What is genomics ?

Donor / Embryo Genomics Patrick Blondin

L’Alliance Boviteq

AABP Embryo TransferSeminar

Montréal, 2012 Sept 19th

What is genomics? A consortium of Universities and North American artificial

insemination centers developed the Illumina BovineSNP50 Beadchip.

This Single Nucleotide Polymorphism (SNP) DNA chip contains more than 43 000 SNPs (this can be seen as 43 000 potential mutation between different individuals).

The pattern of SNPs obtained following SNP50 hybridization have been correlated to production, fertility and health traits.

Genomic values are then generated and used to rank animals in terms of genetic potential.

How to use GENOMICS to increase genetic gain ?

Option #1 (current situation):Mating of elite animals

Transfers

Calves genotyping to keep the best subjects

Option #2 (future…?):Mating of elite animals

Embryo genotyping to transfer only the most promising embryos

Getting DNA for genotyping

DNAExtraction

Bovine SNP50Hybridization

DNA

Extraction

Bovine SNP50

Hybridiza

tion

Impossible….10-12 cells in embryo

biopsies

10 000 times lowerthan requirementsfor Bovine SNP50

hybridization

Solution?

Pre-amplification of genomic DNA recovered from biopsies> 10 000 fold before SNP50 hybridization

GENTLE DNA« Liberation »

AMPLIFICATION STEP• Different methods:

• PCR based• Isothermal amplification

• Plenty of commercial kits:• Qiagen• Nugen• Sigma• New England Biolabs• GE healthcare

• Release DNA• Decompact DNA• Avoid DNA breakage• Avoid loosing material

Be careful with commercial kits

10X 100X 1000X 10 000X

• EFFICIENT (Enough DNA for SNP50 hybridization) • INNACURATE (Too much inconsistencies VS starting template)

Most of them were designed for bigger samples than embryo biopsies

An EFFICIENT and ACCURATE method is absolutely required

Filling the holes… Even with optimized conditions, some discrepancies

between the end results and the starting templace are found following the amplification.

IMPUTATION is then a indispensable tool to fill the holes in the genome of the amplified biopsies generated by the amplification.

Imputation is done through FImpute V2.0, a software developped in house and optimized for embryo biopsies. This software performs a combined family and population imputation and reconstruct the genome using the data generated from the parents on the 50K.

Efficiency of the method We developped our own procedure for DNA extraction and

amplification to get the most accurate coverage of the genome from embryo biopsies.

Phi-29 based isothermal amplification

So far:

681 out of 709 biopsies were successfully amplified (96% efficiency)

Call rate (% of SNPs that generated a signal on the SNP50 array): 91 ± 5%

583 samples were sent to USDA/CDN for genomic evaluation and 572 (98%) passed USDA/CDN quality controls and generated results

Variation intra embryo flushes Some crosses generated very variable Direct Genomic Values

(DGV) between different embryos (flushes #1 and #2) For others, the DGV obtained were very similar (flush #3). In some cases, very large divergences (1842 pts of DGV) were

found between two embryos of a same flush (flush #4).

1 2 3 4 5 6 7 8 9 10 11 12 1 2 3 4 5 6 7 8 9 10 11 12 1 2 3 4 5 6 7 8 9 10 1 2 3 4 5 6

Flush 1 Flush 2 Flush 3 Flush 410001500200025003000350040004500

DGV calculated from different embryos of a same flush incomparison to the Genomic Parent Average

Embryos

Dire

ct G

enom

ic Va

lue

(DGV

)

= Genomic Parent Average

Parentage possible Multiple sires can be used for one embryo production in good

donors since pedigree validation is part of the process and then, sire identification is executed. It is then possible to try many crosses in a shorter period of time while being able to know the pedigree of each embryo.

