Genome Biology最新文献

Background: The regulation of gene expression in plants is governed by complex interactions between cis-regulatory elements and epigenetic modifications such as histone marks. While deep learning models have achieved success in predicting regulatory features from DNA sequence, their cross-species generalizability in plants remains largely unexplored.

Results: We systematically evaluate the ability of deep learning models to predict histone modifications across plant species using a multi-stage framework based on the Sei architecture. We train species-specific models for Arabidopsis (Arabidopsis thaliana), rice (Oryza sativa), and maize (Zea mays), achieving high within-species predictive performance and strong agreement between predictions and experimental ChIP-seq profiles. However, cross-species predictions show reduced performance with increasing phylogenetic distance, highlighting limited model transferability between monocots and dicots. To improve generalization, we construct a Poaceae family-level model by jointly training on rice and maize, and an Arabidopsis-trained model based solely on Arabidopsis. These models demonstrate robust predictive power in completely unprofiled species that are not used in training set, highlighting the model's adaptability to novel plant genomes based solely on conserved regulatory syntax. In contrast, cross-family models produce less consistent results, with reliable performance only in species sharing conserved regulatory features. We also develop an easy-to-use pipeline that predicts genome-wide chromatin signals directly from DNA sequences.

Conclusions: Our findings demonstrate that phylogenetically informed model training significantly improves cross-species epigenomic prediction, offering a scalable computational strategy for functional annotation in non-model and agriculturally important plants.

Background: Root rot disease caused by fungal pathogens of wine grapevines poses a serious threat to their growth and results in a substantial economic impact on grape industry. The rhizosphere microbiome recruited to plants is critical for mitigating soil-borne pathogens. However, how beneficial microbes influence disease resistance remains unclear.

Results: We investigate the composition and gene functions of microorganisms in wine grapevines with root rot disease and healthy controls by amplicon and metagenomic sequencing. We use culturomics and in vivo experiments to verify the pathogen and beneficial strains to improve plant health. We find that root rot disease in grapevines significantly affects rhizosphere microbiome diversity and composition. The microbial interkingdom network indicates that the disease destabilizes the bacteria-fungi co-occurrence network. We find that plants recruit the potentially beneficial bacteria Pseudomonas, Bacillus and Streptomyces in healthy rhizosphere soil. By culturomics, we confirm that Fusarium solani is the main pathogen causing root rot disease. We further observe that these three key beneficial bacteria from the co-occurrence networks enhance the resistance of grapevines to pathogens. Furthermore, metagenomic analysis reveals that beneficial bacterial strains suppress pathogens by enriching potential functional genes in pathways involved in disease resistance.

Conclusions: Our findings highlight the critical role of disease resistance pathways of potentially beneficial microorganisms in fighting disease and supporting plant health, offering new insight for the exploration of beneficial microbial resources and providing a basis for the development of biological control of grape root rot disease.

There is an urgent need to increase sustainable crop production. The application of molecular marker technologies such as genomic selection and machine learning based approaches are aiding accelerated crop improvement. Conventional molecular marker technologies use single nucleotide polymorphisms to predict traits, however these do not capture local epistasis and can be challenging for machine learning applications. With the growth of genome sequence data, it is possible to define haplotypes that can account for local epistatic effects and are more suitable for machine learning models. This review discusses the different methods for defining haplotype blocks and their application in plant breeding.

Background: Centromere function is fundamental and conserved across eukaryotes, despite highly divergent DNA sequences, even among closely related species. These regions often contain rapidly evolving repeats and retrotransposons, yet play a crucial role in chromosome segregation. Soybean, which harbors two distinct types of centromeric satellite repeats, is an ideal model for studying centromeric repeat organization and function.

Results: Here we generate the complete map of centromeric satellite repeats revealing the organizational patterns of different types of centromeric satellite repeats within centromeres. These maps are constructed using three recently available telomere-to-telomere soybean genomes. We find that certain centromeric satellite repeats exhibit chromosome-specific evolutionary trajectories and may serve distinct functional roles in centromere activity. We further analyze the potential relationship between centromere-specific histones H3 (CENH3) and centromeric satellite repeats, identifying consensus motifs associated with CENH3-binding sites. We also analyze the higher-order tandem repeats of the centromere and propose a hypothetical model of centromeric DNA replication.

Conclusions: We conclude that CentGm-1 and CentGm-4 evolve independently. The observation that completely identical CentGm-4 sequences consistently appear on the same chromosome across different soybean varieties indicates a stronger chromosome-specific preference for CentGm-4. We propose a model in which replication templates within the centromere region originate from multiple CENH3-nucleosome complexes bound to CentGm sequences. Both CentGm-1 and CentGm-4 contain similar motifs with the potential to bind CENH3 protein. The findings provide a new insight into the mechanisms behind centromere diversity and dynamics.

Understanding human blood metabolites is essential for deciphering systemic physiology and disease mechanisms, yet remains challenging due to diverse origins and dynamic regulation. In this study, we develop HUBMet ( https://hubmet.app.bio-it.tech/home ), an open-access web server that includes 3,950 metabolites and 129,814 metabolite-protein associations, with four analytical modules: Over-Representation Analysis (ORA) for enrichment analysis; Metabolite Set Enrichment Analysis (MSEA) for quantitative data analysis; Tissue Specificity Analysis (TSA) for assessing metabolite-tissue relevance; Metabolite-Protein Network Analysis (MPNet) for identifying key metabolite-protein associations and functional modules. HUBMet's utility is demonstrated through a COVID-19 case study revealing metabolic signatures associated with disease severity.

