Supplemental data

Supplemental data

Supplemental data Susceptibility to amoxicillin-clavulanate-induced liver injury is influenced by multiple HLA class I and class II alleles M. Isabel...

1MB Sizes 0 Downloads 3 Views

Recommend Documents

affinity analysis
Kinetic/affinity analysis of FK binding to immobilized CD46 was performed using. 4 three models in combination with the

SAT® Student Data Questionnaire Supplemental Instructions for
This document is to assist you in completing the SAT Student Data Questionnaire. Completing the Student Data. Questionna

RNY Supplemental Data - RNY Property Trust
Mar 4, 2015 - Green Realty Corp., at which time an affiliate of RXR Realty LLC (“RXR”) acquired. Reckson's interest

Incorporating Supplemental Nutrition Assistance Program Data in
Incorporating Supplemental Nutrition. Assistance Program Data in Earned. Income Tax Credit Administration: A Florida Cas

Supplemental Data Ancient Voyaging and Polynesian Origins
Indonesia/East Malaysia. Philippines. Trobriand Islands. North coast of New Guinea. Micronesia. Bismarck Archipelago. We

Supplemental Data Neural Predictors of Purchases
13.5. 15.4. Portrait of a University Book. 9. 0. 13.5. 3.8. Reno 911 DVD Set: Season 1 ... on representative slices from

Supplemental Data MicroRNA Signature Predicts Survival and
Sung-Liang Yu, Hsuan-Yu Chen, Gee-Chen Chang, Chih-Yi Chen, ... Singh, Chiou-Ling Cheng, Chong-Jen Yu, Yung-Chie Lee, Ha

Supplemental Operating and Financial Data Roseland - Mack-Cali
(operating). Marbella. (operating). URL® Harborside – I. (in-construction). San Remo. (future development). Marbella

Supplemental Data A Two-Step Process for Cytokine Production
Page 1 of 2. Supplemental Data. A Two-Step Process for Cytokine Production ... were infected with H. polygyrus and perip

Recent data show continued growth in Supplemental Nutrition
Jan 3, 2013 - programs have played a key role in supporting vulnerable families. Perhaps one of the most efficient and e

Supplemental data Susceptibility to amoxicillin-clavulanate-induced liver injury is influenced by multiple HLA class I and class II alleles

M. Isabel Lucena, Mariam Molokhia, Yufeng Shen, Thomas J. Urban, Guruprasad P. Aithal, Raúl J. Andrade, Christopher P. Day, Francisco Ruiz-Cabello, Peter T. Donaldson, Camilla Stephens, Munir Pirmohamed, Manuel Romero-Gomez, Jose Maria Navarro, Robert J. Fontana, Michael Miller, Max Groome, Emmanuelle Bondon-Guitton, Anita Conforti, Bruno H. C. Stricker, Alfonso Carvajal, Luisa Ibanez, Qun-Ying Yue, Michel Eichelbaum, Aris Floratos, Itsik Pe'er, Mark J. Daly, David B. Goldstein, John F. Dillon, Matthew R. Nelson, Paul B. Watkins, Ann K. Daly for the Spanish DILI Registry, EUDRAGENE, DILIN, DILIGEN and International SAEC

1

Appendix. Collaborators and Contributors to case recruitment Members of the SAEC scientific management committee Arthur L Holden [SAEC], Brian Spear [Abbott], Anahita Bhathena, PhD [Abbott],Joe Walker [DaiichiSankyo], Lon Cardon [GlaxoSmithKline], Matt Nelson [GlaxoSmithKline], Nadine Cohen [Johnson & Johnson], Joanne Meyer [Novartis], Steve Lewitzky [Novartis], Sally John [Pfizer], Klaus Lindpaintner [Roche], Steven Kovacs [Sanofi-Aventis], Meng Zhang [Sanofi-Aventis], Leonardo Sahelijo [Takeda], Michael Dunn [Wellcome Trust], Maha Karnoub [Wyeth], and Michael E Burczynski [Wyeth] DILIGEN collaborators Investigators: A.K. Daly (PI), C.P. Day, P.T. Donaldson, C. Donaldson (Newcastle University); G.P. Aithal (Nottingham Digestive Diseases Centre); M. Pirmohamed, B.K. Park, S. Khoo, I. Gilmore (University of Liverpool); W. Bernal (Kings College Hospital, London). Research nurses: J. Henderson (Newcastle University); C Davies (Nottingham Digestive Diseases Centre); K. Hawkins, A. Hanson, J. Evely (University of Liverpool). Other contributors to case recruitment: H. Hussaini (Truro), P. Mills (Glasgow), W. Griffiths (Addenbrooks Hospital, Cambridge), J. Collier (John Radcliffe Infirmary, Oxford), A. Brind (North Staffordshire), N. Fisher (Dudley), J. Shearman (South Warwick), E. Elias (Birmingham), A. Grant (Leicester Royal Infirmary), A. Austin (Derby), F. Gordon (Bristol), M.Cramp (Plymouth), S. Saksena (North Durham), H J McMurtry (Chorley). DILIN collaborators Paul Watkins (University of North Carolina- Chapel Hill), Robert Fontana (University of Michigan), Naga Chalasani (Indiana University), Herb Bonkovsky (University of Connecticut), Timothy Davern (University of California- San Francisco), James Rochon (Duke Clinical Research Institute), Jay Hoofnagle, Jose Serrano (Senior Project officers, National Institutes of Health).

