S4.3. Association Mapping, Breeder Ready markers and Genomic Selection

1,861 views
1,660 views

Published on

Presentacion de 11th Asian Maize Conference which took place in Beijing, China from November 7 – 11, 2011.

Published in: Education
0 Comments
2 Likes
Statistics
Notes
  • Be the first to comment

No Downloads
Views
Total views
1,861
On SlideShare
0
From Embeds
0
Number of Embeds
105
Actions
Shares
0
Downloads
254
Comments
0
Likes
2
Embeds 0
No embeds

No notes for slide

S4.3. Association Mapping, Breeder Ready markers and Genomic Selection

  1. 1. Association Mapping, Breeder Ready markers and Genomic SelectionRaman Babu, Jill Cairns, Gary Atlin, PH Zaidi, Pichet Grudloyma, George Mahuku, Sudha K Nair, Natalia Palacios, Pixley Kevin, Jose Crossa, BM Prasanna and all the Breeders of CIMMYT
  2. 2. Outline Association Mapping for Drought Tolerance – CIMMYT‟s experience ● Are there large effect genes for GY_stress? ● Should we bother about “rare alleles” that have large effects? Association Mapping for Disease Resistance Association Mapping (Candidate-gene based) for Carotenoids „Breeder-ready‟ markers for disease resistance and ProA Integrating Genomic Selection in the breeding Pipeline
  3. 3. LD and Population structurein DTMA-AM panel based on55K SNPs Average distance between two markers is 55kb and Average EM- R2 is 0.26 LD in DTMA panel is low and hence suitable for association mapping Population structure is ‘mild’ and LaPosta Seq association results were corrected DTP for structure (through PCA) and kinship by MLM CIM-CALI
  4. 4. DTMA-AM panel and 55K SNPs can identify largeeffect genes – 1. Grain Color Psy1 92 – Yellow lines (1) R² = 37% 186 – white lines (0) SNP with largest significant association with grain color located within one of the exons of Phytoene Synthase1 (psy1) on chr.6
  5. 5. DTMA-AM panel and 55K SNPs can identify largeeffect genes – 2. QPM 10 – QPM lines (1) 268 – Normal lines (0) Opaque2 Ask2 at 7.01 R² = 8% R² = 16% Besides opaque-2 and ask-2, several minor QTL regions influencing kernel modification and tryptophan content identified that overlap with previously reported regions…
  6. 6. Mapping Drought ToleranceStrategy GWASAM-panel ~ 300 inbreds – TCd with CML312Known DT sources La Posta Sequia C7; DTP C9, MBR etc.Phenotyping 10 locations – Stress & OptimalHeritabilities Kiboko-10-Late (0.64), M-10 –Tlalti (0.67), Thailand- 10 (0.49), M-Tlalti-09 (0.54), Zim-10 (0.22) Across locations: 0.35Phenotype used in Combined BLUPs of TC_GY under stress, corrected forGWAS anthesis dateGenotyping Genome-wide, high density markers – 55K SNPs and GBS markers (500K SNPs)Statistical Model General Linear Model (PCA correction) and Mixed Linear Model (PCA + Kinship – Q+K)
