a line
Lines in the same inbreeding generations can have
very different levels of homogeneity!

• Example consider 2 lines in the F6. One derived
  from a single F3 plant, the other from an F5
  plant.
• In the F3 plant, 75 of loci are homozygous
• In the F5 plant, 93.75 of loci are homozygous
• ? An F3 derived F6 line is therefore more
  homogeneous than an F5 derived F6 line, although
  the plants in each are equally homozygous.
• For the example above
• The F3 derived line in the F6 is denoted as
  F3,6
• The F5 derived line in the F6 is denoted as F
  5,6

14
What is your opinion on Pedigree versus bulk
selection

• Advancing in bulk to F3 is almost always more
  efficient than pure pedigree selection
• Even using bulk method, mild selection for plant
  type, grain shape can be applied
• About 2000 plants should be carried per generation

15
How many generations should pedigree selection
continue?

• 2 factors to consider
• Phenotypic (visible) uniformity
• ?If the line is not visibly uniform, it needs
  reselection
• Genetic heterogeneity remaining within the line.
• ?Genetic heterogeneity (variability) is the
  fuel of selection
• ?Selecting among genetically homogeneous plants
  is pointless

16
Exercise Construct a table for lines established
from the F3 through the F8 generation, indicating
the proportion of segregating loci expected
within the line.
Generation Proportion of segregating loci
F3
F4
F5
F6
F7
F8
25
12.5
6.25
3.13
1.56
0.78
17
Controlling field variability in pedigree
nurseries

• Field variability can have a big impact on even
  highly heritable traits like height, DTF etc.
• Use repeated checks, but not too many
• Select among nearby lines (gridding)
• Treat groups of advanced sister lines as
  replicates of a family select on the basis of
  family means rather than individual line
  performance (or select the best lines from the
  best families)

18
Selection based on family means
Conventional arrangement
Range 1 IR70000-1-1-1 IR70000-1-1-2 IR70000-1-1-3
IR70000-2-14-1 IR70000-2-14-2 IR70000-2-14-3
Range 2 IR70002-3-1-1 IR70002-3-1-2 IR70002-3-1-3
IR70005-5-7-1 IR70005-5-7-2 IR70005-5-7-3
Range 3 IR70007-6-4-1 IR70007-6-4-2 IR70007-6-4-3
IR70010-3-8-1 IR70010-3-8-2 IR70010-3-8-3
19
Selection based on family means

• Blocking by families

Block 1 IR70000-1-1-1 IR70000-2-14-1 IR70002-3-2-
1 IR70005-5-7-1 IR70007-6-4-1 IR70010-3-8-1
Block 2 IR70000-1-1-2 IR70000-2-14-2 IR70002-3-2-
2 IR70005-5-7-2 IR70007-6-4-2 IR70010-3-8-2 (rando
mize)
Block 3 IR70000-1-1-3 IR70000-2-14-3 IR70002-3-2-
3 IR70005-5-7-3 IR70007-6-4-3 IR70010-3-8-3 (rando
mize)
20
Any questions or comments?
21
Summary 1

• Most successful crosses have at least 1
  high-quality, adapted parent that is preferred by
  farmers in the TPE
• BC1- or BC2-derived populations may be efficient
  for generating high-quality, high-yield lines
  because they leave adapted gene blocks in the
  elite recurrent parents intact
• At least 2000 plants should be screened in the F2
• A standard pedigree description system should be
  used
• Bulk inbreeding in the F2 and F3 generations is
  used by some breeders to inexpensively produce
  uniform lines

22
Summary 2

• Pedigree selection should focus only on highly
  heritable, easily-scored traits (unless family
  selection is used.)
• There is little genetic variability among F6
  sister lines derived from the same F5 plant
• Regularly occurring checks should be included in
  pedigree nurseries, but their frequency should
  not exceed 10 of the total number of plots
• Selection among families of closely-related lines
  can be used in advanced pedigree generations.
  Individual lines in a family can be treated as
  replicates, and planted in different blocks to
  overcome the effect of field variability