Spatial rice-AMF

Source code used to analyse the data and generate graphics for the publication:

Spatiotemporal regulation of arbuscular mycorrhizal symbiosis at cellular resolution

Tania Chancellor^1,§ and Gabriel Ferreras Garrucho^1,§, Garo Z. Akmakjian², Héctor Montero¹, Emma Wallington³, Samik Bhattacharya⁴, Christian Korfhage⁴, Julia Bailey-Serres², Uta Paszkowski^1,5,*

¹Crop Science Centre, Department of Plant Sciences, University of Cambridge, 93 Lawrence Weaver Rd, Cambridge, United Kingdom, CB3 0LE. ²University of California-Riverside, Department of Botany and Plant Sciences and Center for Plant Cell Biology, Riverside, CA 92521-0217, USA. ³NIAB, Park Farm, Villa Road, Histon, Cambridge, CB24 9NZ, UK. ⁴Resolve BioSciences GmbH, Creative Campus Monheim, Creative-Campus-Allee 12, 40789 Monheim, Germany. ⁵RIKEN, Center for Sustainable Resource Science, Yokohama, Kanagawa 230-0045, Japan. ^*Corresponding author: up220@cam.ac.uk ^§These authors contributed equally to this article.

---

Repository Contents

Molecular Cartography Analysis (MolecularCartography/)

This directory contains scripts for analyzing spatial transcriptomics data generated using the Resolve BioSciences Molecular Cartography platform.

• MC_singlecell_Seurat.R: Single-cell clustering analysis and quality control on all detected cells in the dataset
• MC_arbuscule_singlecell_Seurat.R: Focused single-cell clustering analysis specifically for arbuscule-containing cells
• MC_colocalization.R: Hierarchical clustering analysis to identify co-localization patterns of gene expression
• MC_DGE_DESeq2.R: Differential gene expression analysis between cell types and conditions
• MC_vesicle_arbuscules.R: Calculates transcripts per unit area in vesicles and arbuscules
• cell_segmentation_for_ImageJ/: Segmented cell coordinates for viewing in raw images (AllCells/ and Arbuscules_Vesicles/ subfolders)
• raw_data/: Contains raw gene count matrices and metadata:
  • Arbuscule_metadata.csv: Metadata for arbuscule-specific analysis
  • Arbuscule_pooled_gene_counts.csv: Gene count matrix for arbuscule samples
  • gene_count_matrix_sample_probe_filtered.csv: Filtered gene count matrix
  • mock_median_matrix.csv: Mock treatment median expression values
  • myc_median_matrix.csv: Mycorrhizal treatment median expression values
  • pooled_ROI_matrix_filteredv2.txt: Filtered region-of-interest matrix
  • sample_info.csv: Sample metadata and annotations
  • vesicle_arbuscule_areas.csv: Consolidated area measurements for vesicles and arbuscules
  • vesicle_arbuscule_transcripts.csv: Consolidated gene transcript counts for vesicles and arbuscules
• results/: Output directory organized by analysis type:
  • single_cell/{plots,tables}/: Single-cell analysis results
  • arbuscule_single_cell/{plots,tables}/: Arbuscule-specific results
  • colocalization/{plots,tables}/: Clustering and co-localization results
  • DGE/{plots,tables}/: Differential expression results
  • vesicle_arbuscule/{plots,tables}/: Vesicle/arbuscule transcript per area analysis results

TRAP-seq Analysis (TRAPseq/)

This directory contains scripts for analyzing translating ribosome affinity purification sequencing (TRAP-seq) data.

• TRAP_DGE_analysis.R: Differential gene expression analysis using DESeq2, including:
  • Quality control visualizations
  • Library complexity assessment
  • Variance-stabilizing transformation (VST)
  • DEG identification and annotation
• TRAP_plotting_results.R: Comprehensive visualization script generating publication-quality figures, including:
  • PCA plots
  • Heatmaps with multiple gene sets
  • Venn diagrams for DEG comparisons
  • GO enrichment analysis and plotting
• gene_count_tables/: Pre-processed gene count matrices:
  • gene_count_matrix.csv: Raw TRAP-seq counts
  • DESeq2_counts.csv: DESeq2-normalized counts
  • DESeq2_normalized_counts.csv: Size-factor normalized counts
  • DESeq2_VSD_counts.csv: Variance-stabilized transformed counts
  • DESeq2_VSD_counts_averages.csv: VST counts averaged by condition
• info_sheets/: Gene annotations, GO terms, and metadata tables
• results/: Output directory organized by analysis type:
  • plots/{Average_expression_plots,Heatmaps,Venn,Correlation,Scatterplots,Gene_count_plots,DEG_number,PCA,GO}/
  • DEG_tables/: Activated and repressed gene lists for each comparison
  • comparison_tables/: Full comparison results including log2FC and adjusted p-values

