MaizeNet: a co-functional network for network-assisted systems genetics in Zea mays

Recent advances in exploring transcriptional regulatory landscape of crops

Crop breeding entails developing and selecting plant varieties with improved agronomic traits. Modern molecular techniques, such as genome editing, enable more efficient manipulation of plant phenotype by altering the expression of particular regulatory or functional genes. Hence, it is essential to thoroughly comprehend the transcriptional regulatory mechanisms that underpin these traits. In the multi-omics era, a large amount of omics data has been generated for diverse crop species, including genomics, epigenomics, transcriptomics, proteomics, and single-cell omics. The abundant data resources and the emergence of advanced computational tools offer unprecedented opportunities for obtaining a holistic view and profound understanding of the regulatory processes linked to desirable traits. This review focuses on integrated network approaches that utilize multi-omics data to investigate gene expression regulation. Various types of regulatory networks and their inference methods are discussed, focusing on recent advancements in crop plants. The integration of multi-omics data has been proven to be crucial for the construction of high-confidence regulatory networks. With the refinement of these methodologies, they will significantly enhance crop breeding efforts and contribute to global food security.

Prioritizing Metabolic Gene Regulators through Multi-Omic Network Integration in Maize

Elucidating gene regulatory networks (GRNs) is a major area of study within plant systems biology. Phenotypic traits are intricately linked to specific gene expression profiles. These expression patterns arise primarily from regulatory connections between sets of transcription factors (TFs) and their target genes. In this study, we integrated publicly available co-expression networks derived from more than 6,000 RNA-seq samples, 283 protein-DNA interaction assays, and 16 million of SNPs used to identify expression quantitative loci (eQTL), to construct TF-target networks. In total, we analyzed ~4.6M interactions to generate four distinct types of TF-target networks: co-expression, protein-DNA interaction (PDI), trans-expression quantitative loci (trans-eQTL), and cis-eQTL combined with PDIs. To improve the functional annotation of TFs based on its target genes, we implemented three different strategies to integrate these four types of networks. We subsequently evaluated the effectiveness of our method through loss-of function mutant and random networks. The multi-network integration allowed us to identify transcriptional regulators of hormone-, metabolic- and development-related processes. Finally, using the topological properties of the fully integrated network, we identified potentially functional redundant TF paralogs. Our findings retrieved functions previously documented for numerous TFs and revealed novel functions that are crucial for informing the design of future experiments. The approach here-described lays the foundation for the integration of multi-omic datasets in maize and other plant systems.

Genome-Wide Identification and Functional Analysis of Nitrate Transporter Genes (NPF, NRT2 and NRT3) in Maize

Nitrate is the primary form of nitrogen uptake in plants, mainly transported by nitrate transporters (NRTs), including NPF (NITRATE TRANSPORTER 1/PEPTIDE TRANSPORTER FAMILY), NRT2 and NRT3. In this study, we identified a total of 78 NPF, seven NRT2, and two NRT3 genes in maize. Phylogenetic analysis divided the NPF family into eight subgroups (NPF1-NPF8), consistent with the results in Arabidopsis thaliana and rice. The NRT2 family appears to have evolved more conservatively than the NPF family, as NRT2 genes contain fewer introns. The promoters of all NRTs are rich in cis-acting elements responding to biotic and abiotic stresses. The expression of NRTs varies in different tissues and developmental stages, with some NRTs only expressed in specific tissues or developmental stages. RNA-seq analysis using Xu178 revealed differential expression of NRTs in response to nitrogen starvation and nitrate resupply. Moreover, the expression patterns of six key NRTs genes (NPF6.6, NPF6.8, NRT2.1, NRT2.5 and NRT3.1A/B) varied in response to alterations in nitrogen levels across distinct maize inbred lines with different nitrogen uptake rates. This work enhances our understanding of the structure and expression of NRTs genes, and their roles in nitrate response, paving the way for improving maize nitrogen efficiency through molecular breeding.

