Fine-mapping of QTL using high-density SNP genotypes

J. B. Cole,1,* P. M. VanRaden,1 D. J. Null,1 T. S. Sonstegard,2 M.
McClure,2 C. P. Van Tassell,2 H. A. Adams,3 and H. A. Lewin4
1Animal Improvement Programs and 2Bovine Functional Genomics Laboratories,
Agricultural Research Service, USDA, Beltsville, MD, USA
3Department of Animal Sciences & Institute for Genomic Biology, University of
Illinois at Urbana-Champaign, Urbana, IL, USA
4Department of Evolution and Ecology, University of California, Davis, Davis, CA
95616, USA
john.cole@ars.usda.gov
Fine-mapping of QTL using
high-density SNP genotypes

16th MAS-QTL Workshop, Alghero, Italy, 24 May 2012 (2)
Illumina genotyping arrays
 BovineSNP50
 54,001 SNPs (version 1)
 54,609 SNPs (version 2)
 BovineHD
 777,962 SNPs
 BovineLD
 6,909 SNPs
 Allows for additional SNPs (e.g.,
GeneSeek Genomic Profiler)
BovineSNP50 v2
BovineLD
BovineHD

Bovine High-Density Bead Chip (HD)
• 778K SNP chosen to
• Be evenly spaced
• Include some Y-specific SNP
• Include mitochondrial SNP
• Utilize across-breed information
• Fine mapping of QTL
• Enhanced performance in Zebu cattle

Estimation of marker effects
 Predict traditional PTA using phenotypes and
pedigree
 Compute SNP effects using deregressed PTA
weighted by reliability
 Bayes A-type model
 Regress small effects to mean
 Allow large effects to grow

We got right to the point…
Cole, J.B. et al. 2009. Distribution and location of genetic effects for dairy traits. ICAR Tech Ser. 13:355–360.

…and found some interesting things
ARS-BFGL-NGS-109285
Cole, J.B., VanRaden, P.M., O'Connell, J.R., Van Tassell, C.P., Sonstegard, T.S., Schnabel, R.D., Taylor, J.F., and
Wiggans, G.R. 2009/ Distribution and location of genetic effects for dairy traits. J. Dairy Sci. 92(6):2931–2946.

Is luck better than skill?
 ARS-BFGL-NGS-109285 at 57,895,121 Mb on BTA18
 Signal consistent in HD data
 Intronic to a putative CD33-related Siglec gene
 This region is gene-rich
 Many Siglecs involved in leptin signaling
 Also affects gestation length
Maltecca, C., Gray, K. A., Weigel, K. A., Cassady, J. P., Ashwell, M. 2011. A genome-
wide association study of direct gestation length in US Holstein and Italian
Brown populations. Animal Genetics 42:1365-2052.

50k SNP are good for regions, not genes
Cole et al. 2009. J. Dairy Sci. 92(6):2931–2946.

Simple strategy
 Compute SNP effects for trait of interest
 Look for peaks
 Perform bioinformatics on regions under
interesting peaks
 NCBI/Ensembl
 Bovine Gene Atlas
 Bovine QTLdb

Can we look at every peak?

Are 777k SNP better than 50k?
VanRaden, P.M., Null, D.J., Sargolzaei, M., Wiggans, G.R., Tooker, M.E., Cole, J.B., Sonstegard, T.S., Connor, E.E.,
Winters, M., van Kaam, J.B.C.H.M., Van Doormaal, B.J., Faust, M.A., and Doak, G.A. Genomic imputation and
evaluation using high density Holstein genotypes. J. Dairy Sci. (Submitted).

Case studies
 Identification of causal variants associated
with two haplotypes related to fertility
 Discovery and validation of HH1 in U.S.
Holstein cattle
 Discovery and validation of JH1 in U.S.
Jersey cattle
 Fine-mapping of the Weaver locus

Recessive defect discovery
 Check for homozygous haplotypes
 7 to 90 expected but none observed
 5 of top 11 are potentially lethal
 3.1% to 3.7% lower conception rates
 Some slightly higher stillbirth rates
 Confirmed Brachyspina same way

Novel haplotypes affecting fertility
Name
Chrom-
osome
Loca-
tion
Carrier
Freq Earliest Known Ancestors
BTA Mbase %
HH1 5 62-68 4.5 Pawnee Farm Arlinda Chief
HH2 1 93-98 4.6 Willowholme Mark Anthony
HH3 8 92-97 4.7 Glendell Arlinda Chief,
Gray View Skyliner
JH1 15 11-16 23.4 Observer Chocolate Soldier
BH1 7 42-47 14.0 West Lawn Stretch Improver
VanRaden, P.M., Olson, K.M., Null, D.J., and Hutchison, J.L. 2011. Harmful
recessive effects on fertility detected by absence of homozygous haplotypes. J.
Dairy Sci. 94(12):6153–6161.