2

EUDRAGENE collaborators Investigators: Mariam Molokhia (Joint PI), LSHTM, UK, Paul McKeigue (Joint PI), University of Edinburgh, Scotland, Data Analysis Committee: Bruno Stricker, Erasmus MC, NL, Qun-Ying Yue, Medical Products Agency, Uppsala, Sweden. Centre investigators: Alfonso Carvajal, Universidad de Valladolid, Spain, Luisa Ibáñez, Fundació Institut Català de Farmacologia; Hospital Universitari Vall d’ Hebron, Universitat Autònoma, Barcelona, Spain, Maryse Lapeyre-Mestre, Université de Toulouse, France, Jean-Louis Montastruc, Université de Toulouse, France, Dr. Anita Conforti, University Hospital, Verona, Italy, Giampaolo Velo, University Hospital, Verona, Italy, Michel Eichelbaum, Dr. Margarete Fischer-Bosch Institute Clinical Pharmacology, Stuttgart. Contributors to case recruitment/other: Emmanuelle Bondon-Guitton, Université de Toulouse, France, Inés Salado, Universidad de Valladolid, Spain, Lourdes Vendrell, Fundació Institut Català de Farmacologia, Barcelona, Spain, Francesca Succurro, University Hospital, Verona, Italy, Marco Smerghetto, University Hospital, Verona, Italy, Mark Eijgelsheim, Erasmus MC, NL, Ramazan Buyukcelik, Erasmus MC, NL, Pascal Arp, Erasmus MC, NL, Richard Jackson, LSHTM, UK, Ferran Orsola, LSHTM, UK. Pharmacovigilance centres involved in case ascertainment (France, Spain, Italy): L'Association Française des Centres Régionaux de Pharmacovigilance (CRPV), Sistema Español de Farmacovigilancia. Agencia Española de Medicamentos y Productos Sanitarios, Italian Pharmacovigilance Centres of Veneto Region. Spanish DILI Registry collaborators Investigators: R.J. Andrade (PI), M.I. Lucena, C. Stephens, E. Ulzurrun, Y. Borraz (Málaga University), F. Ruiz-Cabello, M.A. Lopez-Nevot (H. Virgen de las Nieves, Granada). Contributors to case recruitment: M. Robles-Diaz, M. García-Cortés, I. Moreno (H. Virgen de la Victoria, Málaga), M. Romero-Gomez, A. Madrazo, R. Corpas, E. Suárez (H. Valme, Sevilla), M.C. Fernández, G. Peláez, M. Casado, J.L. Vega, F. Suárez, M. González-Sánchez (H. Torrecárdenas, Almería), J.M. Navarro, (H. Costa del Sol, Málaga), C. Guarner, G. Soriano, E.M. Roman (H. San Pau, Barcelona), R. Planas, J. 3