  7. 7. 12 -15 significant genomic regions identified for DT 7.0% 5.8% 5.7% 7.3% 5.7% 6.2% 5.5% 5.1% 5.1% 4.9% Only 147 SNPs (~15 genomic regions) had R2 values more than 5%
  8. 8. Significant Genomic Regions associated with TC_GY_Stress Average GY of the stress trials – 1.3 t/ha Heritability across locations – 0.32 Effect SNP Chr Position P value MAF R2 (%) (kg/ha) Candidate GeneSYN39332 10 142655119 9.62E-06 0.49 7.6 29.3 Starch SynthasePZE-107042377 7 72216348 1.49E-05 0.32 7.3 35.6 Myb family transcription factor-related proteinPZE-108046876 8 77237318 2.33E-05 0.35 6.9 -34.4PZE-110029252 10 50842298 7.77E-05 0.40 6.8 -26.3PZE-107032355 7 45011599 6.62E-05 0.38 6.2 38.6SYN37988 2 146399448 3.84E-05 0.26 6.1 49.5 TSA: Zea mays contig27975, mRNA sequencePZE-101090321 1 80757998 1.67E-04 0.46 4.9 -31.4PZE-109041733 9 62608362 1.35E-04 0.42 4.9 25.5PZE-104047052 4 78536398 1.21E-04 0.32 4.7 30.1
  9. 9. Rare Alleles with Large Effects Average for Average for Average for Marker Chr Position P Minor Allele MAF DD Dd dd Effect (kg/ha) DD Dd ddPZE-104042524 4 67259441 3.70E-03 A 0.14 1499.5 1414.1 1382.8 116.6 7 59 188PZE-101066401 1 49827350 1.54E-02 A 0.04 1487.8 1391.5 96.3 10 0 257 SYN36769 4 4914023 7.83E-03 A 0.06 1479.7 1355.9 1391.7 88.0 14 2 249 SYN26515 1 63053588 2.42E-03 A 0.06 1472.1 1253.0 1391.0 81.0 15 1 251 SYN1035 5 5786027 3.24E-02 G 0.07 1463.1 1395.2 1390.3 72.8 16 2 240PZE-110053356 10 100124247 4.70E-03 A 0.11 1331.7 1381.2 1401.8 -70.1 17 24 224PZE-104113536 4 194565443 7.31E-04 A 0.13 1334.7 1282.7 1405.3 -70.6 33 3 231PZE-102096857 2 107898705 3.07E-03 G 0.08 1329.7 1400.4 -70.6 20 0 238PZE-109074314 9 116545321 2.21E-04 G 0.08 1328.8 1400.3 -71.5 20 0 246PZE-105127701 5 183968110 2.24E-04 A 0.07 1323.4 1400.7 -77.4 19 0 249PZE-102121069 2 162773047 8.92E-04 G 0.06 1320.1 1400.2 -80.1 17 0 250PZE-106064720 6 116886483 1.09E-02 A 0.07 1318.2 1307.6 1401.3 -83.1 17 1 248 SYN14434 2 15813081 1.40E-03 A 0.08 1314.6 1433.4 1398.6 -84.0 19 1 221PZE-106056703 6 107499158 1.98E-04 G 0.06 1310.2 1307.6 1401.2 -91.0 15 1 251 SYN8914 3 194356323 4.25E-03 G 0.08 1307.9 1381.6 1400.1 -92.2 9 25 226
  10. 10. PZE-101066401 1Rare Alleles with 49827350 SYN36769 4 GY 4914023Positive Large A POB.502 c3 F2 10-3-2-1-BBBBBB-B (kg/ha) 1429.0 A GY (kg/ha) [SYN-USAB2/SYN-ELIB2]-12-1-1-2-Effects POB.502c3 F2 9-14-1-2-B-B-B-B CLQ-RCYQ28=(CLQ6502*CLQ6601)- 1482.4 BBB 1497.3 [CML440/[[[K64R/G16SR]-39-1/[K64R/G16SR]- 20-2]-5-1-2-B*4/CML390]-B-39-2-B-4-#-1- B-34-2-2-B*6-B 1476.1 B//ZM303c1-243-3-B-1-1-B]-9-1PZE-104042524 DTPWC9-F24-4-3-1-B-B-B 1554.0 [[KILIMA ST94A]-30/MSV-03-1-10-B- DTPWC9-F115-1-4-1-1-B-B-B 1483.4 1-B-B-1xP84c1 F27-4-1-6-B-5-B] F8-3-4 2-2-1 x G16SeqC1F47-2-1-2-1-BBBB-67259441 DTPWC9-F103-2-1-1-1-B-B-B 1469.6 B-xP84c1 F26-2-2-6-B-3-B]-3-1- DTPYC9-F46-3-4-1-1-B-B-B 1535.9 B/CML395]-1-1 1419.5 GY [Pob.SEW-HG"B"c0F39-1-1-1-1xMBR DTPYC9-F46-3-9-1-1-B-B 1461.7 C5 Bc F22-2-1-4-B-B-B-B-2-2-B-B-A (kg/ha) DTPYC9-F46-1-2-1-2-B-B 1606.1 B/CML442]-1-1 1333.290[SPMATC4/P500(SELY)]#-B-4-2-B-B 1483.8 DTPYC9-F13-2-1-1-2-B-B 1379.5 [Cuba/Guad C3 F34-2-1-1-B-B-B x CML264Q]-1-1 1376.4DTPYC9-F46-3-9-1-1-B-B 1461.7 CML-322 1428.5La Posta Seq C7-F125-2-1-1-2-B-B-B 1436.8 SYN26515 DTPWC9-F115-1-4-1-1-B-B-B 1483.4 1 DTPWC9-F31-1-3-1-1-B-B-B 1492.0La Posta Seq C7-F103-2-2-2-1-B-B-B 1626.9 DTPWC9-F67-1-2-1-2-B-B-B 1506.5 63053588La Posta Seq C7-F180-3-1-1-1-B-B-B 1593.5 GY DTPWC9-F104-5-4-1-1-B-B-B 1454.3 DTPYC9-F46-3-4-1-1-B-B-B 1535.9La Posta Seq C7-F96-1-1-1-B-B 1482.1 A (kg/ha) DTPYC9-F46-3-9-1-1-B-B 1461.7DTPYC9-F72-1-2-1-1-B-B 1411.4 CML444-B 1501.9 DTPYC9-F46-1-2-1-1-B-B 1552.7 S87P69Q(SIYF) 109-1-1-4-B 1518.4 DTPYC9-F46-1-2-1-2-B-B 1606.1 DTPWC9-F67-2-2-1-B-B-B 1568.7 CLQ-RCYQ40 = (CML165 x CLQ-6203)-B- 9-1-1-B*8 1509.3 CML497=[CL-00331*v]-3-B-3-2-1-B*6 1443.1 DTPWC9-F115-1-4-1-1-B-B-B 1483.4 DTPWC9-F109-2-6-1-1-B-B-B 1467.8 DTPWC9-F67-1-2-1-2-B-B-B 1506.5 DTPWC9-F104-5-4-1-1-B-B-B 1454.3 DTPWC9-F128-1-1-1-1-B-B-B 1390.9 DTPYC9-F143-5-4-1-2-B-B-B 1442.1 DTPYC9-F143-1-6-1-B-B 1414.6 DTPWC9-F67-2-2-1-B-B-B 1568.7
  11. 11. PZE-106056703 SYN14434Rare Alleles with 6 107499158 2 15813081Negative Large G [CML444/CML395//DTPWC8F31-4-2-1- 6]-2-1-1-1-B*4 1331.949 A P501SRc0-F2-47-3-2-1-B-B [CML444/CML395//DTPWC8F31-1-1-2-2- 1268.038Effects [(CML395/CML444)-B-4-1-3-1- B/CML395//DTPWC8F31-1-1-2-2]-5-1- BB]-4-2-2-2-2-BB-B [CML444/CML395//DTPWC8F31-1-1-2-2- 1267.39 2-2-BB 1346.993 BB]-4-2-2-2-1-BB-B 1408.142 CML 384xMBR/MDR C3 Bc F58-2-1-3- 02SADVL2B-#-17-1-1-B 1419.196SYN8914 B-B-B-B-3-1-B-B-BB-B 1344.688 [CML440/[[[K64R/G16SR]-39-1/[K64R/G16SR]-20-2]-3 MBR C6 Bc F280-2-B-#-1-1-B-B-B-B-B- 5-1-2-B*4/CML390]-B-39-2-B-4-#-1-B//ZM303c1-243- B 1256.056 3-B-1-1-B]-9-1194356323 [G16SeqC1F47-2-1-2-1-BBBB-B-xP84c1 [CML144/[CML144/CML395]F2-5sx]-1-3-1-G F27-4-1-6-B-5-B] F23-2-1-2-3 x P43C9- 3-B*4 1397.445[CML198/ZSR923S4BULK-2-2-X-X-X-X-1- 1-1-1-1-1-BBBB-1-xP84c1 F26-2-2-6-B- [CML198/ZSR923S4BULK-2-2-X-X-X-X-1-BB]-3-3-1-1-2-B*7 1196.562 3-B]-2-1-B/CML395]-1-1 1258.137 BB]-3-3-1-1-2-B*7 1196.562S99TLWQ-B-8-1-B*5 1245.322 [M37W/ZM607#bF37sr-2-3sr-6-2-X]-8- [CML144/[CML144/CML395]F2-8sx]-1-1-1- 2-X-1-BB-B-xP84c1 F27-4-3-3-B-1-B] B*5 1171.7594001 1292.372 F29-1-2-1-6 x [KILIMA ST94A]-30/MSV- [CML144/[CML144/CML395]F2-8sx]-1-2-3-CLA41 1389.549 03-2-10-B-1-B-B-xP84c1 F27-4-1-6-B-5- 2-B*5 1203.073(A.I.Z.T.V.C. 20-3-1-1-2-B-B x B]3-1-2-B/CML442]-1-1 1190.413 CLA222 1337.217A.I.Z.T.V.C.PR93A-17-1-3-1-1-B-B)-B- [Pob.SEW-HG"B"c0F39-1-1-1-1xMBR [M37W/ZM607#bF37sr-2-3sr-6-2-X]-8-2-X- C5 Bc F22-2-1-4-B-B-B-B-2-2-B-B- 1-BB-B-xP84c1 F27-4-3-3-B-1-B] F29-1-2-1-14TL-1-3-B-B 1252.957 B/CML442]-1-1 1333.209 6 x [KILIMA ST94A]-30/MSV-03-2-10-B-1-B-[G16SeqC1F47-2-1-2-1-BBBB-B-xP84c1 [MBR Et/MBR Bc C1 F4-1-1-3-B-B-B- B-xP84c1 F27-4-1-6-B-5-B]3-1-2-F27-4-1-6-B-5-B] F23-1-3-1-1 x [KILIMA Bx1760B B1 Bco x Comp.