Researcher school genetics / Forskarskule genetikk

1
(No Transcript)
2
RESEARCH SCHOOL GENETICS / FORSKARSKULE GENETIKK

• Tormod Ådnøy, leader
• 5.9.2007

3
Welcome! Program today

• 1415 On the Research school genetics at UMB.
  Tormod Ådnøy
• 1438 Lille lørdag Åsmund Bjørnstad
• 1440 Group work Who we are, what we know, and
  what we want. Groups of 3-5 participants.
• 1455 Brief summing up by the youngest in every
  group.
• 1500 Pizza, beer, .. (free) (Husdyrkantina)
• Discussions over tables, and all together.
  (Genetic small talk permitted)

4
What we are

• An application for funding (web - in Norwegian) /
  100.000 nok
• But people and ideas are more important than
  money
• Email list 42 PhD, 45 Advisors and researchers
• Web page http//www.umb.no/22912
• Preliminary board (3 PhD, 2 advisors)
• Silje Brenna Hansen (PhD Cigene)
• Marianne Haraldsen (PhD IHA Forskargruppe
  genetikk og avl)
• Simen Rød Sandve (PhD IPM Genetikk og
  plantebiologi)
• Morten Lillemo (postdoc IPM)
• Tormod Ådnøy, leader research school (assoc.prof.
  IHA)
• A secretary Anne Golten, IHA
• This gathering today, and first Wednesday every
  month
• Focus on PhD students

5
What we will be remains to be seen

• Send an email for new participants to join the
  school
• So far membership is not exclusive
• Meetings are open
• Peer review groups?
• Reader groups?
• Nordic collaboration?
• Research grant applications?
• Include MSc students?
• ECTS for some activities?
• Presentation of own work for others in the
  Research school
• May help self-image
• Will give useful training
• Future courses in the Research school?
• Summer courses?
• ...

6
What I will talk about now

• SCHOOL
• Institution
• Knowledge
• Feeling ok
• Clowns
• Paradigms

• GENETICS
• Genes
• DNA, mRNA, ..
• SNP
• Genotypes, haplotypes
• Regulatory nets
• BREEDING
• Regression
• Additive relationship

7
GENETICS

• Gene
• Gene maps, DNA, mRNA, , amino acids and
  proteins,
• SNPs identifying genes in single individuals
• We know a lot more now than some years ago
• Molecular lab people have a lot of information
  and will have a lot more!How can it be used?
• Can it be used to find the best future individual
  for a trait we want to improve?
• What combination of genes is best?

8
How is the gene expressed in a trait?

• We dont see a genes value, we see an
  individuals complete genomes value!
• Genotype value for the trait
• How do we express a genes value
• or
• How do we know which genes to combine to have a
  better individual in the future?

9
If two individuals were the same except two
alleles in a locus

• We could say that the difference in the two
  individuals genotype values was the difference
  of the two allele effects
• But the alleles may interact with other genes, or
  the environment
• And normally we have a lot more differences
  between two individuals than just two different
  alleles

10
Numbers of genotypes

• ..very many
• How do we know which individuals/genotypes to
  select for future breeding?
• What is you answer?
• May we predict what value a not yet existing
  genotype (of infinitely many) will get for a
  trait?

11
.. to have better individuals in the future ..

• Select for single gene effects
• Select for haplotype effects
• Select for combination of gene effects
  (dominance, epistasis, heterosis)?
• Select for best genotypes today breeding

12
Casein genes in Norwegian goats

• DNA from 436 bucks in national breeding scheme
• Analyzed 39 snps (single nucleotide
  polymorphisms) in 4 casein genes (on same
  chromosome)
• Haplotypes deduced from snp genotypes and
  relationship
• Milk (kg), and protein-, fat-, lactose- from
  daughters in Goat dairy control
• Hayes,Ben Hagesæther,Nina Ådnøy,Tormod
  Pellerud,Grunde Berg,Paul R. Lien,Sigbjørn
  (2006) Haplotype structure of casein genes in
  Norwegian goats and effects on production traits.
  Genetics 174, 455-464.

13
Casein snp genotypes excerpt of the 436 bucks
14
Results haplotypes effects on fat-, protein-
and milk, not significant for lactose
15
Results of single snp not significant
16
Finding gene effects / Number of genotypes

• We deduced 21 haplotypes on bucks, and found
  additive effects on daughters production.
• All possible combinations of 39 snps is 339gt1018,
  but number of individuals observed was 436. All
  genotypes may not be modeled, only the ones
  observed.
• Modeling additive effects of all 39 snp-s
  simultaneously led to collinearity problems, but
  we could analyze for one snp at a time.
• Even to find all haplotype combinations
  represented in a sample will be difficult
  212120/2231 potential genotypes. (Some
  haplotypes are rare.) How important is haplotype
  dominance?

