Recent advances in sequencing and informatic technologies have led to a deluge of publicly available genomic data. While it is now relatively easy to sequence, assemble, and identify genic regions in diploid plant genomes, functional annotation of these genes is still a challenge. Over the past decade, there has been a steady increase in studies utilizing machine learning algorithms for various aspects of functional prediction, because these algorithms are able to integrate large amounts of heterogeneous data and detect patterns inconspicuous through rule‐based approaches. The goal of this review is to introduce experimental plant biologists to machine learning, by describing how it is currently being used in gene function prediction to gain novel biological insights. In this review, we discuss specific applications of machine learning in identifying structural features in sequenced genomes, predicting interactions between different cellular components, and predicting gene function and organismal phenotypes. Finally, we also propose strategies for stimulating functional discovery using machine learning–based approaches in plants.

Plants’ ability to chemically modify core structures of specialized metabolites is the main reason why the plant kingdom contains such a wide and rich array of diverse compounds. One of the most important types of chemical modifications of small molecules is the addition of an acyl moiety to produce esters and amides. Large-scale phylogenomics analyses have shown that the enzymes that perform acyl transfer reactions on the myriad small molecules synthesized by plants belong to only a few gene families. This review is focused on describing the biochemistry, evolutionary origins, and chemical ecology implications of one of these families—the BAHD acyltransferases. The growth of advanced metabolomic studies coupled with next-generation sequencing of diverse plant species has confirmed that the BAHD family plays critical roles in modifying nearly all known classes of specialized metabolites. The current and future outlook for research on BAHDs includes expanding their roles in synthetic biology and metabolic engineering.

Pyrrolizidine alkaloids (PAs) are toxic secondary metabolites that are found in several distantly related families of the angiosperms. The first specific step in PA biosynthesis is catalyzed by homospermidine synthase (HSS), which has been recruited several times independently by duplication of the gene encoding deoxyhypusine synthase, an enzyme involved in the posttranslational activation of the eukaryotic initiation factor 5A. HSS shows highly diverse spatiotemporal gene expression in various PA-producing species. In comfrey (Symphytum officinale; Boraginaceae), PAs are reported to be synthesized in the roots, with HSS being localized in cells of the root endodermis. Here, we show that comfrey plants activate a second site of HSS expression when inflorescences start to develop. HSS has been localized in the bundle sheath cells of specific leaves. Tracer feeding experiments have confirmed that these young leaves express not only HSS but the whole PA biosynthetic route. This second site of PA biosynthesis results in drastically increased PA levels within the inflorescences. The boost of PA biosynthesis is proposed to guarantee optimal protection especially of the reproductive structures.

Systematic mappings of protein interactome networks have provided invaluable functional information for numerous model organisms. Here we develop PCR-mediated Linkage of barcoded Adapters To nucleic acid Elements for sequencing (PLATE-seq) that serves as a general tool to rapidly sequence thousands of DNA elements. We validate its utility by generating the ORFeome for Oryza sativa covering 2,300 genes and constructing a high-quality protein–protein interactome map consisting of 322 interactions between 289 proteins, expanding the known interactions in rice by roughly 50%. Our work paves the way for high-throughput profiling of protein–protein interactions in a wide range of organisms. Significance Although recent advances in next-generation sequencing have facilitated the construction of whole genomes from hundreds of organisms, considerable barriers still restrict the functional understanding of the genes that they contain. An initial prerequisite for such an understanding is the availability of high-quality gene libraries (ORFeomes) amenable to high-throughput functional experiments. Here we develop a massively parallel next-generation sequencing method, PLATE-seq, and leverage the method to construct an ORFeome for rice, providing a toolkit for systematic functional study in an agricultural species. To demonstrate the utility of these resources, we present a map of rice protein–protein interactions.