Decoding the Hidden Secrets of SNP Data: Revealing Ancestral Origins, Genomic Predictions, and Polygenic Risk Score

The study of Single Nucleotide Polymorphism (SNPs) data provides a gateway to uncovering profound insights into ancestral origins, enomic predictions, and polygenic genetic scores.
We embarked on an exploration of SNP data using two distinct datasets available for the plant Arabidopsis thaliana. Firstly, we employed a constructed artificial dataset to simulate and analyse mutation increments over multiple generations. This artificial dataset allowed us to investigate the dynamics of genetic variation and its implications for ancestral lineage tracing by Principal Component Analysis (PCA). Secondly, we incorporated real data comprising 27,081 non-redundant SNPs. Leveraging this extensive dataset, our investigations aimed to explore the intricate genetic landscape of Arabidopsis thaliana and reveal crucial details about population structure, genetic diversity, and the potential functional implications of identified SNPs in relation to various metabolites.
We developed a Polygenic Risk Score (PRS) tool implemented in Python - PGine: Py/Bioconda software package for the calculation of polygenic risk scores in plants that may be useful for new breeding strategies. Subsequently, we integrated diverse computational approaches to achieve Genomic Prediction models.
Our study reveals the hidden information embedded in SNP data, thereby improving our understanding of general genetic variation and its implications for different fields, including predictive modelling, population genetics and evolutionary biology.