Plant metabolism underpins many traits of ecological and agronomic importance. Plants produce numerous compounds to cope with their environments but the biosynthetic pathways for most of these compounds have not yet been elucidated. To engineer and improve metabolic traits, we need comprehensive and accurate knowledge of the organization and regulation of plant metabolism at the genome scale. Here, we present a computational pipeline to identify metabolic enzymes, pathways, and gene clusters from a sequenced genome. Using this pipeline, we generated metabolic pathway databases for 22 species and identified metabolic gene clusters from 18 species. This unified resource can be used to conduct a wide array of comparative studies of plant metabolism. Using the resource, we discovered a widespread occurrence of metabolic gene clusters in plants: 11,969 clusters from 18 species. The prevalence of metabolic gene clusters offers an intriguing possibility of an untapped source for uncovering new metabolite biosynthesis pathways. For example, more than 1,700 clusters contain enzymes that could generate a specialized metabolite scaffold (signature enzymes) and enzymes that modify the scaffold (tailoring enzymes). In four species with sufficient gene expression data, we identified 43 highly coexpressed clusters that contain signature and tailoring enzymes, of which eight were characterized previously to be functional pathways. Finally, we identified patterns of genome organization that implicate local gene duplication and, to a lesser extent, single gene transposition as having played roles in the evolution of plant metabolic gene clusters.

Specialized metabolites (also called natural products or secondary metabolites) derived from bacteria, fungi, marine organisms and plants constitute an important source of antibiotics, anti-cancer agents, insecticides, immunosuppressants and herbicides. Many specialized metabolites in bacteria and fungi are biosynthesized via metabolic pathways whose enzymes are encoded by clustered genes on a chromosome. Metabolic gene clusters comprise a group of physically co-localized genes that together encode enzymes for the biosynthesis of a specific metabolite. Although metabolic gene clusters are generally not known to occur outside of microbes, several plant metabolic gene clusters have been discovered in recent years. The discovery of novel metabolic pathways is being enabled by the increasing availability of high-quality genome sequencing coupled with the development of powerful computational toolkits to identify metabolic gene clusters. To provide a comprehensive overview of various bioinformatics methods for detecting gene clusters, we compare and contrast key aspects of algorithmic logic behind several computational tools, including 'NP.searcher', 'ClustScan', 'CLUSEAN', 'antiSMASH', 'SMURF', 'MIDDAS-M', 'ClusterFinder', 'CASSIS/SMIPS' and 'C-Hunter' among others. We also review additional tools such as 'NRPSpredictor' and 'SBSPKS' that can infer substrate specificity for previously identified gene clusters. The continual development of bioinformatics methods to predict gene clusters will help shed light on how organisms assemble multi-step metabolic pathways for adaptation to various ecological niches.

The Summary: Plants and microbes produce numerous compounds to cope with their environments but the biosynthetic pathways for most of these compounds have yet to be elucidated. Some biosynthetic pathways are encoded by enzymes collocated in the chromosome. To facilitate a more comprehensive condition and tissue-specific expression analysis of metabolic gene clusters, we developed METACLUSTER, a probabilistic framework for characterizing metabolic gene clusters using context-specific gene expression information.

Linkage mapping is one of the most commonly used methods to identify genetic loci that determine a trait. However, the loci identified by linkage mapping may contain hundreds of candidate genes and require a time-consuming and labor-intensive fine mapping process to find the causal gene controlling the trait. With the availability of a rich assortment of genomic and functional genomic data, it is possible to develop a computational method to facilitate faster identification of causal genes. We developed QTG-Finder, a machine learning based algorithm to prioritize causal genes by ranking genes within a quantitative trait locus (QTL). Two predictive models were trained separately based on known causal genes in Arabidopsis and rice. An independent validation analysis showed that the models could recall about 64% of Arabidopsis and 79% of rice causal genes when the top 20% ranked genes were considered. The top 20% ranked genes can range from 10 to 100 genes, depending on the size of a QTL. The models can prioritize different types of traits though at different efficiency. We also identified several important features of causal genes including paralog copy number, being a transporter, being a transcription factor, and containing SNPs that cause premature stop codon. This work lays the foundation for systematically understanding characteristics of causal genes and establishes a pipeline to predict causal genes based on public data.