Multi-trait and multi-environment genomic prediction for flowering traits in maize: a deep learning approach

Maize (Zea mays L.), the third most widely cultivated cereal crop in the world, plays a critical role in global food security. To improve the efficiency of selecting superior genotypes in breeding programs, researchers have aimed to identify key genomic regions that impact agronomic traits. In this study, the performance of multi-trait, multi-environment deep learning models was compared to that of Bayesian models (Markov Chain Monte Carlo generalized linear mixed models (MCMCglmm), Bayesian Genomic Genotype-Environment Interaction (BGGE), and Bayesian Multi-Trait and Multi-Environment (BMTME)) in terms of the prediction accuracy of flowering-related traits (Anthesis-Silking Interval: ASI, Female Flowering: FF, and Male Flowering: MF). A tropical maize panel of 258 inbred lines from Brazil was evaluated in three sites (Cambira-2018, Sabaudia-2018, and Iguatemi-2020 and 2021) using approximately 290,000 single nucleotide polymorphisms (SNPs). The results demonstrated a 14.4% increase in prediction accuracy when employing multi-trait models compared to the use of a single trait in a single environment approach. The accuracy of predictions also improved by 6.4% when using a single trait in a multi-environment scheme compared to using multi-trait analysis. Additionally, deep learning models consistently outperformed Bayesian models in both single and multiple trait and environment approaches. A complementary genome-wide association study identified associations with 26 candidate genes related to flowering time traits, and 31 marker-trait associations were identified, accounting for 37%, 37%, and 22% of the phenotypic variation of ASI, FF and MF, respectively. In conclusion, our findings suggest that deep learning models have the potential to significantly improve the accuracy of predictions, regardless of the approach used and provide support for the efficacy of this method in genomic selection for flowering-related traits in tropical maize.

Identification of growth regulators using cross-species network analysis in plants

With the need to increase plant productivity, one of the challenges plant scientists are facing is to identify genes that play a role in beneficial plant traits. Moreover, even when such genes are found, it is generally not trivial to transfer this knowledge about gene function across species to identify functional orthologs. Here, we focused on the leaf to study plant growth. First, we built leaf growth transcriptional networks in Arabidopsis (Arabidopsis thaliana), maize (Zea mays), and aspen (Populus tremula). Next, known growth regulators, here defined as genes that when mutated or ectopically expressed alter plant growth, together with cross-species conserved networks, were used as guides to predict novel Arabidopsis growth regulators. Using an in-depth literature screening, 34 out of 100 top predicted growth regulators were confirmed to affect leaf phenotype when mutated or overexpressed and thus represent novel potential growth regulators. Globally, these growth regulators were involved in cell cycle, plant defense responses, gibberellin, auxin, and brassinosteroid signaling. Phenotypic characterization of loss-of-function lines confirmed two predicted growth regulators to be involved in leaf growth (NPF6.4 and LATE MERISTEM IDENTITY2). In conclusion, the presented network approach offers an integrative cross-species strategy to identify genes involved in plant growth and development.

easyMF: A Web Platform for Matrix Factorization-Based Gene Discovery from Large-scale Transcriptome Data

With the development of high-throughput experimental technologies, large-scale RNA sequencing (RNA-Seq) data have been and continue to be produced, but have led to challenges in extracting relevant biological knowledge hidden in the produced high-dimensional gene expression matrices. Here, we develop easyMF ( https://github.com/cma2015/easyMF ), a web platform that can facilitate functional gene discovery from large-scale transcriptome data using matrix factorization (MF) algorithms. Compared with existing MF-based software packages, easyMF exhibits several promising features, such as greater functionality, flexibility and ease of use. The easyMF platform is equipped using the Big-Data-supported Galaxy system with user-friendly graphic user interfaces, allowing users with little programming experience to streamline transcriptome analysis from raw reads to gene expression, carry out multiple-scenario MF analysis, and perform multiple-way MF-based gene discovery. easyMF is also powered with the advanced packing technology to enhance ease of use under different operating systems and computational environments. We illustrated the application of easyMF for seed gene discovery from temporal, spatial, and integrated RNA-Seq datasets of maize (Zea mays L.), resulting in the identification of 3,167 seed stage-specific, 1,849 seed compartment-specific, and 774 seed-specific genes, respectively. The present results also indicated that easyMF can prioritize seed-related genes with superior prediction performance over the state-of-art network-based gene prioritization system MaizeNet. As a modular, containerized and open-source platform, easyMF can be further customized to satisfy users' specific demands of functional gene discovery and deployed as a web service for broad applications.

