Supplementary TABLES
Supplementary Table 1: Characteristics of all samples including affected individuals
Discovery- sample / Replication-1 / Replication-2Number of individuals a (after QC) / 200 (200) / 186 (183) / 324 (316)
Age (years)
(distribution (mean ± sd)) / 8 – 19
(12.53 ± 2.22) / 8 – 18
(11.40 ± 2.19) / 7 – 11
(9.47 ± 0.80)
Male / female / 150 / 50 / 130 / 56 / 170 / 154
Degree of dyslexia severity b (distribution (mean ± sd)) / 2.01 – 4.27
(2.65 ± 0.45) / 1.09 – 2.87
(1.82 ± 0.30) / 0.76 – 3.45
(1.47 ± 0.47)
Mathematical calculation
(distribution (mean ± sd)) / -5.81 – 1.51
(-0.29 ± 1.02) / -0.80 – 2.17
(-0.13 ± 1.10) / -3.53 – 2.13
(-0.49 ± 0.84)
Numerosity judgement
(distribution (mean ± sd)) / -4.42– 1.73
(-0.147 ± 1.05) / -4.00 – 2.33
(-0.06 ± 1.05) / -7.96 – 2.08
(-3.70 ± 0.89)
Basic mathematical abilities factor
(distribution (mean ± sd)) / -10.81 – 3.75
(-0.92 ± 2.14) / -7.04 – 4.30
(-0.36 ± 1.90) / -5.06 – 1.53
(-0.43 ± 0.70)
a Numbers of individuals included. Numbers for whom genotypic and phenotypic data were available are given in parentheses.
b Measured as x-fold standard deviation between expected and observed spelling score (discovery and replication-1), or between expected and observed reading score (replication-2).
Abbreviations: QC – quality control
Supplementary Table 2: Characteristics for samples including non-affected/non-screened individuals
Number of individuals a (after QC) / 421 (391) / 1 081 (1 080)
Age (years)
(distribution (mean ± sd)) / 7 – 11
(9.45 ± 0.79) / 7, 9 and 10b
Male / female / 211 / 210 / 469 / 612
Basic mathematical abilities factor / MA_UK
(distribution (mean ± sd)) / -2.56 – 2.97
(0.42 ± 0.79) / -3.36 – 2.25
(0.07±0.83)
a Numbers of individuals included. Numbers for whom genotypic and phenotypic data were available are given in parentheses.
b Children of the TEDS sample were tested at the ages of 7, 9, and 10 years (MA_UK). All children with phenotypic data from all three time-points were included.
Abbreviations: QC – quality control
Supplementary Table 3: Power analysis for all samples
α= 0.05 / α= 7.2x 10-08 / α= 0.05 / α= 7.2x 10-08
Discovery-sample / 200 / 100.00% / 68.25% / 88.49% / 1.30%
Replication-1 / 183 / 73.93% / 0.27% / 85.60% / 0.91%
Replication-2 / 316 / 38.65% / 0.01% / 97.79% / 7.87%
All cases / 699 / 100.00% / 69.89% / 100.00% / 69.89%
German/Austrian control / 391 / 5.49% / <0.01% / 5.31% / <0.01%
TEDS / 1 080 / 38.87% / 0.01% / 5.87% / <0.01%
All controls / 1 471 / 6.19% / <0.10% / 6.19% / <0.10%
a Respective effect sizes as calculated in Table 1
b Weighted average of effect size over all samples (R²=4.87% for cases, R²=0.007% for controls)
Supplementary Table 4: Demographic information and global sulcus measurements of the MRI sample
AA/AG / GG
N / 56 / 23 / n.a.
Age, years b / 44.5 (14.5) / 39.8 (13.6) / 0.187
Gender, %women c / 47.8 / 48.2 / 1.0
Right depthavg, mm d / 15.72 (0.83) / 15.59 (0.78) / 0.163
Left depthavg, mm d / 15.56 (1.06) / 15.64 (0.71) / 0.803
Right SVtotal, mm3 d / 22 459 (2,683) / 22 078 (2,355) / 0.158
Left SVtotal, mm3 d / 21 786 (3,244) / 21 607 (2,543) / 0.767
a mean values and standard deviations per genotype group
b two-tailed t-test
c Pearson-Chi-Square test
d Analysis of covariance using gender as covariate.
