Exploring the oasis of plant genomes: From genetic variations to function and beyond

For decades, plant geneticists have been striving to elucidate the biological mechanisms by which genetic variation (GV) influences complex traits. Plant breeders, in turn, create, select, and combine desirable GVs based on a full evaluation of GVs' effects. However, deciphering the functional consequences of GVs within the vast plant genome is akin to searching for an oasis in the desert-a daunting task characterized by overwhelming complexity. Here, we review recent advances in addressing the variation-to-function (V2F) problem and introduce the Plant V2F database (PV2F; http://rgmi.hzau.edu.cn/pv2f). Leveraging the emergence of large-scale genomic datasets and advances in modern artificial intelligence, we summarize a suite of new methods and applications involving molecular quantitative trait locus studies, causal gene prioritization, variant effect prediction, sequence-to-function models, and protein design, all of which are poised to uncover the mechanistic relationships across layers of cellular biology. Looking ahead, we outline future directions in plant breeding, emphasizing that design breeding and functional allele selection will pave the way for smart plant breeding, thereby addressing critical challenges in food security and climate change.