Costa, A. Barriocanal, F.García-Góngora (H. German Trias i Pujol, Badalona), T. Muñoz-Yague, J.A. Solís-Herruzo (H. 12 de Octubre, Madrid), R. Pérez-Álvarez (H. Central de Asturias, Oviedo), A. Castiella, E.M. Zapata (H. de Mendaro, Guipuzcuo), F. Pons (H. Marqués de Valdecilla, Santander), M. Jiménez, R. González-Grande (H. Carlos Haya, Málaga), M. Moreno (H. la Laguna, Tenerife), J.L. Calleja, J. de la Revilla (H. Puerta de Hierro, Madrid), J. Salmeron, A. Gila (H. Universitario San Cecilio), R. Solá (H. del Mar, Barcelona), O. Lo Lacono (H. del Tajo, Madrid), E. Martín (H. La Línea, Cadiz), J.A. Durán, I. Carmona, A. Melcón de Dios, M. Jiménez-Sáez, J. Alanis-López, M. Villar (H. Virgen de la Macarena, Sevilla)

4

Supplemental Materials and Methods Genome-wide association study QC QC was conducted at both single marker and subject levels. Any marker that did not pass the following criteria was excluded from analysis: (i) genotype call rate in the batch of subjects greater than 95%, (ii) minor allele frequency greater than 1%, (iii) p-value for Hardy-Weinberg equilibrium greater than 10-7 in controls (if applicable). After applying these criteria, 249,893 markers (including SNPs and CNV probes) were discarded and 822,927 SNPs were left for downstream analysis. Any subject that did not pass the following criteria was excluded from analysis: (i) missing genotype rate < 0.1 among the SNPs that passed QC; (ii) not a sample duplicate or closely related based on estimated identity-by-descent (IBD) using PLINK v 1.05.22 In total, 10 cases were discarded on this basis. HLA genotyping Typing was performed by sequencing PCR products for exons 2 and 3 for HLA-A and B and exon 2 for DRB1, DQA1 and DQB1 on both strands using cycle sequencing with BigDye V3.1 (Applied Biosystems, Foster City, CA) chemistry. For HLA-A and B, typing, generic PCR amplifications were carried out for each locus.1 HLA-A had additional group-specific amplifications using an A2-specific site in intron 1. The A2-group contains the most common alleles among the HLA-A alleles and this additional allele sequencing separated A2-carrying heterozygotes and reduced ambiguous allele combinations. For HLA-B, six subgroup amplifications, TA, CG (Di-allelic sites in intron 1)2, B7, B15, B13, and Bw4 groups, were performed. HLA class II sequencing was carried out on amplicons produced by group-specific amplifications. A total of 17 gene- and group-specific DRB1 amplifications were performed using exon-based

5

amplification primers: DRB1*01, DRB1*15/16, DRB1*03/11/13/14, DRB1*03/11/13/14-86V, DRB1*11, DRB1*04, DRB1*04-86V, DRB1*07, DRB1*08/12, DRB1*09, DRB1*10. In addition to exon-based group-specific sequencing, intron based group-specific sequencing was used for DRB1*01-In, DRB1*15-In, DRB1*03/11/13/14-In, DRB1*04-In, DRB1*08-In, and DRB1*10-In. These newly added intron-based amplifications not only improved the resolution but also the accuracy of the typing by capturing variations at the beginning of exon 2 sequences which were missing in the exon-based group-specific amplicons. Sorting DRB1 sub-groups was accomplished by DRB generic typing. DRB generic typing was carried out by a hybridization-based sequence-specific oligonucleotide probe (SSO) method using AP-conjugated oligo probes. These SSO results were used for sorting DRB1 subgroups and also for quality assurance purposes. DQA1 high resolution typing was achieved by sequencing amplicons generated by DQA1 generic PCR products. DQB1 high resolution typing was achieved by sequencing amplicons generated by DQB1 generic, and three other DQB1 subgroups. Sequencing data files were analyzed using Histogenetics’ proprietary analysis programs Histomatcher and HistoMagic for HLA typing assignment of each sample. Allele assignments are based on IMGT/HLA Database release version 2.21.0 and Release date April 2008 (http://www.ebi.ac.uk/imgt/hla/). References 1.

Cereb N, Maye P, Lee S, Kong Y, Yang SY. Locus-specific amplification of HLA class I genes from genomic DNA: locus-specific sequences in the first and third introns of HLA-A, -B, and -C alleles. Tissue Antigens 1995;45:1-11.

2.

Cereb N, Yang SY. Dimorphic primers derived from intron 1 for use in the molecular typing of HLA-B alleles. Tissue Antigens 1997;50:74-6.

6

Figure S1.