Economics of Next Generation Sequencing
Bovine genome = 2.85 billion bases
30X Coverage = about 90 billion bases
1 run on HiSeq2000 = 600 billion bases
20 animals sequenced to 30X coverage in 9 days (2 x 100 bp reads)
Cost about €19,600

HH1 in Holstein cattle
 75 SNPs spanning 7 Mbp on Bos taurus
chromosome 5
 Traced to Pawnee Farm Arlinda Chief
 Very popular bull
 >16,000 daughters
 >500,000 granddaughters
 >2 million recorded great-granddaughters

Sequencing and haplotype reconstruction
= Eight bulls in study derived from purchased
semen- ~30X seq. coverage
= Previously sequenced-6X on
454 Roche
*
* **
*= Four bulls used to identify
mutation causative for HH1

Locating HH1 causal variant
Reference
genome
C/T
Original75 SNP HH1 haplotype
Refined 38 SNP
haplotype

Sequence analysis of HH1 SNP

Annotation of mutated gene
 APAF1 - Bos taurus apoptotic peptidase activating factor 1
 ATP binding factor
 Gene expression for APAF1 in murine development begins
between 7 and 9 d in heart, mesenchyme, periderm, and primitive
intestine (Muller et al., 2005)
 Gene knockout of APAF1 in mice leads to embryonic lethality
(Muller et al., 2005)
 Proteins required for this
pathway/cascade are important
for neural tube closure in vivo

SNP validation in HH1 interval
 Designed Sequenome 24 assays for 12 SNP
in the HH1 interval
 Encompasses all SNP for coding, 3’ UTR,
and downstream regions – and 5 other SNP
covering all other genes in the interval
 Tested wide diversity of haplotypes for
HH1 region – use SNP50 DNA archive

Concordance of genotype to HH1 statusConcordancetoHH1haplotypes(12,32)
Genome Coordinates UMD3.1
Exonic SNP
1.2% false positives at intron SNP
One Mbp Interval

HH1 - Conclusions
 HH1 Haplotype 12 100% concordant with
heterozygosity at one exonic SNP location (N=486)
 HH1 Haplotype 32 100% concordant with het status
at same exonic SNP location (N=11)
 Other 256 non-carrier animals for HH1 interval were
homozygous normal at this exonic SNP
 One intronic SNP was 98.8% concordant to HH1
carrier status
 No other individual SNP were viable candidates
 Based on collaborative work with Harris Lewin

NGS sequencing of JH1 carrier animals
naab otherhap otherfreq ANIM_NAME byr
014JE00244 8 2.07073 MEADOWAY CHERRY GARCIA 1990
007HO06417 13 2.10855 AMES CHOCOLATE SURVILLE 1978
008JE00361 14 0.48222 GCG BRASS PRINCE 1985
008JE00266 38 0.27421 FAIRWAY MORGAN KREMLIN-ET 2004
008JE00230 45 4.51967 GATES SKY LINE CONGA-ET 1993
007JE00657 49 0.53896 PEARLMONT HALLMARK CATAMOUNT- 2002
014JE00411 75 0.01891 GABYS FAIR ROCK-ET 2000
011JE00762 80 0.12292 CLOVER FARMS TACO SUPREME 2001
007JE00254 91 0.69024 WILSONVIEW ARTISTIC ROMEO 2005
200JE07039 117 0.06619 BUSHLEA BROOK BIESTAR 1999
 Sequencing run included Chocolate Soldier (JH1 founder) and
Oman (for HH3)
 30X coverage of whole genome per animal
 Funded by American Jersey Cattle Association
 Found 38 candidate SNP in 492 kbp interval

SNP Validation in JH1 Interval
 Designed 30 Sequenom assays for 15 unique
SNP in the JH1 interval
 Only 1 SNP in a gene
− Stopgain mutation in CWC15
spliceosome-associated protein
− Not expressed in every bovine tissue
 Test all JH1 carriers in the SNP50 repository
 185 normal, 546 carriers

Spliceosome structure and CWC15
Will and Lührmann. 2011.
Spliceosome structure and
Function. Cold Spring
Harb Perspect Biol.