17
BREEDING

• To generate the best future individuals We want
  to change the population mean
• Info used
• Phenotypic observations
• Additive relationship
• ..best genotypes/ population (for a future
  environment)
• Given
• Existing populations,
• Existing knowledge about the populations,
• Existing techniques for breeding (AI, blup, ..)
• Focus is on population, less on individuals

18
How do we know which individuals/genotypes to
select for future breeding?

• Select best phenotypes Mass selection
• Select individuals with best offspring
• Other methods
• Breeders use genes as an alibi they dont need
  them!
• Statistics linear regression of offspring
  phenotypes on parents phenotypes
• Additive inheritance of genes is a motive for
  relationship matrix
• Include info on genes
• Meuwissen, T. H. E., Hayes, B. J., Goddard, M.
  E. (2001). Prediction of total genetic value
  using genome-wide dense marker maps. Genetics
  157, 1819-1829.

19
The infinitesimal model

• Many genes
• Small effects
• Independently distributed
• No change in gene frequency
• Equilibrium of gene frequencies
• All assumptions are violated in breeding
  programs, normally
• Shaky foundation of Fisherian genetics SWO
• Why does it work so well?

20
Additivity

• A crucial question
• To what extent are gene effects additive?
• How does deviations from additivity affect the
  Parent-Offspring relationship Cov (P,O)
  0.5Additive variance ?

21
Genotype values
22
Genetic variance pure additive model

• For two loci with two alleles each (H h K k), and
  only additive gene effects (in figure aH2 aK3,
  while ahak0)
• Let Hi1 when H-allele is present, Hi0 when
  h-allele
• Then the genotype value is
• y(i,j) H(i)H(j)aH K(i)K(j)aK
• (0, 2, 4, , 10 in figure)
• The mean genotype value is
• EY sum p(i,j) y(i,j) 2pHaH 2pKaK
• The variance of the genotype values, with random
  mating and same disequilibrium in parents
  gametes (D dHKpHK-pHpK)
• VY E(Y2)(EY)2 2pHphaH22pKpkaK24dHKaHa
  K
• VY0 VYd
• Avery, P. J. Hill, W. G. (1978). The effect of
  linkage disequilibrium on the genetic variance of
  a quantitative trait. Adv. Appl. Prob. 4-6. /
• Ådnøy, T. (1981). Selection in few-locus models /
  Seleksjon i få-lokus modellar. PhD-dissertation
  at Dept Mathem Statist, Agric Univ Norway. 1-218.
  
• Even in the additive model, disequilibrium over
  loci will change the variance.Linkage,
  selection, .. may lead to disequilibrium.

23

• If selection is for an additive trait we should
  expect that the best allele is fixed in every
  locus
• Should be no genetic variation left
• This does not happen normally
• There is genetic variation left for most traits
  even after much selection
• Why?

24
Bridging the gap genes phenotypes (Cigene in
eVita)

• Arne Gjuvsland (Cigene) PhD dissertation October
  2
• linking regulatory gene networks to additivity
  and dominance

25
SCHOOL

• To learn
• Something is not known by students
• It is normal that students dont know thats
  why they attend a school
• The most important is the process in the
  students heads
• Transfer of knowledge from lectures, books, ..
• It helps to know what you already know
• Generation of knowledge
• Important science may generate new schools
  (paradigms)
• The shaky foundations of Fisherian genetics
  SWO
• Creativeness is good in a research school / new
  ideas

26
To feel OK

• Emotions are important for learning
• We learn more when we fell ok
• We (most of us) need to know if what we are doing
  is good/relevant/useful/
• We may not always rely on our self-evaluation
• We need external evaluations
• Norwegians are good at belittling themselves
• We need to compare to what others do
• What do you need to trust that you are doing ok?
• If I tell you youre clever do you believe me?
• Others input may correct our learning make us
  better students
• Dont be afraid to tell what you dont know!
• Helps other feel helpful / builds their
  self-image
• Clowns help us relax
• May help us see ourselves in a new light

27
I have talked about

• Genetics
• Breeding
• School

28
Litle laurdag / Lille lørdag Wednesdaylittle
Saturday

• Now professor Åsmund Bjørnstad will tell a story?
  
• Fanfare!!
• In comes the clown??

29
(No Transcript)
30
Groups

• Divide in groups

31
GROUP ASSIGNMENT

• Who we are
• Present yourself to the group
• Name, birth, occupation,
• What we know
• What techniques do you use? What courses are you
  taking?
• Variance components, Linear models, Molecular
  lab, mRNA, micromatrices, HFA401,
• How does your discipline find the best
  individuals for the future?
• What we want
• How can a research school be useful?
• What can we contribute yourself and what can we
  get/buy from others?
• Present two topics where you think our school may
  be helpful to the whole group at 1455. (By
  youngest in group.)

32
(No Transcript)
33
Food and drink

• Pizza
• 8 assorted kinds
• 1 vegetarian
• 1 muslim
• Salad with vinaigrette
• One bottle of drink
• Apple drink
• Clausthaler Beer without alcohol
• Green Tuborg
• I need two voluntaries to help with the dishes
  afterwards

34
(No Transcript)