Population structure of the cases and genetically-matched controls. Principal

component analysis plot for 201 cases and their 532 genetically matched controls. Blue dots: UK cases; Purple squares: DILIN cases; Green dots: Malaga; Red dots: Eudragene; Black: POPRES controls

7

(a)

(b)

(c)

Figure S2. Manhattan plots for the three groups separately. (a) UK cases (n=74) versus Northwestern European controls (n=306) (b) DILIN cases (n=51) versus Northwestern European controls (n=306) (c)Spanish DILI Registry cases (n=46) versus Spanish controls (n=160) 8

A.

B.

Figure S3

Quantile-quantile plot of p-values from case-control analyses. A. The comparison of

empirical p-value quantiles from the genome-wide analysis to the expected quantiles under the uniform distribution (null hypothesis). The black solid line represents the null model where observed p-values

9

match the expected values. The red dots represent observed p-values versus the expected values. B The same as in A, after conditioning on the top class I and II SNPs as in Figure 1D.

10

p = 0.071. Figure S4. Test of heterogeneity on A*0201 between Northwestern European and Spanish subjects.

11

Table S1.

Comparison of clinical characteristics between cohorts

Clinical characteristics

Test

UK

Spanish DILI Registry 0.87

DILIN

Fisher’s exact Two sample t-test Two sample t-test Two sample t-test Chisquare

0.24

EUDRAGENE 1

0.75

0.054

0.66

0.24

0.0035

0.042

0.19

0.021

Causality (CIOMS)

Peak Bilirubin

Gender Age Total days on drug Time to onset Pattern of injury

Peal ALT Peak ALP

Note

0.16

UK vs DILIN: 0.0028 -8 1.6x10 DILIN vs UK: 8.7x10-8 0.20 Combining cholestatic and mixed yields similar result 0.00091 DILIN contains more probable cases 0.017 In log scale

0.00065 0.0042

0.016

0.20

0.20

0.22

Two sample t-test

0.08

0.94

0.19

Two sample t-test Two sample t-test Two sample t-test

0.34

0.29

0.46

0.60

0.26

0.70

0.45

In log scale

0.31

0.59

0.14

0.79

In log scale

Each cell is the p-value from cohort vs all other cohorts.

12

Table S2.

Top SNPs among reported GWAS hits of autoimmune diseases

Overall Top SNPs p-value rs2476601 (chr: 1p13) rs6679677 (chr: 1p13)

1.3x10-4 2.3x10-4

OR (95% CI) 2.1 (1.5 – 3.2) 2.1 (1.4 – 3.1)

UK (74 cases vs 306 controls) OR p-value (95% CI) 2.0 0.025 (1.1-3.6) 2.0 0.025 (1.1-3.6)

US (51 cases vs 306 controls)

Spain (46 cases vs 160 controls)

p-value

OR (95% CI)

p-value

8.7x10-6

3.9 (2.2-7.0)

0.48

8.7x10-6

3.9 (2.2-7.0)

0.64

13

OR (95% CI) 1.4 (0.50 - 3.4) 1.2 (0.50 - 3.0)

MAF in nwEU 0.16 vs. 0.065 0.16 vs. 0.065

MAF in Spanish 0.087 vs. 0.063 0.076 vs. 0.063

Table S3.

Pair-wise r2 between top SNPs and HLA alleles

allele 1

allele 2

rs9274407/Minor

Spanish

Northwestern European

cases

controls

cases

controls

DQB1*0602

0.8

0.93

0.88

0.64

rs9274407/Minor

rs3135388/T

0.76

0.77

0.82

0.59

rs9274407/Minor + rs3135388/C

DQB1*0402

0.89

1

0.65

0.85

rs3135388/T

DQB1*0602

0.94

1

0.88

0.9

rs3135388/T

DRB1*1501

0.98

1

0.94

1

rs2523822/C

A*0201

0.96

0.91

0.64

0.87

DRB1*1501

DQB1*0602

0.91

1

0.94

0.90

14

Table S4. Frequencies of HLA alleles and haplotypes. HLA haplotypes inferred from unphased single gene alleles using Beagle Five gene haplotype (Class I and Class II) NW-EU Controls (n=107)

NW-EU Cases (n=134)

Spanish controls(n=121)

Spanish cases (n=48)

alleles

carriers

carriers

carriers

carriers

20

17

10

10

freq (%carriers) 4.13% (8.26%)

alleles

18

freq (%carriers) 7.46% (12.69%)

alleles

18

freq (%carriers) 8.41% (16.82%)

alleles

49

Overall frequency 0.0598

1

1

38

0.0463

2

2

0.93% (1.87%)