NetREx: Network-based Rice Expression Analysis Server for abiotic stress conditions

Recent focus on transcriptomic studies in food crops like rice, wheat and maize provide new opportunities to address issues related to agriculture and climate change. Re-analysis of such data available in public domain supplemented with annotations across molecular hierarchy can be of immense help to the plant research community, particularly co-expression networks representing transcriptionally coordinated genes that are often part of the same biological process. With this objective, we have developed NetREx, a Network-based Rice Expression Analysis Server, that hosts ranked co-expression networks of Oryza sativa using publicly available messenger RNA sequencing data across uniform experimental conditions. It provides a range of interactable data viewers and modules for analysing user-queried genes across different stress conditions (drought, flood, cold and osmosis) and hormonal treatments (abscisic and jasmonic acid) and tissues (root and shoot). Subnetworks of user-defined genes can be queried in pre-constructed tissue-specific networks, allowing users to view the fold change, module memberships, gene annotations and analysis of their neighbourhood genes and associated pathways. The web server also allows querying of orthologous genes from Arabidopsis, wheat, maize, barley and sorghum. Here, we demonstrate that NetREx can be used to identify novel candidate genes and tissue-specific interactions under stress conditions and can aid in the analysis and understanding of complex phenotypes linked to stress response in rice.

Database URL: https://bioinf.iiit.ac.in/netrex/index.html

Integration of probabilistic functional networks without an external Gold Standard

BACKGROUND

Probabilistic functional integrated networks (PFINs) are designed to aid our understanding of cellular biology and can be used to generate testable hypotheses about protein function. PFINs are generally created by scoring the quality of interaction datasets against a Gold Standard dataset, usually chosen from a separate high-quality data source, prior to their integration. Use of an external Gold Standard has several drawbacks, including data redundancy, data loss and the need for identifier mapping, which can complicate the network build and impact on PFIN performance. Additionally, there typically are no Gold Standard data for non-model organisms.

RESULTS

We describe the development of an integration technique, ssNet, that scores and integrates both high-throughput and low-throughout data from a single source database in a consistent manner without the need for an external Gold Standard dataset. Using data from Saccharomyces cerevisiae we show that ssNet is easier and faster, overcoming the challenges of data redundancy, Gold Standard bias and ID mapping. In addition ssNet results in less loss of data and produces a more complete network.

CONCLUSIONS

The ssNet method allows PFINs to be built successfully from a single database, while producing comparable network performance to networks scored using an external Gold Standard source and with reduced data loss.

A novel probabilistic generator for large-scale gene association networks

MOTIVATION

Gene expression data provide an opportunity for reverse-engineering gene-gene associations using network inference methods. However, it is difficult to assess the performance of these methods because the true underlying network is unknown in real data. Current benchmarks address this problem by subsampling a known regulatory network to conduct simulations. But the topology of regulatory networks can vary greatly across organisms or tissues, and reference-based generators-such as GeneNetWeaver-are not designed to capture this heterogeneity. This means, for example, benchmark results from the E. coli regulatory network will not carry over to other organisms or tissues. In contrast, probabilistic generators do not require a reference network, and they have the potential to capture a rich distribution of topologies. This makes probabilistic generators an ideal approach for obtaining a robust benchmarking of network inference methods.

RESULTS

We propose a novel probabilistic network generator that (1) provides an alternative to address the inherent limitation of reference-based generators and (2) is able to create realistic gene association networks, and (3) captures the heterogeneity found across gold-standard networks better than existing generators used in practice. Eight organism-specific and 12 human tissue-specific gold-standard association networks are considered. Several measures of global topology are used to determine the similarity of generated networks to the gold-standards. Along with demonstrating the variability of network structure across organisms and tissues, we show that the commonly used "scale-free" model is insufficient for replicating these structures.

AVAILABILITY

This generator is implemented in the R package "SeqNet" and is available on CRAN (https://cran.r-project.org/web/packages/SeqNet/index.html).

Genetic Basis of Maize Resistance to Multiple Insect Pests: Integrated Genome-Wide Comparative Mapping and Candidate Gene Prioritization

Several species of herbivores feed on maize in field and storage setups, making the development of multiple insect resistance a critical breeding target. In this study, an association mapping panel of 341 tropical maize lines was evaluated in three field environments for resistance to fall armyworm (FAW), whilst bulked grains were subjected to a maize weevil (MW) bioassay and genotyped with Diversity Array Technology's single nucleotide polymorphisms (SNPs) markers. A multi-locus genome-wide association study (GWAS) revealed 62 quantitative trait nucleotides (QTNs) associated with FAW and MW resistance traits on all 10 maize chromosomes, of which, 47 and 31 were discovered at stringent Bonferroni genome-wide significance levels of 0.05 and 0.01, respectively, and located within or close to multiple insect resistance genomic regions (MIRGRs) concerning FAW, SB, and MW. Sixteen QTNs influenced multiple traits, of which, six were associated with resistance to both FAW and MW, suggesting a pleiotropic genetic control. Functional prioritization of candidate genes (CGs) located within 10-30 kb of the QTNs revealed 64 putative GWAS-based CGs (GbCGs) showing evidence of involvement in plant defense mechanisms. Only one GbCG was associated with each of the five of the six combined resistance QTNs, thus reinforcing the pleiotropy hypothesis. In addition, through in silico co-functional network inferences, an additional 107 network-based CGs (NbCGs), biologically connected to the 64 GbCGs, and differentially expressed under biotic or abiotic stress, were revealed within MIRGRs. The provided multiple insect resistance physical map should contribute to the development of combined insect resistance in maize.

