DNA-based prediction of eye color in Latin American population applying Machine Learning models

Area of Science:

• Genetics and Bioinformatics
• Computational Biology
• Machine Learning Applications

Background:

• Decreasing costs of DNA sequencing and genotyping enable large-scale genetic databases.
• These databases facilitate analysis of genotype-phenotype relationships, including visible traits.
• Understanding genetic determinants of traits like eye color is an active research area.

Purpose of the Study:

• To enhance the prediction accuracy of human eye color using genetic information.
• To evaluate the efficacy of various Machine Learning models for genotype-based trait prediction.
• To explore the application of AI in predicting observable human characteristics from DNA.

Main Methods:

• Utilized a dataset comprising genetic data from 308 volunteers in Buenos Aires, Argentina.
• Applied and compared multiple Machine Learning algorithms for genotype-to-phenotype prediction.
• Focused on predicting the specific trait of eye color.

Main Results:

• Achieved competitive prediction accuracy for eye color based on genetic markers.
• Demonstrated the practical utility of Machine Learning models in this predictive task.
• The models showed significant potential for real-world applications.

Conclusions:

• Artificial intelligence (AI) offers a powerful tool for analyzing human genetic data.
• Accurate prediction of visible traits like eye color from genotype is feasible.
• This research supports AI's role in advancing fields such as personalized medicine, biometrics, and forensics.