30

30

11.19% (22.39%)

1

1

0.41% (0.83%)

5

5

5.21% (10.42%)

22

0.0268

9

8

4.21% (7.48%)

6

6

2.24% (4.48%)

3

3

1.24% (2.48%)

4

4

4.17% (8.33%)

17

0.0207

5

5

2.34% (4.67%)

10

10

3.73% (7.46%)

2

2

0.83% (1.65%)

0

0

0.0% (0.0%)

17

0.0207

4

4

1.87% (3.74%)

2

2

0.75% (1.49%)

8

7

3.31% (5.79%)

3

3

3.13% (6.25%)

8

0.0098

2

2

0.93% (1.87%)

5

5

1.87% (3.73%)

0

0

0.0% (0.0%)

1

1

1.04% (2.08%)

Haplotype

count

A*0101B*0801DRB1*0301DQA1*0501DQB1*0201 A*0201B*0702DRB1*1501DQA1*0102DQB1*0602 A*0301B*0702DRB1*1501DQA1*0102DQB1*0602 A*0201B*4402DRB1*0401DQA1*0301DQB1*0301 A*2902B*4403DRB1*0701DQA1*0201DQB1*0201 A*0201B*5701DRB1*0701-

15

freq (%carriers) 1.04% (2.08%)

DQA1*0201DQB1*0303 A*2501B*1801DRB1*1501DQA1*0102DQB1*0602 A*0101B*5701DRB1*0701DQA1*0201DQB1*0303 A*3201B*1401DRB1*0701DQA1*0201DQB1*0201 A*0101B*0702DRB1*1501DQA1*0102DQB1*0602 A*0201B*4001DRB1*1302DQA1*0102DQB1*0604 A*1101B*2705DRB1*0101DQA1*0101DQB1*0501 A*2301B*4403DRB1*0701DQA1*0201DQB1*0201 A*3301B*1402DRB1*0102-

8

0.0098

2

2

0.93% (1.87%)

5

5

1.87% (3.73%)

0

0

0.0% (0.0%)

1

1

1.04% (2.08%)

7

0.0085

3

3

1.4% (2.8%)

1

1

0.37% (0.75%)

2

2

0.83% (1.65%)

1

1

1.04% (2.08%)

7

0.0085

3

3

1.4% (2.8%)

0

0

0.0% (0.0%)

3

3

1.24% (2.48%)

1

1

1.04% (2.08%)

6

0.0073

1

1

0.47% (0.93%)

5

5

1.87% (3.73%)

0

0

0.0% (0.0%)

0

0

0.0% (0.0%)

6

0.0073

1

1

0.47% (0.93%)

5

5

1.87% (3.73%)

0

0

0.0% (0.0%)

0

0

0.0% (0.0%)

6

0.0073

4

4

1.87% (3.74%)

0

0

0.0% (0.0%)

2

2

0.83% (1.65%)

0

0

0.0% (0.0%)

6

0.0073

3

3

1.4% (2.8%)

1

1

0.37% (0.75%)

2

2

0.83% (1.65%)

0

0

0.0% (0.0%)

6

0.0073

0

0

0.0% (0.0%)

0

0

0.0% (0.0%)

6

6

2.48% (4.96%)

0

0

0.0% (0.0%)

16

DQA1*0101DQB1*0501 A*0101B*0801DRB1*0101DQA1*0101DQB1*0501 A*0201B*0801DRB1*0301DQA1*0501DQB1*0201 A*0201B*1501DRB1*0101DQA1*0101DQB1*0501 A*0201B*4001DRB1*0404DQA1*0301DQB1*0302 A*0201B*4402DRB1*1301DQA1*0103DQB1*0603 A*0201B*4403DRB1*0701DQA1*0201DQB1*0201 A*2402B*3502DRB1*1104DQA1*0501DQB1*0301 A*2902B*4403DRB1*1501-

5

0.0061

3

3

1.4% (2.8%)

2

2

0.75% (1.49%)

0

0

0.0% (0.0%)

0

0

0.0% (0.0%)

5

0.0061

0

0

0.0% (0.0%)

4

4

1.49% (2.99%)

1

1

0.41% (0.83%)

0

0

0.0% (0.0%)

5

0.0061

0

0

0.0% (0.0%)

4

4

1.49% (2.99%)

0

0

0.0% (0.0%)

1

1

1.04% (2.08%)

5

0.0061

3

3

1.4% (2.8%)

2

2

0.75% (1.49%)

0

0

0.0% (0.0%)

0

0

0.0% (0.0%)

5

0.0061

1

1

0.47% (0.93%)

2

2

0.75% (1.49%)

1

1

0.41% (0.83%)

1

1

1.04% (2.08%)

5

0.0061

1

1

0.47% (0.93%)

2

2

0.75% (1.49%)

2

2

0.83% (1.65%)

0

0

0.0% (0.0%)

5

0.0061

0

0

0.0% (0.0%)

1

1

0.37% (0.75%)

2

2

0.83% (1.65%)

2

2

2.08% (4.17%)

5

0.0061

0

0

0.0% (0.0%)

2

2

0.75% (1.49%)

0

0

0.0% (0.0%)

3

3

3.13% (6.25%)

17

DQA1*0102DQB1*0602 A*3002B*1801DRB1*0301DQA1*0501DQB1*0201 Other

5

0.0061

0

572

0.697560976

149

0

0.0% (0.0%)

1

0.696261682

158

1

18

0.37% (0.75%)

2

0.589552239

195

2

0.83% (1.65%)

2

0.805785124

70

2

2.08% (4.17%)

0.729166667

Table S5. The relationship between phenotypes and the genotypes of top hits. Clinical SNP or HLA characteristics allele Bilirubin rs9274407/Minor (log(umol/L)) rs2523822/C DQB1*0602 DQB1*0402 B*1801 B*1801

Nw-EU

Spanish

Overall

0.83

0.63

0.70

0.57 0.92 0.75 0.092 0.30

0.14 0.55 0.72 0.56 0.31

0.77 0.82 0.94 0.11 0.27

Age

rs9274407/Minor rs2523822/C DQB1*0602 DQB1*0402 B*1801 rs9274407/Minor

0.92 0.43 0.83 0.56 0.88 0.46

0.9 0.32 0.61 0.26 0.38 0.87

0.84 0.93 0.62 0.31 0.36 0.42

rs2523822/C DQB1*0602 DQB1*0402 B*1801

0.94 0.57 0.62 0.87

0.96 0.52 0.24 0.0056

0.90 0.39 0.36 0.029

rs9274407/Minor 0.050

0.43

0.049

rs2523822/C DQB1*0602 DQB1*0402 B*1801

0.89 0.34 0.43 0.8

0.63 0.81 0.049 0.15

ALT (log(U/L))

ALP (log(U/L))

0.71 0.45 0.050 0.15

Each cell shows the p-value from linear regression.

19

Notes

hepatocellular cases only

ALT is positively correlated with B*1801 alleles in Spanish cases ALP is negatively correlated with rs9274407/Minor

Table S6. Test characteristics based on selected alleles comparing carriers to non-carriers and assuming the probability of AC-DILI to be 0.014%

Allele(s) rs9274407/Minor

rs2523822/C

rs9274407/Minor and rs2523822/C

DQB1*0602

A*0201

A*0201 and DQB1*0602

B*1801

B*1801 and DQB1*0602

specificity sensitivity PPV 1 - NPV specificity sensitivity PPV 1 - NPV specificity sensitivity PPV 1 - NPV specificity sensitivity PPV 1 - NPV specificity sensitivity PPV 1 - NPV specificity sensitivity PPV 1 - NPV specificity sensitivity PPV 1 - NPV specificity sensitivity PPV 1 - NPV

nw-EU 73.53% 58.40% 0.03% 0.008% 51.63% 74.40% 0.02% 0.007% 88.24% 47.20% 0.06% 0.008% 79.44% 51.56% 0.03% 0.008% 59.81% 74.22% 0.03% 0.006% 94.39% 41.41% 0.1% 0.009% N/C N/C N/C N/C N/C N/C N/C N/C

Spanish 84.38% 36.96% 0.03% 0.01% 51.25% 73.91% 0.02% 0.007% 92.50% 28.26% 0.05% 0.01% 91.07% 40.43% 0.06% 0.009% 56.25% 59.57% 0.02% 0.01% 95.54% 25.53% 0.08% 0.01% 90.18% 27.66% 0.04% 0.01% 99.11% 8.51% 0.13% 0.01%

20