Abbreviations: N – number of individuals; n.a. - not applicable; avg. - average; SV - sulcus volume
Supplementary Table 5: Haplotype P-values for BMAF in the German sample.
First SNP / Last SNP / Alleles / Haplotypic P-values in German sample a,bdiscovery / replication-1 / combined
Two marker haplotype
rs3859865 / rs133885 / AA / 1.50E-004 / 1.26E-001 / 6.82E-005
GA / 1.35E-002 / 1.75E-002 / 6.55E-004
GG / 3.65E-007 / 1.50E-003 / 1.80E-009
rs133885 / rs4822649 / AA / 1.27E-002 / 1.57E-001 / 1.07E-002
AC / 4.50E-004 / 9.52E-003 / 3.17E-006
GC / 3.65E-007 / 1.50E-003 / 1.80E-009
Three marker haplotype
rs133871 / rs133885 / GAA / 1.50E-004 / 9.47E-002 / 4.51E-005
GGA / 7.58E-001 / 1.64E-001 / 2.95E-001
AGA / 2.24E-003 / 7.26E-002 / 9.99E-004
AGG / 5.71E-007 / 2.94E-001 / 8.64E-010
rs3859865 / rs4822649 / GAA / 1.78E-002 / 1.30E-001 / 1.12E-002
AAC / 2.37E-004 / 1.15E-001 / 7.84E-005
GAC / 4.26E-001 / 9.62E-002 / 4.64E-002
GGC / 3.65E-007 / 1.19E-003 / 1.29E-009
rs133885 / rs13058434 / ACT / 6.71E-001 / 2.18E-001 / 3.18E-001
AAC / 1.27E-002 / 1.27E-001 / 8.67E-003
ACC / 8.32E-005 / 4.55E-002 / 9.01E-006
GCC / 3.30E-007 / 1.26E-003 / 1.49E-009
Four marker haplotype
rs9620553 / rs133885 / CGAA / 2.19E-002 / 1.29E-002 / 3.21E-004
TGAA / 3.23E-003 / 8.14E-001 / 1.74E-002
CGGA / 7.77E-001 / 1.74E-001 / 3.28E-001
TAGA / 1.53E-003 / 7.24E-002 / 6.51E-004
CAGG / 4.43E-002 / 2.77E-001 / 1.48E-002
TAGG / 1.31E-005 / 6.42E-004 / 2.15E-008
rs133871 / rs4822649 / AGAA / 9.09E-003 / 1.24E-001 / 6.15E-003
GAAC / 2.37E-004 / 9.53E-002 / 6.35E-005
GGAC / 9.35E-001 / 1.56E-001 / 2.07E-001
AGAC / 1.73E-001 / 3.60E-001 / 9.90E-002
AGGC / 6.56E-007 / 5.96E-004 / 9.06E-010
rs3859865 / rs13058434 / GACT / 6.74E-001 / 2.44E-001 / 3.44E-001
GAAC / 1.78E-002 / 1.30E-001 / 1.12E-002
AACC / 2.37E-004 / 1.34E-001 / 8.59E-005
GACC / 9.42E-002 / 1.65E-001 / 3.50E-002
GGCC / 3.21E-007 / 1.19E-003 / 1.08E-009
rs133885 / rs133903 / AACA / 3.00E-002 / 1.54E-001 / 1.91E-002
ACCA / 4.97E-002 / 1.06E-001 / 8.02E-003
GCCA / 8.62E-001 / 4.10E-003 / 3.90E-002
ACTC / 8.72E-001 / 1.49E-001 / 2.15E-001
AACC / 1.17E-001 / 5.04E-001 / 1.35E-001
ACCC / 3.00E-004 / 2.02E-001 / 1.88E-004
GCCC / 1.23E-007 / 2.03E-001 / 2.31E-006
Five marker haplotype
rs133866 / rs133885 / TCGAA / 2.17E-002 / 1.32E-002 / 3.12E-004
CTGAA / 1.61E-003 / 9.29E-001 / 1.57E-002
TCGGA / 9.95E-001 / 2.39E-001 / 2.59E-001
CTAGA / 1.55E-003 / 7.21E-002 / 6.51E-004
CTAGG / 1.92E-005 / 5.59E-004 / 2.92E-008
