This module provides a gene-centered integrative view of genetic variation and regulatory features in rice. It summarizes local genetic variants, functional and phenotypic annotations, chromatin accessibility, model-predicted regulatory effects of non-coding variants, and population genetic statistics within the genomic neighborhood of a queried gene. Quantitative chromatin accessibility and transcriptome profiles across multiple tissues, developmental stages, and three reference varieties are jointly presented to support functional, regulatory, and evolutionary interpretation.

Gene Expression Profiles: Quantitative gene expression levels across multiple tissues and developmental stages are shown where available. These profiles enable direct comparison between transcriptional output and local regulatory activity, facilitating interpretation of tissue-specific gene regulation.

Chromatin Accessibility Landscape: Tissue-resolved chromatin accessibility profiles are visualized across the queried genomic region using ATAC-seq data. Accessibility signals are quantified from Tn5 insertion events aggregated into fixed genomic windows of 250 bp with a sliding step of 100 bp. Heatmap-style visualization highlights spatial and tissue-specific patterns of chromatin openness, revealing putative cis-regulatory elements and their tissue specificity.

Regulatory Effect Predictions for Non-coding Variants: Allele-specific chromatin accessibility effects are predicted using Basenji (Kelley et al., 2018), a deep learning sequence-to-signal model trained to predict chromatin accessibility from genomic sequence. The signed local sequence activity log-ratio is LSAR = log₂(P_ALT + 1) − log₂(P_REF + 1), where P_ALT and P_REF are local predictions averaged across the central eight 128-bp output bins (1,024 bp total). Positive values indicate that the alternative allele is predicted to increase chromatin accessibility, whereas negative values indicate reduced accessibility relative to the reference allele. These scores enable systematic prioritization of non-coding variants with potential regulatory impact.

High-impact regulatory variants (HEVs): HEV cutoffs depend on both the reference genome and the variant set used for calibration. A symbol is shown for a tissue-specific score only when its absolute unrounded value reaches the applicable cutoff. Both predicted increases and decreases can therefore be marked.

• Nipponbare (IRGSP-1.0; vg) SNP and non-SNP variants: A light-red ★ denotes an HEV at 0.052 (exact value 0.051688551903), a cutoff calibrated from the SNP-only q90 and applied to both SNP and non-SNP scores. A red ★ denotes a strong HEV at the combined SNP and non-SNP q90 cutoff of 0.106 (exact value 0.10640232). Strong HEVs are displayed as ★ only.
• MH63RS3 (vm) SNPs: a light-red ★ denotes an HEV at the SNP-only q90 cutoff of 0.052 (exact value 0.052074253559). No strong-HEV tier is used.
• ZS97RS3 (vz) SNPs: a light-red ★ denotes an HEV at the SNP-only q90 cutoff of 0.047 (exact value 0.047439455986). No strong-HEV tier is used.

These model-derived HEVs are distinct from snpEff HIGH-impact or PolyPhen-2 damaging annotations. “Strong” describes the predicted effect magnitude relative to the combined calibration cutoff; it is not a confidence score.

Population differentiation and diversity statistics: Population genetic statistics are displayed across the gene neighborhood to provide evolutionary context. Genetic differentiation is quantified using F_ST, calculated in sliding windows of 10 kb with a step size of 1 kb between predefined rice populations. Nucleotide diversity (π) is computed using the same window and step size to measure local genetic variation within populations. Joint visualization of F_ST and π highlights genomic regions under potential selection and links regulatory variation with population-level evolutionary processes.

Genetic Variants in the Gene Neighborhood: All genetic variants located within the gene body and flanking regions are listed and visualized along the genomic coordinate system. Variants are annotated with sequence context, functional annotations, and population-level allele statistics, enabling detailed inspection of local genetic diversity and variant distribution.