SNP2Prot – Protein Variant Effect Prediction for Arabidopsis Thaliana

Genomic variants in protein-coding genes, such as single nucleotide polymorphisms (SNPs), can lead to alterations in the primary structure of the encoded protein. Such non-synonymous SNPs (nsSNPs) may significantly affect the structure, dynamics, and function of a protein. Thanks to advances in DNA sequencing technologies and extensive research efforts, more than 1,100 genomes of the model organism Arabidopsis thaliana have been sequenced, providing a solid foundation to study genomic variation.

While numerous studies in recent years have focused on the relationship between genomic variation and phenotype in plants, the role of proteins, standing between genotype and phenotype, has often been overlooked. However, recently developed deep learning based methods, such as AlphaFold2 and ESMFold, enable highly accurate protein structure predictions, allowing us to investigate the impact of genomic variation on protein structure and its influence on protein function.

Understanding the connection between genomic variation, protein function and phenotype will not only provide insights into evolutionary principles, but also offers opportunities to engineer plants with desirable traits. This is particularly crucial in addressing the challenges posed by climate change, resource limitations, and population growth, which threaten global plant production.

Our goal is to better understand the landscape and effects of nsSNPs on protein structure and function in the model plant A.thaliana. We leverage existing tools and develop new AI methods to predict the impact of SNPs on protein properties such as structure, thermostability, or binding affinities. This research will provide a foundation for designing functionally enhanced proteins in the future.