rs9620553 / rs4822649 / TAGAA / 4.22E-003 / 1.25E-001 / 2.87E-003
CGAAC / 1.87E-002 / 1.39E-002 / 3.28E-004
TGAAC / 5.79E-003 / 8.14E-001 / 2.27E-002
CGGAC / 9.36E-001 / 1.67E-001 / 2.36E-001
TAGAC / 1.73E-001 / 3.79E-001 / 1.04E-001
TAGGC / 1.28E-005 / 6.08E-004 / 1.66E-008
rs133871 / rs13058434 / GGACT / 7.98E-001 / 1.62E-001 / 2.21E-001
AGAAC / 8.77E-003 / 1.24E-001 / 6.00E-003
GAACC / 2.37E-004 / 1.09E-001 / 8.17E-005
AGACC / 1.02E-001 / 1.77E-001 / 3.58E-002
AGGCC / 6.23E-007 / 5.96E-004 / 8.89E-010
rs3859865 / rs133903 / GAACA / 3.98E-002 / 1.35E-001 / 1.99E-002
AACCA / 1.05E-001 / 1.65E-001 / 2.03E-002
GACCA / 2.64E-001 / 4.07E-001 / 1.97E-001
GGCCA / 8.50E-001 / 3.76E-003 / 3.69E-002
GACTC / 9.06E-001 / 1.67E-001 / 2.30E-001
GAACC / 1.43E-001 / 6.26E-001 / 1.98E-001
AACCC / 9.64E-004 / 3.46E-001 / 8.79E-004
GGCCC / 1.84E-007 / 1.99E-001 / 2.27E-006
rs133885 / rs3848858 / ACTCG / 9.66E-001 / 1.94E-001 / 1.53E-001
ACCCG / 1.51E-003 / 2.17E-002 / 1.83E-004
GCCCG / 3.88E-003 / 1.31E-001 / 3.83E-003
AACAA / 2.71E-002 / 1.65E-001 / 1.85E-002
ACCAA / 5.29E-002 / 9.80E-002 / 7.58E-003
GCCAA / 8.36E-001 / 2.64E-003 / 3.05E-002
ACTCA / 7.45E-001 / 3.99E-001 / 6.65E-001
AACCA / 1.42E-001 / 7.41E-001 / 3.39E-001
GCCCA / 5.03E-004 / 8.80E-001 / 8.75E-004
a Sliding-window approach on up to five SNPs, always including rs133885.
b P-values are depicted in bold if genome-wide significant (P<7.2×10-08) in the discovery / combined sample, or P0.05 in the replication-1 sample.
Supplementary Table 6: Association analysis for rs133885 and the mathematical subphenotypes in dyslexia samples
mathematical calculation MC
(risk allele b) / Distinct test batteries of mathematical calculation / Association P-value a
numerosity judgement
(risk allele b)
Addition
(risk allele b) / Multiplication (risk allele b)
Discovery-sample / 7.38 x 10-08 (G) / 7.88 x 10-08 (G) / 5.44 x 10-06 (G) / 2.24 x 10-07(G)
Replication-1 sample / 6.87 x 10-02 (G) / 1.49 x 10-02 (G) / 1.82 x 10-01 (G) / 7.57 x 10-02 (G)
Replication-2 sample c / 1.28 x 10-01 (G) / 1.48 x 10-01 (G) / 1.26 x 10-01 (G) / 5.43 x 10-02 (G)
All cases d / 1.31 x 10-05 (G) / 9.45 x 10-06 (G) / 5.33 x 10-04 (G) / 5.90 x 10-06 (G)
a P-values are one-sided (except for discovery-sample and combined case sample), based on the carrier A-model.
b Risk allele defined as the allele contributing to a weaker task performance.
c Sample origin included as a covariate.
d Sample origin, spelling and word reading were used as covariates.
