Deep learning methodologies have revolutionized prediction in many fields and show potential to do the same in molecular biology and genetics. However, applying these methods in their current forms ignores evolutionary dependencies within biological systems and can result in false positives and spurious conclusions. We developed two approaches that account for evolutionary relatedness in machine learning models: ( i ) gene-family–guided splitting and ( ii ) ortholog contrasts. The first approach accounts for evolution by constraining model training and testing sets to include different gene families. The second approach uses evolutionarily informed comparisons between orthologous genes to both control for and leverage evolutionary divergence during the training process. The two approaches were explored and validated within the context of mRNA expression level prediction and have the area under the ROC curve (auROC) values ranging from 0.75 to 0.94. Model weight inspections showed biologically interpretable patterns, resulting in the hypothesis that the 3′ UTR is more important for fine-tuning mRNA abundance levels while the 5′ UTR is more important for large-scale changes.

Plants have evolved a repertoire of strategies collectively termed the shade-avoidance syndrome to avoid shade from canopy and compete for light with their neighbors. However, the signaling mechanism governing the adaptive changes of adult plant architecture to shade is not well understood. Here, we show that in Arabidopsis, compared with the wild type, several PHYTOCHROME-INTERACTING FACTORS (PIFS) overexpressors all display constitutive shade-avoidance syndrome under normal high red to far-red light ratio conditions but are less sensitive to the simulated shade, whereas the MIR156 overexpressors exhibit an opposite phenotype. The simulated shade induces rapid accumulation of PIF proteins, reduced expression of multiple MIR156 genes, and concomitant elevated expression of the SQUAMOSA-PROMOTER BINDING PROTEIN-LIKE (SPL) family genes. Moreover, in vivo and in vitro assays indicate that PIFs bind to the promoters of several MIR156 genes directly and repress their expression. Our results establish a direct functional link between the phytochrome-PIFs and miR156-SPL regulatory modules in mediating shade-avoidance syndrome.

Our era has witnessed tremendous advances in plant genomics, characterized by an explosion of high-throughput techniques to identify multi-dimensional genome-wide molecular phenotypes at low costs. More importantly, genomics is not merely acquiring molecular phenotypes, but also leveraging powerful data mining tools to predict and explain them. In recent years, deep learning has been found extremely effective in these tasks. This review highlights two prominent questions at the intersection of genomics and deep learning: 1) how can the flow of information from genomic DNA sequences to molecular phenotypes be modeled; 2) how can we identify functional variants in natural populations using deep learning models? Additionally, we discuss the possibility of unleashing the power of deep learning in synthetic biology to create novel genomic elements with desirable functions. Taken together, we propose a central role of deep learning in future plant genomics research and crop genetic improvement.