JH1 - SNP Validation Results
 JH1 Haplotype 99.3% concordant with
CWC15 stopgain mutation
 5 non-concordant samples
 False negative JH1 haplotypes (2 cases)
 DNA misplating? Retest samples
 2 other SNP in complete LD with JH1-
CWC15 stopgain
 Working with Jersey to complete an
independent validation

Annotation is a big problem
 Function of CWC15 in the bovine is unknown
 CWC15 ortholog in mice, mED1, expressed in
early embryos (Duan et al., 2010)
 No analagous studies in large mammals
 Funding
 Time
 Perceived impact

Preliminary fine-mapping of Weavers
 35,353 SNP on BTA4
 69 Brown Swiss bulls with HD genotypes
 20 cases and 49 controls
 No affected animals!
 Microsatellite-mapped to the interval 43.2–
51.2 cM
 Phenotype based on name

Preliminary sliding-window analysis
BTA4_43-60Mb
0
50
100
150
200
250
300
43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60
Mb
-log10p

HD analysis and NGS
 GWAS with Bovine HD
 20 carrier and 50 controls
 Collaboration with Italian consortium
 Refined historic interval from 46-56Mb to 48-53Mb on BTA4
 Weaver similar to ALS in humans
 18 annotated genes in interval
− 7 of them interact with major gene responsible for heritable ALS
 NGS performed on a pool of 10 Normal and 10 Carriers
resulting in ~30x coverage
 117 SNV in the Weaver locus,
− 1 synonymous, 3 non-synonymous
− Test all SNP

Validation of Weavers
 5 Sequenom multiplexes tested 114 SNV
 715 Brown Swiss, 26 Carora, 4 Angus, 4 Holstein, 4 Jersey, 3 Senepol, 3
Herefords
 Phenotype mapping refined locus to 35 SNP
 The test is accurate
 5 Carora carriers based upon the 35 SNP
 Multiple ‘assumed normal’ BS are carriers/affected
 A few BS diagnosed normals were carriers
 Found 1 Holstein heterozygous for the right 30 SNP and homozygous
normal for the left 5 SNP
− Related (6.25%) to BSUSA000000183023 (MEADOW VIEW MATT ALEX) who is a
confirmed Weaver carrier

Future application of NGS sequencing
Researchers Catalog Loss-of-Function Variants in
Human Protein-Coding Genes
NEW YORK (GenomeWeb News) – A Wellcome Trust Sanger Institute and
Yale University-led team has sifted through data from three 1000
Genomes Project pilot efforts to find a set of authentic loss-of-function
variants in the human genome.
"Each of us can be walking around with at least 20 genes basically
inactivated," the study's first author, Daniel MacArthur, told
GenomeWeb Daily News.
If this is true for humans, then is it true for cattle???

Conclusions
 Targeted resequencing is effective
 Genotyping-by-sequencing can identify loss-
of-function mutations
 Functional studies of protein function in vivo
are needed
 New precision mating strategies should be
developed

Acknowledgements
• AIPL, Beltsville, MD (USDA-ARS)
– George Wiggans and Tabatha Cooper
• BFGL, Beltsville, MD (USDA-ARS)
– Larry Shade
– George Liu, Derek Bickhart. Rueben Anderson, Sasho A.
– Steve Schroeder
– Alicia Beavers
• University of Illinois, Urbana-Champaign
– Denis Larkin
• Meat Animal Research Center, Clay Center, NE (USDA-ARS)
• Tim Smith, Tara McDaneld, John Keele

Fine-mapping of QTL using high-density SNP genotypes

More Related Content

What's hot (20)

Similar to Fine-mapping of QTL using high-density SNP genotypes (20)

More from John B. Cole, Ph.D. (20)

Recently uploaded (20)

Fine-mapping of QTL using high-density SNP genotypes