Supplementary Table 7: Association results for BMAF in dyslexia samples after inverse rank-based transformation
Discovery / 200 / 5.78 x 10-09 / 1.04 x 10-07
Replication-1 / 183 / 4.91 x 10-03 / 4.19 x 10-03
Replication-2 c / 316 / 4.46 x 10-02 / 8.82 x 10-02
All cases d / 699 / 7.71 x 10-10 / 8.43 x 10-10
a Number of individuals included in the statistical analysis
b P-values are one-sided, based on the carrier A-model (except for discovery-sample and combined sample).
c Sample origin included as a covariate.
d Sample origin, spelling and word reading were used as covariates.
Abbreviations: N – number of individuals
Supplementary Table 8: Association analysis for rs133885 and the mathematical sub-measurements in the TEDS sample
Risk allele a / G / G / G
P-value b / 0.0294 / 0.071 / 0.055
0.041
a defined as the allele contributing to a weaker task performance.
b P-values are one-sided, based on the carrier A-model. Bold if significant (P<0.05)
Supplementary Table 9: Statistics for non-synonymous variants within the N-terminus of MYO18Blong in the German discovery-sample.
Mathematical calculation / Numerosity judgement / BMAF
(A) Single-marker analysis b,c
rs61734946 d / 5.60 x 10-04 / 3.29 x 10-04 / 8.11 x 10-05
Conditioned on rs133885 / total / 0.3568 / 0.2744 / 0.3607
Subgroup A-carrier / 0.2192 / 0.3701 / 0.1736
Subgroup non-A-carrier / 0.6637 / 0.1537 / 0.7231
rs13058434 / 0.9416 / 0.4452 / 0.6733
Conditioned on rs133885 / total / 0.7522 / 0.0847 / 0.2538
Subgroup A-carrier / 0.0668 / 0.6241 / 0.0158
Subgroup non-A-carrier / 0.6661 / 0.5745 / 0.6229
(B) Haplotype analysis
rs61734946 - rs133885 / 1.56 x 10-05 / 3.62 x 10-07 / 1.03 x 10-07
rs133885 - rs13058434 / 1.27 x 10-05 / 4.37 x 10-06 / 9.19 x 10-08
rs61734946 - rs133885 - rs13058434 / 3.85 x 10-05 / 8.10 x 10-06 / 3.51 x 10-07
(C) Interaction analysis
rs61734946 - rs133885 / 0.0833 / 0.0290 / 0.0338
rs133885 - rs13058434 / 0.0979 / 0.1200 / 0.1118
rs61734946 - rs13058434 / 0.4261 / 0.1125 / 0.1812
a P-values in bold if nominally significant
b Single-marker analysis in two scenarios: (i) independent of; or (ii) as a function of rs133885 genotype.
c Linkage disequilibrium between the three variants, as measured by r2, was low (rs61734964-rs13058434: r²=0.012; rs61734964-rs133885: r²=0.321; rs13058434-rs133885: r²=0.059).
d Genotype distribution deviated from Hardy-Weinberg (P=0.004). However, as this sample was comprised of selected cases, this was not used as an exclusion criterion.
Abbreviations: BMAF – basic mathematical abilities factor
Supplementary Table 10: Effect sizes for different grades of spelling disability in the German sample
Degree of severity a / Number of individuals / Genotypic distributionAA+AG / GG b / Effect size c
SD > 1 / 383 / 258 (0.67) / 125 (0.33) / 9.4%
SD > 1.5 / 356 / 237 (0.67) / 119 (0.33) / 10.0%
SD > 2.0 / 241 / 157 (0.65) / 84 (0.35) / 10.8%
SD > 2.5 / 112 / 71 (0.63) / 41 (0.37) / 19.1%
SD > 3.0 / 41 / 28 (0.68) / 13 (0.32) / 25.1%
a Grades were defined according to the spelling discrepancy score, i.e. the standard deviation of observed spelling score from that expected on the basis of IQ and age
b according to carrier-A model; absolute numbers (frequency)
c measured according to the R²-goodness-of-fit coefficient (carrier-A model)
Abbreviations: SD – standard deviation