Supplementary Material

Genetic and environmental components of family history

in type 2 diabetes

Marilyn C Cornelis,Noah Zaitlen, Frank B. Hu, Peter Kraft, and Alkes L. Price

M.C. Cornelis

Department of Preventive Medicine, Northwestern University Feinberg School of Medicine

680 N Lake Shore, Suite 1400, Chicago IL 60611, USA

Telephone: 312-503-4548

Fax: 312-908-9588

Email:

Supplementarymaterials and methods

Nested case-control GWAS of type 2 diabetes

Assessment of type 2 diabetes

Diabetescases were defined as self-reported diabetes confirmed by avalidated supplementary questionnaire(Hu et al. 2001; Manson et al. 1991). For cases before 1998, diagnosis was made using criteria consistentwith those proposed by the National Diabetes Data Group (Classification and diagnosis of diabetes mellitus and other categories of glucose intolerance. National Diabetes Data Group 1979). We used the American Diabetes Associationdiagnostic criteria for diagnosis of diabetes cases duringthe 1998 and 2000 cycles(Report of the Expert Committee on the Diagnosis and Classification of Diabetes Mellitus 1997). Controls were defined as those free of diabetes at the time of diagnosis of the case. Although controls were originally matched per case (by gender, year of birth, month of blood collection, and fasting status), matched pairs were broken because not all subjects gave informed consent for submission of their GWAS data to dbGaP.

References

National Diabetes Data GroupClassification and diagnosis of diabetes mellitus and other categories of glucose intolerance. National Diabetes Data Group (1979). Diabetes 28:1039-1057

Hu FB, Leitzmann MF, Stampfer MJ, Colditz GA, Willett WC, Rimm EB (2001) Physical activity and television watching in relation to risk for type 2 diabetes mellitus in men. Arch Intern Med 161:1542-1548

Manson JE et al. (1991) Physical activity and incidence of non-insulin-dependent diabetes mellitus in women. Lancet 338:774-778

Report of the Expert Committee on the Diagnosis and Classification of Diabetes Mellitus (1997). Diabetes Care 20:1183-1197

Supplementary Tables

Table S1 Variance in T2D risk explained by family history of diabetes (FH) and known genetic (G1) and environmental (E1) risk factors under different definitions of FH in the Nurses’ Health Studya

Model / FH definition
parental (y/n) / sibling (y/n) / maternal (y/n) / paternal (y/n) / parental or sibling (y/n) / maternal (y/n) + paternal (y/n) + sibling (y/n)
FH / 0.069 / 0.024 / 0.052 / 0.024 / 0.078 / 0.085
G1 / 0.032 / 0.032 / 0.032 / 0.032 / 0.032 / 0.032
E1 / 0.179 / 0.179 / 0.179 / 0.179 / 0.179 / 0.179
FH, G1 / 0.093 / 0.052 / 0.077 / 0.053 / 0.099 / 0.105
FH, E1 / 0.222 / 0.195 / 0.214 / 0.194 / 0.227 / 0.234
G1, E1 / 0.207 / 0.207 / 0.207 / 0.207 / 0.207 / 0.207
FH, G1, E1 / 0.244 / 0.220 / 0.237 / 0.220 / 0.248 / 0.254

aObserved-scaled σ2 from linear regressions of FH, G1 (T2D GRS) and E1 (age, BMI, smoking status, diet quality, alcohol consumption and physical activity) and their combinations (specified in column 1) on T2D case-status.

Table S2 Variance in T2D risk explained by family history of diabetes (FH) and known genetic (G1) and environmental (E1) risk factors under different definitions of FH in the Health Professionals Follow-up Studya

