Harnessing Artificial Intelligence and Machine Learning for Identifying Quantitative Trait Loci (QTL) Associated with...

Area of Science:

• Genetics and Genomics
• Plant Breeding
• Computational Biology

Background:

• Seed quality traits are vital for crop productivity and nutritional value.
• Traditional quantitative trait loci (QTL) mapping methods face challenges with complex genetic architectures and environmental interactions.
• Artificial intelligence (AI) and machine learning (ML) offer advanced solutions for trait prediction and marker-trait association.

Purpose of the Study:

• To provide an integrated overview of AI/ML applications in QTL mapping and seed trait prediction.
• To highlight key AI/ML methodologies and their effectiveness across diverse crop species.
• To explore the integration of multi-omics data for enhanced QTL resolution.

Main Methods:

• Review of AI/ML techniques including LASSO regression, Random Forest, Gradient Boosting, ElasticNet, Convolutional Neural Networks (CNNs), and Graph Neural Networks (GNNs).
• Case study on soybean seed mineral nutrient accumulation using ML models.
• Exploration of AI-driven integration of genomics, transcriptomics, metabolomics, and phenomics data.

Main Results:

• AI/ML models effectively identify significant single nucleotide polymorphisms (SNPs) associated with seed traits, as demonstrated in soybean.
• LASSO and ElasticNet models showed superior predictive accuracy compared to tree-based methods in soybean.
• AI/ML has successfully enhanced QTL detection in various crops like wheat, lettuce, rice, and cotton.

Conclusions:

• AI/ML significantly enhances the accuracy and efficiency of QTL mapping and seed trait prediction.
• Integration of multi-omics data with AI/ML offers higher resolution in genetic analyses.
• AI holds transformative potential for accelerating genomic-assisted breeding and developing improved crop varieties.