ZEAMAP, a Comprehensive Database Adapted to the Maize Multi-Omics Era

As one of the most extensively cultivated crops, maize (Zea mays L.) has been extensively studied by researchers and breeders for over a century. With advances in high-throughput detection of various omics data, a wealth of multi-dimensional and multi-omics information has been accumulated for maize and its wild relative, teosinte. Integration of this information has the potential to accelerate genetic research and generate improvements in maize agronomic traits. To this end, we constructed ZEAMAP, a comprehensive database incorporating multiple reference genomes, annotations, comparative genomics, transcriptomes, open chromatin regions, chromatin interactions, high-quality genetic variants, phenotypes, metabolomics, genetic maps, genetic mapping loci, population structures, and populational DNA methylation signals within maize inbred lines. ZEAMAP is user friendly, with the ability to interactively integrate, visualize, and cross-reference multiple different omics datasets.

•

Functional annotations and comparative genomics of maize and teosinte genomes

•

Multi-omics data generated from the same maize inbred lines panel

•

Interactive tools to query, cross-refer, and visualize the omics data

BiomeNet: a database for construction and analysis of functional interaction networks for any species with a sequenced genome

Motivation

Owing to advanced DNA sequencing and genome assembly technology, the number of species with sequenced genomes is rapidly increasing. The aim of the recently launched Earth BioGenome Project is to sequence genomes of all eukaryotic species on Earth over the next 10 years, making it feasible to obtain genomic blueprints of the majority of animal and plant species by this time. Genetic models of the sequenced species will later be subject to functional annotation, and a comprehensive molecular network should facilitate functional analysis of individual genes and pathways. However, network databases are lagging behind genome sequencing projects as even the largest network database provides gene networks for less than 10% of sequenced eukaryotic genomes, and the knowledge gap between genomes and interactomes continues to widen.

Results

We present BiomeNet, a database of 95 scored networks comprising over 8 million co-functional links, which can build and analyze gene networks for any species with the sequenced genome. BiomeNet transfers functional interactions between orthologous proteins from source networks to the target species within minutes and automatically constructs gene networks with the quality comparable to that of existing networks. BiomeNet enables assembly of the first-in-species gene networks not available through other databases, which are highly predictive of diverse biological processes and can also provide network analysis by extracting subnetworks for individual biological processes and network-based gene prioritizations. These data indicate that BiomeNet could enhance the benefits of decoding the genomes of various species, thus improving our understanding of the Earth’ biodiversity.

Availability and implementation

The BiomeNet is freely available at http://kobic.re.kr/biomenet/.

Supplementary information

Supplementary data are available at Bioinformatics online.

BarleyNet: A Network-Based Functional Omics Analysis Server for Cultivated Barley, Hordeum vulgare L

Cultivated barley (Hordeum vulgare L.) is one of the most produced cereal crops worldwide after maize, bread wheat, and rice. Barley is an important crop species not only as a food source, but also in plant genetics because it harbors numerous stress response alleles in its genome that can be exploited for crop engineering. However, the functional annotation of its genome is relatively poor compared with other major crops. Moreover, bioinformatics tools for system-wide analyses of omics data from barley are not yet available. We have thus developed BarleyNet, a co-functional network of 26,145 barley genes, along with a web server for network-based predictions (http://www.inetbio.org/barleynet). We demonstrated that BarleyNet's prediction of biological processes is more accurate than that of an existing barley gene network. We implemented three complementary network-based algorithms for prioritizing genes or functional concepts to study genetic components of complex traits such as environmental stress responses: (i) a pathway-centric search for candidate genes of pathways or complex traits; (ii) a gene-centric search to infer novel functional concepts for genes; and (iii) a context-centric search for novel genes associated with stress response. We demonstrated the usefulness of these network analysis tools in the study of stress response using proteomics and transcriptomics data from barley leaves and roots upon drought or heat stresses. These results suggest that BarleyNet will facilitate our understanding of the underlying genetic components of complex traits in barley.