Model / FH definition
parental (y/n) / sibling (y/n) / maternal (y/n) / paternal (y/n) / parental or sibling (y/n) / maternal (y/n) + paternal (y/n) + sibling (y/n)
FH / 0.062 / 0.040 / 0.033 / 0.035 / 0.063 / 0.085
G1 / 0.031 / 0.031 / 0.031 / 0.031 / 0.031 / 0.031
E1 / 0.185 / 0.185 / 0.185 / 0.185 / 0.185 / 0.185
FH, G1 / 0.084 / 0.066 / 0.060 / 0.061 / 0.085 / 0.105
FH, E1 / 0.226 / 0.213 / 0.206 / 0.209 / 0.227 / 0.243
G1, E1 / 0.212 / 0.212 / 0.212 / 0.212 / 0.212 / 0.212
FH, G1, E1 / 0.247 / 0.237 / 0.230 / 0.232 / 0.248 / 0.263

aObserved-scaled σ2 from linear regressions of FH, G1 (T2D GRS) and E1 (age, BMI, smoking status, diet quality, alcohol consumption and physical activity) and their combinations (specified in column 1) on T2D case-status.

Table S3 Variance in T2D risk explained by family history of diabetes (FH) and known environmental (E1) risk factors under different definitions of FHa based on the full Nurses’ Health Study cohort (84,008 women at baseline and 8,591 incident cases of T2D in follow-up) a

Model / FH definition
parental (y/n) / sibling (y/n) / maternal (y/n) / paternal (y/n) / parental or sibling (y/n) / maternal (y/n) + paternal (y/n) + sibling (y/n)
FH / 0.0204 / 0.008 / 0.014 / 0.009 / 0.022 / 0.027
E1 / 0.086 / 0.086 / 0.086 / 0.086 / 0.086 / 0.086
FH, E1 / 0.100 / 0.092 / 0.095 / 0.093 / 0.101 / 0.104

aObserved-scaled σ2 from linear regressions of FH and E1 (age, BMI, smoking status, diet quality, alcohol consumption and physical activity) and their combination (specified in column 1) on T2D case-status.

Table S4 Variance in T2D risk explained by family history of diabetes (FH) and known environmental (E1) risk factors under different definitions of FHa based on the full Health Professional Follow-up Study cohort (39,694 men at baseline and 2979 incident cases of T2D in follow-up) a

Model / FH definition
parental (y/n) / sibling (y/n) / maternal (y/n) / paternal (y/n) / parental or sibling (y/n) / maternal (y/n) + paternal (y/n) + sibling (y/n)
FH / 0.019 / 0.014 / 0.012 / 0.010 / 0.022 / 0.029
E1 / 0.051 / 0.051 / 0.051 / 0.051 / 0.051 / 0.051
FH, E1 / 0.066 / 0.062 / 0.060 / 0.059 / 0.069 / 0.075

aObserved-scaled σ2 from linear regressions of FH and E1 (age, BMI, smoking status, diet quality, alcohol consumption and physical activity) and their combination (specified in column 1) on T2D case-status.

Supplementary Figures

ab

Fig.S1Family shared environment () under simulation models with varying number of siblings. Values plotted are the best estimates of ξ for a dataset (,see Materials and Methods),where the heritability of type 2 diabetes (σG12 + σG22) is 0.40, σG12=0.03, σE12=0.18, σG22=0.37, σE22=0.42 and x=0.0 or 0.5. Results for NHS (a) and HPFS (b).
a

b

Fig.S2 Estimates of family shared environment (). Results for NHS (a) and HPFS (b) under simulation models, where the heritability of type 2 diabetes (σG12 + σG22) is 0.40,σG12=0.03, σE12=0.18, σG22=0.37, σE22=0.42, number of first degree relatives= 3 (mother, father, sibling) and x=0.0, 0.4, 0.5 or 0.6.

ab

Fig.S3 Family shared environment () under simulation models with varying heritability estimates of type 2 diabetes. Heritability estimates range from 0.20 to 0.60 with increments of 0.05. σG12 and σE12 are held constant with values of 0.03 and 0.18, respectively. Results for NHS (a) and HPFS (b).

Fig.S4 Family shared environment () under simulation models with varying proportions of E1 explained by genetics (x). σG12=0.03, σE12=0.18, σG22=0.37, σE22=0.42. E1 is partitioned into a genetic component E1G (σE1G2=σE12(x)) and an environmental component E1E (σE1E2=σE12-σE1G2), where x ranges from 0.0 to 0.75 with 0.05 increments.

1