Harnessing the predicted maize pan-interactome for putative gene function prediction and prioritization of candidate genes for important traits

Authors: PORETSKY, ELLY - Oak Ridge Institute For Science And Education (ORISE); CAGIRICI, BUSRA - Stanford University; Andorf, Carson; Sen, Taner

Submitted to: Genetics
Publication Type: Peer Reviewed Journal
Publication Acceptance Date: 3/8/2024
Publication Date: 3/16/2024
Citation: Poretsky, E., Cagirici, B.H., Andorf, C.M., Sen, T.Z. 2024. Harnessing the predicted maize pan-interactome for putative gene function prediction and prioritization of candidate genes for important traits. Genetics. https://doi.org/10.1093/g3journal/jkae059.
DOI: https://doi.org/10.1093/g3journal/jkae059

Interpretive Summary: Maize, an essential crop with significant agricultural importance, has been the subject of extensive research, resulting in a wealth of genomic and phenotypic data. The recent release of the genome assemblies and annotations for the 26 maize inbred lines have enabled large-scale pan-genomic comparative studies. These studies have expanded the understanding of agronomically important traits by identifying additional loci using the NAM mapping population and exploring trait-specific co-expression networks through pan-transcriptomic approaches. Our study not only provides a comprehensive resource of predicted protein-protein interaction networks for all 26 maize genomes but also offers a means to predict protein functions and prioritize gene candidates through the analysis of interactome clusters. To facilitate exploration of candidate gene lists associated with the pan-, and core-interactome clusters, we developed a user-friendly Python Dash web-application. Our study will facilitate the understanding of genotype-phenotype relationship and help breeding efforts to develop plants with desired traits.

Technical Abstract: Zea mays (maize), an essential crop with significant agricultural importance, has been the subject of extensive research, resulting in a wealth of publicly available data. Recent advancements in genome assemblies and annotations for the 26 maize nested association panel (NAM) inbreds have enabled large-scale pan-genomic comparative studies. These studies have expanded our understanding of agronomically important traits by identifying additional loci using the NAM mapping population and exploring trait-specific co-expression networks through pan-transcriptomic approaches. Despite the availability of high-throughput next-generation sequencing-based data, obtaining reliable protein-protein interaction (PPI) data has remained a challenge due to its high cost and complexity. As a solution, we generated predicted PPI networks for each of the 26 maize NAM inbreds using the established STRING database. These NAM-interactomes were then integrated to generate the core- and pan-interactomes. We employed the PPI clustering algorithm, ClusterONE, to identify numerous putative clusters of functionally associated proteins within the NAM-, core-, and pan-interactomes. Gene ontology (GO) enrichment analysis was used to annotate the generated clusters, demonstrating a diverse range of enriched GO terms across different clusters. Additional supporting information was generated by integrating the cluster PPI with gene co-expression analysis network. We show that the functionally annotated PPI clusters are a useful framework for protein function inference and prioritization of candidate genes associated with agronomically important traits. In summary, our study not only provides a comprehensive resource of predicted PPIs for 26 maize genomes but also offers a means to predict protein functions and prioritize gene candidates through the analysis of interactome clusters.