Author summary Plants thrive in highly heterogenous soils. How they compute a multitude of contrasting stimuli and mount an adaptive response without a centralized information processing unit is an intriguing question. For instance, below ground, roots can sense and respond to the single or multiple nutrient stresses, and adjust its growth rate accordingly. Nevertheless, the genetic architecture of root growth responses under single and combined stress remains poorly understood. To fill this gap in our understanding about such crucial phenomenon for plant survival, we explored the natural variation of root growth rate (RGR) in Arabidopsis grown under single and combined nutritional stress, including deficiencies of iron (-Fe), zinc (-Zn), phosphate and iron (-P-Fe) and phosphate and zinc (-P-Zn). Our GWAS revealed distinct genetic architectures underlying root growth responses to single or combined nutrient stresses. By integrating GWAS and coexpression networks, we identified and validated genes controlling the variation of root growth to combined nutrient-deficiency, namely VARIANT IN METHYLATION 1, FORMIN-LIKE-PROTEIN-6 and VOLTAGE-DEPENDENT ANION-SELECTIVE CHANNEL PROTEIN 3. Our findings provide a framework to accelerate future research aiming at better understanding how plants sense and respond to multiple environmental inputs, and promise to help designing new agronomical and biotechnological strategies to improve root growth.

Senna tora is a widely used medicinal plant. Its health benefits have been attributed to the large quantity of anthraquinones, but how they are made in plants remains a mystery. To identify the genes responsible for plant anthraquinone biosynthesis, we reveal the genome sequence of S. tora at the chromosome level with 526 Mb (96%) assembled into 13 chromosomes. Comparison among related plant species shows that a chalcone synthase-like (CHS-L) gene family has lineage-specifically and rapidly expanded in S. tora. Combining genomics, transcriptomics, metabolomics, and biochemistry, we identify a CHS-L gene contributing to the biosynthesis of anthraquinones. The S. tora reference genome will accelerate the discovery of biologically active anthraquinone biosynthesis pathways in medicinal plants. Anthraquinones are aromatic polyketides and have been used for treating various diseases, but the biosynthetic pathway is unclear. Here, the authors assemble the genome of an anthraquinone-producing medicinal plant Senna tora and show the evidences that CHS-like genes may be involved in anthraquinone biosynthesis.

Linkage mapping has been widely used to identify quantitative trait loci (QTL) in many plants and usually requires a time-consuming and labor-intensive fine mapping process to find the causal gene underlying the QTL. Previously, we described QTG-Finder, a machine-learning algorithm to rationally prioritize candidate causal genes in QTLs. While it showed good performance, QTG-Finder could only be used in Arabidopsis and rice because of the limited number of known causal genes in other species. Here we tested the feasibility of enabling QTG-Finder to work on species that have few or no known causal genes by using orthologs of known causal genes as the training set. The model trained with orthologs could recall about 64% of Arabidopsis and 83% of rice causal genes when the top 20% ranked genes were considered, which is similar to the performance of models trained with known causal genes. The average precision was 0.027 for Arabidopsis and 0.029 for rice. We further extended the algorithm to include polymorphisms in conserved non-coding sequences and gene presence/absence variation as additional features. Using this algorithm, QTG-Finder2, we trained and cross-validatedSorghum bicolorandSetaria viridismodels. TheS. bicolormodel was validated by causal genes curated from the literature and could recall 70% of causal genes when the top 20% ranked genes were considered. In addition, we applied theS. viridismodel and public transcriptome data to prioritize a plant height QTL and identified 13 candidate genes. QTL-Finder2 can accelerate the discovery of causal genes in any plant species and facilitate agricultural trait improvement.

In this Perspective, Diane Dickel and colleagues review recent progress and opportunities in applying single-cell sequencing and microfluidics methods to plants. The authors highlight the need for new tools developed with plants in mind, and advocate for the creation of a centralized, open-access database to house plant single-cell data.