-B-1-1-1-1-B- B/CML442]-1-1 1190.413ST94A]-30/MSV-03-2-10-B-1-B-B- B-B/CML395]-1-1 1354.8 [Cuba/Guad C3 F34-2-1-1-B-B-B xxP84c1 F27-4-1-6-B-5-B]-2-1- [CML 329/MBR C3 Am F103-1-1-2-B-B CML264Q]-1-1 1376.38 x CML486]-1-1 1346.293 CA00344 / PAC777F2-6-1-1-BB-B-B-BB 1321.875B/CML395]-1-1 1270.448 [(87036/87923)-X-800-3-1-X-1-B-B-1-1- P44 C10MH8-30-4-B-4-1-B-B-B-B- 1329.436POB.501c3 F2 13-8-2-1-BBBB 1383.065 1-B-B-xP84c1 F26-2-2-4-B-2-B] F47-3- P147-#136-5-1-B-1-BBB 1356.154CL-RCY031=(CL-02410*CML-287)-B-9-1- 1-1-3 x M37W/ZM607#bF37sr-2-3sr-6- CLQ-6211=P62QC6HC13-1-3-BBB-6-B-7-6-1-2-B*7 1433.411 2-X]-8-2-X-1-BB-B-xP84c1 F27-4-3-3-B- BBBB-7-9-B-B-B-B 1311.726 1-B]-3-2-B x P33c3 F64-1-1-4-BB]-1-1 1295.392 CML269=P25STEC1F13-6-1-1-#-BBB-f-##- P390amC3/285x287 F73-3-2- B*6-B 1407.819 3xMIRTC5Am F96-1-1-1-3-1)-1-1-B 1399.776 CL-02143 P21C6S1MH247-5-B-1-1-2-BBB- CL-G1837=G18SeqC2-F141-2-2-1-1-1- 1-##-B*10 1471.196 2-##-2-B*4 1275.469 CML421=P31DMR#1-55-2-3-2-1-B*18-B 1252.385 CML421=P31DMR#1-55-2-3-2-1-B*18- DTPWC9-F66-2-1-1-2-B-B-B 1291.755 B 1252.385 DTPWC9-F73-2-1-1-1-B-B-B 1329.332
  12. 12. Rare Alleles – Candidate genes Candidate genes Putative functionidentified by Rare Allelesupstream of a DNA biding/membranebound receptor Many membrane bound receptors like Rpk1, shown to confer DT in AT. Less documented helicase domain proteins in AT proved for DT in CKDEAD box Helicase domain dependent pathways cross-talk between ethylene signalling and drought response pathways well-related to ethyline insensitive2 documented glyco poteins rich in hydroxy proline was first studied in TracheophytesExtensin like cell wall protein which can with stand severe stressAnnexin IV domain Role of Annexins in DT well-documented in ATPeroxidase protein known for involvement in DT in rice, AT etc.Major Facilitator Superfamily (MFS)Transporters plays key roles in different stress conditions over expression of Aspartate aminotransferase along with otherAminotransferase genes has been patented for DTCREB domain containing TF Known component in stress related pathwaysUbiquitin subgroup known component in drought tolerance pathways
  13. 13. Traits for which AM analysis accomplished inDTMA-AM panel GY_Stress_BLUPs MSV GLS NDVI Senescence SPAD Canopy senescence ASI Root traits (Shovelomics!) Anaerobic Emergence % reduction in shoot weight under waterlogged conditions % reduction in root weight under water logged conditions
  14. 14. Following up the AM results● BC-NILs for validation of important genomic regions● Identify MARS progenies with contrasting genotypes and check the drought phenotypes● Genotype the DH lines from DT x Normal crosses and check the phenotypes● Introgress validated genomic regions into tester lines through MAS