1 2 3 4 5 6 7 8 1 2 3 4 5 6 7 8 9 10 11 1 2 3 4 1 2 3Sire 1 Sire 2 Sire 1 Sire 2

Flush 1 Flush 2

0500

1000150020002500300035004000

Example of embryos produced by the use of many siresat fertilisation in a single flush

Dir

ect

Gen

omic

Val

ue (

DG

V)

Evolution of DGVs during pregnancy Parental average and genomic values may change over time. What is the effect of those changes on the genomic on the

embryos since the data are generated 9 months before birth…

Even if variations were observed between both groups, the overall pattern remained the same as the best embryos of the cohort were still the best embryos 9 months later.

1 2 3 4 5 6 7 8 9 10 11 12 13 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19

Flush 1 Flush 20

50010001500200025003000

Variation accross time (Pregnancy duration) of the DGV calculated from the embryonic biopsy

July 2011At biopsyApril 2012At Calving

EmbryosDire

ct G

enom

ic Va

lue

(DGV

)

DGV differences between embryo and corresponding calf:

The first calves born from embryos genotyped were genotyped to measure the accuracy of our amplication method (Figure 4).

A mean divergence of 106 ± 68 pts of DGV was calculated for the first 25 samples (4.3 ± 3.6% from the DGV).

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 250500

100015002000250030003500400045005000

DGV differences between embryos and their corresponding calf

Embryo DGVat calvingCalf DGV

Embryo-Calf Pairs

Dire

ct G

enom

ic Va

lue

(DGV

)

Importance of accurate genotype and imputation

42 503 SNPs analysed in the 25 Calf-Embryo pairs

Before imputation: 4909 ± 1190 missing SNPs (holes) 5286 ± 1439 missing and WRONGS SNPs

After imputation: Basically, all missing SNPs (holes in the genome) are filled! However, some errors persist: 266 ± 188 wrong calls.

Amplification accuracy and genotype errors

After amplification and hybridization, the % of called SNPs is an indicator of the quality of the overall genotype generated.

Indeed, the call rate decreases, the % of errors in the SNPs generated increases!

86% 88% 90% 92% 94% 96% 98%0.0%0.5%1.0%1.5%2.0%2.5%

Call Rates VS Bad calls before imputation

Call Rate% E

rror

in c

alle

d SN

Ps

Impact of amplification quality on the accuracy of imputation Errors generated in the amplification impacts the imputation:

When call rate decreases (amplification of lower quality), the # of divergent SNPs between amplified material and corresponding calf increases!

86% 88% 90% 92% 94% 96% 98%0

100200300400500600

Call Rates VS discrepancies between embryos its calf after imputation

Call Rate

# of

dive

rgen

t SNP

s

Genotype quality VS Genomic evaluation Amplification quality impacts:

Exactitude of the SNPs generated Accuracy of the imputation results

What are the consequences on the DGV?

Better the amplification is, more reliable is the DGV

0100200300

Call Rates and DGV diver-gences between embryo and

calf

% of called SNPs following hybridiza-tion

DGV

dive

rgen

ces

Conclusion A robust and accurate amplification procedure has been

developped to generate high quantities of DNA from embryo biopsies.

The amplified DNA can be used for hybridization on the BovineSNP50 beadchip to generate genomic bovine evaluation from embryos.

Even if some changes happen during pregnancy, the best embryos of a flush remain the best of their cohort over time and thus, decisions made at the embryonic level are still valid at the time of calving.

Very small divergences were observed between the genomic evaluations predicted from the embryos and the ones obtained from the resulting calves.

This confirmed the accuracy of the amplification and imputation methods developed by our group.

Warnings… Imputation « repairs » a lot of errors generated

during the amplification. However, its accuracy is impaired by poor quality

genotype. Therefore, biaised amplification impacts the final

genomic evaluation. Plenty of factors could impact the amplification:

Bad biopsy (too small, degraded material) DNA extraction Method of amplification Commercial kits (none of them performed well

enough to be used as it)

Opportunities… Genomic at the embryonic level is now possible! This technology can be combined to embryo

freezing so breeders can accelerate their genetic gain by mutiplying crosses in a short period of time and only transfer embryos with higher potential.

Using multiple sires per session with good donors multiply the chances of getting very high profiles embryos.

Finally, this amplification procedure could be use for any diagnostic test involving DNA directly at the embryonic level.