Dataset Information

DeepWheat: Predicting the Effects of Genomic Variants on Gene Expression and Regulatory Activities Across Tissues and Varieties in Wheat Using Deep Learning [ChIP-Seq]

ABSTRACT: Accurate prediction of genomic variant effects and gene expression is essential for identifying functional variations and enabling precise genome editing of cis-regulatory elements (CREs). Spatiotemporal gene expression patterns are fundamental to the formation of key traits, yet tissue-specific predictions remain inaccurate, particularly in large-genome crops like wheat. In this study, we developed DeepWheat, a suite of two models for predicting epigenomic features and gene expression in wheat. DeepEXP, a deep learning model, integrates epigenomic and transcriptomic data across various wheat tissues, achieving Pearson correlation coefficients (PCC) over 0.8 and outperforming sequence-only models, especially for tissue-specific genes. DeepEPI predicts epigenomic features from DNA sequences, helping identify regulatory sequences and facilitating model transfer across wheat varieties. Using chromatin accessibility and transcriptomic data from 9 additional wheat varieties, we validated the model’s accuracy and transfer efficiency. Our analysis further revealed that indels have a greater impact on gene expression than SNPs, and that, compared to promoter regions, the 5’UTR, 3’UTR, and introns exert even stronger regulatory effects on gene expression. These models also identified mutations that alter gene expression, supporting precise CRE editing. They provide valuable tools for tissue-specific predictions, regulatory sequence identification, and saturation mutagenesis to pinpoint high-effect sites.

ORGANISM(S): Triticum aestivum

PROVIDER: GSE289179 | GEO | 2025/09/05

REPOSITORIES: GEO

ACCESS DATA

Dataset's files

Source:

			Action	DRS
		Other

Items per page:

1 - 1 of 1

Similar Datasets

Project description:Introduction: Pre-harvest Sprouting (PHS) seriously affects wheat quality and yield. However, to date there have been limited reports. It is of great urgency to breed resistance varieties via quantitative trait nucleotides (QTNs) or genes for PHS resistance in white-grained wheat. Methods: 629 Chinese wheat varieties, including 373 local wheat varieties from 70 years ago and 256 improved wheat varieties were phenotyped for spike sprouting (SS) in two environments and genotyped by wheat 660K microarray. These phenotypes were used to associate with 314,548 SNP markers for identifying QTNs for PHS resistance using several multi-locus genome-wide association study (GWAS) methods. Their candidate genes were verified by RNA-seq, and the validated candidate genes were further exploited in wheat breeding. Results: As a result, variation coefficients of 50% and 47% for PHS in 629 wheat varieties, respectively, in 2020-2021 and 2021-2022 indicated large phenotypic variation, in particular, 38 white grain varieties appeared at least medium resistance, such as Baipimai, Fengchan 3, and Jimai 20. In GWAS, 22 significant QTNs, with the sizes of 0.06% ~ 38.11%, for PHS resistance were stably identified by multiple multi-locus methods in two environments, e.g., AX-95124645 (chr3D:571.35Mb), with the sizes of 36.390% and 45.850% in 2020-2021 and 2021-2022, respectively, was detected by several multi-locus methods in two environments. As compared with previous studies, the AX-95124645 was used to develop Kompetitive Allele-Specific PCR marker QSS.TAF9-3D (chr3D:569.17Mb~573.55Mb) for the first time, especially, it is available in white-grain wheat varieties. Around this locus, nine genes were significantly differentially expressed, and two of them (TraesCS3D01G466100 and TraesCS3D01G468500) were found by GO annotation to be related to PHS resistance and determined as candidate genes. Discussion: The QTN and two new candidate genes related to PHS resistance were identified in this study. The QTN can be used to effectively identify the PHS resistance materials, especially, all the white-grained varieties with QSS.TAF9-3D-TT haplotype are resistant to spike sprouting. Thus, this study provides candidate genes, materials, and methodological basis for breeding wheat PHS resistance in the future.

Project description:Background: Waterlogging was one of the most serious abiotic stresses in wheat-growing regions of China. There were great differences in waterlogging tolerance among different wheat varieties, and the mechanism of waterlogging tolerance of wheat seeds during germination was unclear. Methods: In order to reveal the adaptability of wheat to waterlogging stress during germination, we analyzed the germination rate and anatomical structure of three wheat seeds, ‘Zhoumai 22’, ‘Bainong 207’ and ‘Bainong 607’. At the same time, Illumina sequencing technology was used to determine the transcriptome of these three wheat varieties during germination. Results: The results showed that there was no significant difference between the germination rate of ‘Bainong 207’ after 3 days of waterlogging treatment and that of the control seeds. However, under waterlogging stress, the degree of emulsification and degradation of endosperm cells was higher than that of the control treatment, and starch granules in endosperm were significantly reduced. Transcriptome data were obtained from seed samples (a total of 18 samples) of three wheat varieties under waterlogging and control treatment. A total of 2,775 differentially expressed genes (DEGs) were identified by comprehensive analysis. In addition, by analyzing the correlation between the expression levels of DEGs and seed germination rates in three wheat varieties under waterlogging stress, it was found that the relative expression levels of 563 and 398 genes were positively and negatively correlated with the germination rate of wheat seeds, respectively. The GO and KEGG analysis found that the difference in waterlogging tolerance of the three wheat varieties was related to the abundance of key genes involved in the glycolysis pathway, the starch and sucrose metabolism pathway, and the lactose metabolism pathway. The ethanol dehydrogenase (ADH) gene in the endosperm of ‘Bainong 607’ was immediately induced after a short period of waterlogging, and the energy provided by glycolysis pathway enabled the seeds of ‘Bainong 607’ to germinate as early as possible, while the expression level of AP2/ERF transcription factor was up-regulated to further enhance its waterlogging tolerance. Conclusions: In general, this study provided a deeper understanding of the mechanisms by which different wheat varieties respond to waterlogging stress during germination.

Dataset Information

DeepWheat: Predicting the Effects of Genomic Variants on Gene Expression and Regulatory Activities Across Tissues and Varieties in Wheat Using Deep Learning [ChIP-Seq]

Dataset's files

Similar Datasets

OmicsDI is part of the ELIXIR infrastructure

Tweets