The transcription regulatory network inside a eukaryotic cell is defined by the combinatorial actions of transcription factors (TFs). However, TF binding studies in plants are too few in number to produce a general picture of this complex network. In this study, we use large-scale ChIP-seq to reconstruct it in the maize leaf, and train machine-learning models to predict TF binding and co-localization. The resulting network covers 77% of the expressed genes, and shows a scale-free topology and functional modularity like a real-world network. TF binding sequence preferences are conserved within family, while co-binding could be key for their binding specificity. Cross-species comparison shows that core network nodes at the top of the transmission of information being more conserved than those at the bottom. This study reveals the complex and redundant nature of the plant transcription regulatory network, and sheds light on its architecture, organizing principle and evolutionary trajectory.

Significance: The same gene is often regulated differently in response to stress in even closely related plant species. Directly measuring stress-responsive gene expression can be financially and logistically challenging in nonmodel species. Here, we show that models trained using data on which genes respond to cold in one species can predict which genes will respond to cold in related species, even when the training and target species vary in their degree of tolerance to cold. The prediction models we used require only genomic sequence and gene models. As a result, data from well-studied model species may be used to predict which genes will respond to stress in less-studied species with sequenced genomes.

The complete assembly of vast and complex plant genomes, like the hexaploid wheat genome, remains challenging. Here we present CS-IAAS, a comprehensive telomere-to-telomere (T2T) gap-free Triticum aestivum L. genome, encompassing 14.51 billion base pairs and featuring all 21 centromeres and 42 telomeres. Annotation revealed 90.8 Mb additional centromeric satellite arrays and 5,611 rDNA units. Genome-wide rearrangements, centromeric elements, transposable element expansion and segmental duplications were deciphered during tetraploidization and hexaploidization, providing a comprehensive understanding of wheat subgenome evolution. Among them, transposable element insertions during hexaploidization greatly influenced gene expression balances, thus increasing the genome plasticity of transcriptional levels. Additionally, we generated 163,329 full-length cDNA sequences and proteomic data that helped annotate 141,035 high-confidence protein-coding genes. The complete T2T reference genome (CS-IAAS), along with its transcriptome and proteome, represents a significant step in our understanding of wheat genome complexity and provides insights for future wheat research and breeding.

The morphological development of the leaf greatly influences plant architecture and crop yields. The maize leaf is composed of a leaf blade, ligule and sheath. Although extensive transcriptional profiling of the tissues along the longitudinal axis of the developing maize leaf blade has been conducted, little is known about the transcriptional dynamics in sheath tissues, which play important roles in supporting the leaf blade. Using a comprehensive transcriptome dataset, we demonstrated that the leaf sheath transcriptome dynamically changes during maturation, with the construction of basic cellular structures at the earliest stages of sheath maturation with a transition to cell wall biosynthesis and modifications. The transcriptome again changes with photosynthesis and lignin biosynthesis at the last stage of sheath tissue maturation. The different tissues of the maize leaf are highly specialized in their biological functions and we identified 15 genes expressed at significantly higher levels in the leaf sheath compared with their expression in the leaf blade, including the BOP2 homologs GRMZM2G026556 and GRMZM2G022606, DOGT1 (GRMZM2G403740) and transcription factors from the B3 domain, C2H2 zinc finger and homeobox gene families, implicating these genes in sheath maturation and organ specialization.

Advances in genome sequencing and annotation have eased the difficulty of identifying new gene sequences. Predicting the functions of these newly identified genes remains challenging. Genes descended from a common ancestral sequence are likely to have common functions. As a result, homology is widely used for gene function prediction. This means functional annotation errors also propagate from one species to another. Several approaches based on machine learning classification algorithms were evaluated for their ability to accurately predict gene function from non‐homology gene features. Among the eight supervised classification algorithms evaluated, random‐forest‐based prediction consistently provided the most accurate gene function prediction. Non‐homology‐based functional annotation provides complementary strengths to homology‐based annotation, with higher average performance in Biological Process GO terms, the domain where homology‐based functional annotation performs the worst, and weaker performance in Molecular Function GO terms, the domain where the accuracy of homology‐based functional annotation is highest. GO prediction models trained with homology‐based annotations were able to successfully predict annotations from a manually curated “gold standard” GO annotation set. Non‐homology‐based functional annotation based on machine learning may ultimately prove useful both as a method to assign predicted functions to orphan genes which lack functionally characterized homologs, and to identify and correct functional annotation errors which were propagated through homology‐based functional annotations.