Background: Our understanding of gene function is often driven by its expression and, a fortiori, based on its RNA abundance in a cell, a tissue, or an organ. This assumption often neglects the limited correlation between RNA and protein abundance, largely due to post-transcriptional and pre-translational regulation. Among these regulatory processes, the spatial distribution of RNA molecules within cells has been reported as a major contributor of cellular function in microbial and animal systems. However, our understanding of the differential cellular distribution of transcripts in plants is very limited.

Results: In this manuscript, using Molecular Cartography™ and Xenium, two high-resolution and sensitive spatial transcriptomic technologies, we comprehensively analyze the differential mapping of millions of plant transcripts in the nuclear and cytoplasmic compartments of various soybean nodule cell types. Our analysis reveals distinct distributions of transcripts between the nuclear and the cytoplasmic compartments of the soybean nodule cell. We also detect variability in cytoplasmic distribution among transcripts encoded by different genes and across cell types.

Conclusions: Our findings reveal the strong diversity in the spatial distribution of transcripts in and between differentiated plant cells. It suggests that transcript localization serves as an additional regulatory layer beyond transcriptional control. By modulating nuclear export and cytoplasmic positioning, plant cells may fine-tune translational efficiency and gene function. This study underscores the importance of incorporating spatial information into transcriptomic analyses and provides new insights into the regulatory architecture of plant RNA biology.

Mutational signatures provide key insights into cancer mutational processes, but the availability of signature catalogues generated by different groups using distinct methodologies underscores a need for standardization. We introduce a Bayesian framework that offers a systematic approach to expanding existing signature catalogues for any type of mutational signature while grouping patients based on shared signature patterns. We demonstrate that this approach can identify both known and novel molecular subtypes across nearly 8000 samples spanning six cancer types and show that stratifications derived from signature yield prognostic groups, further enhancing the translational potential of mutational signatures.

Background: Selective DNA elimination occurs across diverse species and plays a crucial role in evolution and development. This process encompasses small deletions, complete removal of chromosomes, or even the elimination of entire parental genomes. Despite its importance, the molecular mechanisms governing selective DNA elimination remain poorly understood. Our study focuses on the tissue-specific elimination of Sorghum purpureosericeum B chromosomes during embryo development.

Results: In situ B chromosome visualisation, complemented by transcriptomic profiling and gene-enrichment analysis, allows us to identify 28 candidate genes potentially linked to chromosome elimination. We show that elimination is a developmentally programmed process, peaking during mid-embryogenesis and nearly completed at later stages, leaving B chromosomes only in restricted meristematic regions. Genome sequencing reveals that the sorghum B chromosome is of multi-A chromosomal origin, has reduced gene density, is enriched in repetitive sequences, and carries a novel centromeric repeat, SpuCL166. Transcriptome analyses identify B-specific variants of kinetochore, cohesion, and checkpoint genes that are expressed during active elimination, while structural modeling of CENH3 and CENP-C indicates functional divergence at the kinetochore interface.

Conclusions: Here, we provide the first comprehensive genomic and transcriptomic characterization of B chromosome and its elimination in Sorghum purpureosericeum. Our findings suggest that B chromosomes express modified mitotic machinery to control their own fate. By establishing a framework of candidate genes, this study opens new avenues for dissecting the molecular mechanisms of chromosome elimination and provides a critical foundation for understanding how genomes evolve to regulate and tolerate supernumerary chromosomal elements.

Background: The fields of ancient DNA research and forensic genetics share both methodological similarities and common challenges, particularly in the analysis of degraded DNA. Leveraging these overlaps, this study evaluates three single nucleotide polymorphisms (SNP)-based genotyping assays for analyzing challenging forensic samples: the FORCE-QIAseq SNP panel, the Twist ancient DNA hybridization capture panel, and whole-genome sequencing.

Results: We analyze twenty skeletal bone and tooth samples from authentic missing person cases, where almost all samples are severely degraded and contain exceptionally low amounts of endogenous DNA, reflected by both reduced quantifiable DNA concentrations and lower proportions of human DNA reads than typically obtained from high-quality forensic samples. Despite these challenging sample characteristics, both the FORCE and Twist assays successfully generate a substantial number of genotypes across many samples, while whole-genome sequencing yields fewer SNP calls. However, techniques like probabilistic genotyping, increase sequencing depth or genotype imputation can further enhance the utility of WGS for forensic use.

Conclusions: This study highlights the effectiveness of incorporating ancient DNA methods into forensic genetics for the analysis of degraded samples. The findings are broadly applicable to both forensic and ancient DNA research disciplines, offering valuable insights into assay selection based on sample condition and investigative goals.

Determining tumor-specificity of MHC-bound peptides is crucial for cancer immunotherapy development, yet current methods struggle with class II peptides and non-reference sequences. We introduce PepQueryMHC, an ultra-fast tool that integrates MHC-bound peptide sequences with translated RNA-seq reads for efficient tumor antigen prioritization. We demonstrate its versatility in prioritizing class I and II tumor antigens, mapping the cellular origins of presented peptides, and resolving uncertainties surrounding the prevalence of proteasome-spliced peptides.