  15. 15. Artesian – Recent Drought TolerantHybrid from Syngenta Base Hybrid Artesian Hybrid
  16. 16. Artesian – how was it developed?Strategy Association mapping (candidate gene-based) BC-MAS of 4-8 QTLsDT source germplasm CML333, CML322, Cateto SP VII (Brazil), Confite Morocho AYA 38 (Peru), or Tuxpeno VEN 692 (Venezula)AM-panel 575 inbreds – 47 different testers (mostly S-2 and S-3 TCHs)Phenotyping 4 locations (Colorado, California and Chile) – Optimal & stress - Yield reduction under stress was 40-60% from optimalGenotyping 85 polymorphisms (corresponding to 57 candidate genes) and 149 random polymorphisms across 600 lines – in total only ~250 markersEffect sizes of identified 60 to 650 kg/hagenomic regionsMinimum P value of any 0.0001significant region
  17. 17. Significant Conclusions – DT mapping LD in DTMA-AM panel is low and hence conducive for association mapping 55K genotype data is capable of identifying large effect genes „Reasonably large effect‟ genomic regions (10-15) do exist for GY_Stress and co-locate with genes, previously implicated for DT in At, rice and maize 9 genomic regions that had robust p-values together explained 35% of phenotypic variance for GY_Stress_Combined Lines with multiple donor segments identified for validation and introgression
  18. 18. Two Key genes in carotenoid biosynthetic pathway identified Association Mapping based on candidate gene sequences Lycopene epsilon cyclase (Harjes et al. 2008; Science) Hydroxylase (CrtRB-1/HydB-1) (Yan et al. 2010; Nature Genetics)
  19. 19. Breeder-ready markers developed and routinely being used in the H+ breeding program of CIMMYT for CrtRB1 and LcyEAM leads to identification of HighKey genes and polymorphisms ProvitA + + = maize! MAS for MAS for Deep orange LycE HydBPolymorphisms validated in earsdiverse tropical geneticbackgrounds and breeder-readyhigh throughput markersdevelopedRoutine use of markers andselection of favorablegenotypes in H+ breedingprogram
  20. 20. Allele Mining for CrtRB1 (HydB1) across variousAssociation Mapping PanelsPanel Genotypes with Fav. White(W)/Yellow(Y) allele/TotalCIMMYT_Syngenta 24*/501 – 16 new sources All Yellow (Y)CAM PanelIMAS 16/430 (6 from ARC, SA and 14-W and 2-Y 3 from KARI)Subtropical Collections 71/1131 – many new sources 24-W and 47-YADP lines of 19/122 – “1” and 23/122 – “H”SYNGENTAPS: * out of 24, 8 were previously fixed for fav. allele of CrtRB1 in the H+ breedingprogram through MAS
  21. 21. Association Mapping for Disease ResistanceMSV – Harare 2010 data (Heritability = 0.79) GLS-combined analysis (Heritability = 0.6)
