What could be the role of genetic tests and machine learning of AXIN2variant dominance in non-syndromic hypodontia? A case-control study in orthodontically treated patients

Autor: Alhazmi, Nora, Alaqla, Ali, Almuzzaini, Bader, Aldrees, Mohammed, Alnaqa, Ghaida, Almasoud, Farah, Aldibasi, Omar, Alshamlan, Hala
Zdroj: Progress in Orthodontics; December 2024, Vol. 25 Issue: 1
Abstrakt: Background: Hypodontia is the most prevalent dental anomaly in humans, and is primarily attributed to genetic factors. Although genome-wide association studies (GWAS) have identified single-nucleotide polymorphisms (SNP) associated with hypodontia, genetic risk assessment remains challenging due to population-specific SNP variants. Therefore, we aimed to conducted a genetic analysis and developed a machine-learning-based predictive model to examine the association between previously reported SNPs and hypodontia in the Saudi Arabian population. Our case–control study included 106 participants (aged 8–50 years; 64 females and 42 males), comprising 54 hypodontia cases and 52 controls. We utilized TaqManTMReal-Time Polymerase Chain Reaction and allelic genotyping to analyze three selected SNPs (AXIN2: rs2240308, PAX9: rs61754301, and MSX1: rs12532) in unstimulated whole saliva samples. The chi-square test, multinomial logistic regression, and machine-learning techniques were used to assess genetic risk by using odds ratios (ORs) for multiple target variables. Results: Multivariate logistic regression indicated a significant association between homozygous AXIN2rs2240308 and the hypodontia phenotype (ORs [95% confidence interval] 2.893 [1.28–6.53]). Machine-learning algorithms revealed that the AXIN2homozygous (A/A) genotype is a genetic risk factor for hypodontia of teeth #12, #22, and #35, whereas the AXIN2homozygous (G/G) genotype increases the risk for hypodontia of teeth #22, #35, and #45. The PAX9homozygous (C/C) genotype is associated with an increased risk for hypodontia of teeth #22 and #35. Conclusions: Our study confirms a link between AXIN2and hypodontia in Saudi orthodontic patients and suggeststhat combining machine-learning models with SNP analysis of saliva samples can effectively identify individuals with non-syndromic hypodontia.
Databáze: Supplemental Index
Externí odkaz: