IADR Abstract Archives
The Heritability of Dental Traits
Objectives: The genetic underpinning of oral health has long been theorized but little information exists on the heritable variance in common dental disease traits explained by high-density genotype data. Our objective was to derive estimates of heritability (h2) in clinical diagnoses and measures of common oral diseases explained by single nucleotide polymorphisms (SNPs) genotyped in the context of a genome-wide association study (GWAS).
Methods: We used existing data from a GWAS of oral health traits conducted in the context of the Dental Atherosclerosis Risk In Communities (ARIC) study. Oral health measures included: number of remaining teeth (NT), dental caries (DMFS index), tooth morbidity (DMTFS index including tooth surfaces lost due to all causes), extent scores of probing depth ≥4mm (EPD4) and attachment loss ≥4mm (EAL4), mean interproximal attachment loss (MALI), and high colonization (top quintile) with periodontal pathogens of the orange (HIOR) and red (HIRED) complex. We used SNPs directly genotyped on the Affymetrix Genome-Wide Human SNP Array 6.0 chip with minor allele frequency of >5% (n=656,292). H2 was calculated with Visscher’s Genome-wide Complex Trait Analysis (GCTA) software, using a random-effects mixed linear model and restricted maximum likelihood (REML) regression adjusting for age, sex and ancestry (10 principal components).
Results: Heritability estimates (standard error) were of modest magnitude and with relatively large standard errors for most clinical traits—NT=0.11 (0.07), EPD4=0.19 (0.07), EAL4=0.08 (0.07), DMFS=0.13 (0.08), DMTFS=0.24 (0.07) and higher for periodontal pathogen colonization—HIOR=0.46 (0.32) and HIRED=0.53 (0.31).
Conclusions: The observed variance explained by GWAS SNPs was modest for most clinical measures and was higher for bacterial colonization traits, although all these estimates were imprecise. The efficient interrogation of oral health and disease genomics will likely benefit from the examination of multiple related or composite traits and biologically-informed endophenotypes in the context of large cohorts or pooled samples.
Methods: We used existing data from a GWAS of oral health traits conducted in the context of the Dental Atherosclerosis Risk In Communities (ARIC) study. Oral health measures included: number of remaining teeth (NT), dental caries (DMFS index), tooth morbidity (DMTFS index including tooth surfaces lost due to all causes), extent scores of probing depth ≥4mm (EPD4) and attachment loss ≥4mm (EAL4), mean interproximal attachment loss (MALI), and high colonization (top quintile) with periodontal pathogens of the orange (HIOR) and red (HIRED) complex. We used SNPs directly genotyped on the Affymetrix Genome-Wide Human SNP Array 6.0 chip with minor allele frequency of >5% (n=656,292). H2 was calculated with Visscher’s Genome-wide Complex Trait Analysis (GCTA) software, using a random-effects mixed linear model and restricted maximum likelihood (REML) regression adjusting for age, sex and ancestry (10 principal components).
Results: Heritability estimates (standard error) were of modest magnitude and with relatively large standard errors for most clinical traits—NT=0.11 (0.07), EPD4=0.19 (0.07), EAL4=0.08 (0.07), DMFS=0.13 (0.08), DMTFS=0.24 (0.07) and higher for periodontal pathogen colonization—HIOR=0.46 (0.32) and HIRED=0.53 (0.31).
Conclusions: The observed variance explained by GWAS SNPs was modest for most clinical measures and was higher for bacterial colonization traits, although all these estimates were imprecise. The efficient interrogation of oral health and disease genomics will likely benefit from the examination of multiple related or composite traits and biologically-informed endophenotypes in the context of large cohorts or pooled samples.