  22. 22. MSV – Harare 2010 data (Heritability = 0.79) Significant chromosomal regions (P < 1.0E-05) associated with MSV resistance (Har-2010 data) based on DTMA-AM panel and 55K genotype data (MLM) Trait Trait FDR (False Minor average average Corr/Trend Corr/Trend discovery R2 Minor Allele Major for Minor for MajorMarker Chr Position P value Chi-square rate) (%) Allele Freq. Allele allele allelePZE-101093951 1 86065123 4.50E-08 29.92 0.002 11.5 A 0.34 G 1.83 3.08PZE-101098418 1 92204598 6.47E-07 24.77 0.011 9.5 G 0.36 A 2.15 2.95SYN36281 1 187128850 1.93E-06 22.67 0.019 8.7 G 0.11 A 2.21 2.72PZE-101094082 1 86384320 2.45E-06 22.21 0.020 8.5 G 0.39 A 1.99 3.10PZE-104024779 4 28770811 4.04E-06 21.24 0.022 8.2 A 0.15 G 2.26 2.73PZE-101098295 1 91837910 5.31E-06 20.72 0.022 8.0 A 0.33 G 2.12 2.92PZE-108038832 8 59948253 5.57E-06 20.63 0.021 7.9 A 0.47 G 2.63 2.70PZE-103070254 3 111066077 6.36E-06 20.38 0.022 7.8 G 0.24 A 3.07 2.52PZE-101094056 1 86365447 6.37E-06 20.37 0.021 7.8 G 0.50 A 2.16 3.16PZE-108039819 8 62905375 7.00E-06 20.19 0.022 7.8 G 0.46 A 2.62 2.69PZE-101090488 1 80905706 7.02E-06 20.19 0.020 7.8 A 0.29 G 1.83 3.00PZE-104016598 4 16339600 7.13E-06 20.16 0.019 7.8 A 0.33 C 2.21 2.87PZE-102080891 2 64845534 7.21E-06 20.14 0.019 7.7 A 0.28 C 2.19 2.84PZE-101098960 1 93244458 7.76E-06 20.00 0.019 7.7 A 0.40 G 3.11 2.36
  23. 23. Validation of AM regions and Breeder-ready markers for MSV PZE01132220936 PHM14104_23PZE0175698629 PZA00529_4 PZA02090_1 PZA03527_1 PZA02614_2 PZA03651_1 Candidate SNPs for MSV Chr.1 Chr.3 Chr.4 Chr.8 Msv1 R R R PZE0186365075 csu1138_4 PZA00944_1 S S S PZE0195148805 PZE01101110579 PZE01111422982 R PZE0175698629 R S S PZE-101093951 S R S R S S R S S R S
  24. 24. Genomic Selection
  25. 25. Genomic SelectionUsing 55K SNP data across 300 individuals in the AMpanels Trait RR-BLUP B-LASSO RP Grain Color (Binary) 0.8 0.82 0.87 QPM (Binary) 0.96 0.96 0.95 ProA - Quant 0.39 0.42 0.6 GLS - Quant 0.52 0.53 0.55 MSV - Quant 0.62 0.61 0.60 GY - Quant 0.34 0.35 0.36
  26. 26. Integrating GS in breeding pipeline (DH + off-season nusery + GS)Season ActivitySummer • Grow 50-100 F2s/BC1s • Select 50 plants/cross and cross to haploid inducerWinter • Chromosome doubling of putative haploids to get DHs • Seed chip (one kernel/DH) 2500 – 5000 DH kernels • Discard disease susceptible DHs through specific marker screening • Select DHs through GY-GEBVs and seed Increase (top 5- 10%)Summer • Test cross GEBV-selected DH lines to one/two testers • Yield trials of DH-TCHs
  27. 27. THANKS
  28. 28. % phenotypic variance explained by structurealone…in DTMA-AM panel % phenotypic variance Trait/Location explained by 10PCs GY_Stress_Combined_BLUP 15.8 MSV (Harrae2010+09-1) 38.2 GLS_Combined 25.1 GLS_Har_10 8.8 GLS_Kakamega 11.5 GLS_Columbia_Scatalina 30.2 GLS_San Pedro_Mexico 29.6 GLS_Acatec_Mex 23 GLS_Paraguacito_Columbia 6.7

×