---

Usage

1. Clone or download this repository locally

   git clone https://github.com/gabriel-ferreras/Spatial_rice_AMF_2025.git
   cd Spatial_rice_AMF_2025

2. Required R packages

   Both analysis pipelines require the following packages (see script headers for complete dependency lists):

   • Bioconductor packages: DESeq2, Seurat, ggplot2, ComplexHeatmap, clusterProfiler
   • CRAN packages: tidyverse, ggpubr, scales, ggvenn, circlize

3. Molecular Cartography Analysis

   The Molecular Cartography scripts process spatial transcriptomics data:

   A. Run MC_singlecell_Seurat.R

   • Loads raw count matrix and sample metadata
   • Performs QC filtering, normalization, and clustering
   • Outputs single-cell clustering plots and tables
   • Results saved to results/single_cell/{plots,tables}/

   B. Run MC_arbuscule_singlecell_Seurat.R

   • Focuses analysis on arbuscule-containing cells
   • Performs subset clustering and analysis
   • Results saved to results/arbuscule_single_cell/{plots,tables}/

   C. Run MC_colocalization.R

   • Performs hierarchical clustering for co-localization analysis
   • Generates clustering heatmaps
   • Results saved to results/colocalization/{plots,tables}/

   D. Run MC_DGE_DESeq2.R

   • Differential expression analysis between cell types/conditions
   • Outputs DEG tables and volcano plots
   • Results saved to results/DGE/{plots,tables}/

   E. Run MC_vesicle_arbuscules.R

   • Analyzes gene expression normalized to structure area (transcripts per μm²)
   • Compares expression levels in vesicles vs arbuscules
   • Generates individual gene plots and multi-gene comparison plots
   • Results saved to results/vesicle_arbuscule/{plots,tables}/
   • Note: Requires raw_data/vesicle_arbuscule_areas.csv and raw_data/vesicle_arbuscule_transcripts.csv to be present. If working from raw per-image data, run the consolidation script raw_data/consolidate_vesicle_arbuscule_data.R first.

4. TRAP-seq Analysis

   The TRAP-seq scripts analyze translating ribosome affinity purification sequencing data:

   A. Run TRAP_DGE_analysis.R

   Sections include:

   1. LOADING PACKAGES - Install and load required R packages

   2. DESEQ2

      • Constructs DESeqDataSet from the count matrix
      • Performs sample QC, pre-filtering, and variance-stabilizing transformation (VST)
      • Saves normalized count matrices for downstream analysis

   3. DEG RESULTS

      • Extracts statistically significant DEGs
      • Generates MA plots and volcano plots
      • Outputs DEG tables with annotations to DEG_tables/ and comparison_tables/

   B. Run TRAP_plotting_results.R to generate publication-quality figures

   Important: For Figure S12 to generate correctly, copy the Molecular Cartography comparison table into the TRAP-seq comparison_tables/ folder before running this script. The code requires the Molecular Cartography results to compare with TRAP-seq results:

   cp MolecularCartography/results/DGE/tables/* TRAPseq/comparison_tables/

   Then uncomment and run the FC and FDR matrix generation code in the script.

   Sections include:

   1. LOADING PACKAGES - Loads required libraries

   2. LOADING DATA

      • Loads gene annotations, GO terms, and gene sets
      • Loads DEG lists and count matrices
      • Prepares FC and p-value matrices

   3. PCA

      • Principal component analysis from VST-transformed counts
      • Figure 5B output

   4. DEG NUMBER PLOTS

      • Visualizes number of differentially expressed genes per comparison
      • Figure 5C output

   5. VENN DIAGRAMS

      • Shows DEG overlap between conditions and published datasets
      • Figures 5D, S10B-D, 6B outputs

   6. NORMALISED GENE COUNT PLOTS

      • Individual gene expression patterns across samples
      • Figure S10A output

   7. HEATMAPS

      • Expression heatmaps for specific gene categories:
        • Cell wall genes (Figure S14B)
        • Strigolactone pathway genes (Figure S13)
        • Cell-type markers (Figure 5E/S11)
        • Nutrient signaling (Figure 6D/S15)
        • Defense and immunity genes (Figure S14A)
        • Resolve gene set comparison (Figure S12)

   8. GO ENRICHMENT

      • Gene ontology enrichment analysis for DEG sets
      • Generates GO term plots by category
      • Figure 6A output

---

Data Availability

Raw sequences, images, and gene count matrices generated in this study are available in the NCBI GEO database under the following accessions:

• TRAP-seq and Total RNA sequencing data: GSE308140

  • Includes raw sequencing reads (fastq files) and processed count matrices for total and TRAP-pulled mRNA samples
  • https://www.ncbi.nlm.nih.gov/geo/query/acc.cgi?acc=GSE308140

• Molecular Cartography data: GSE316219

  • Includes raw image files and processed spatial transcriptomics count matrices
  • https://www.ncbi.nlm.nih.gov/geo/query/acc.cgi?acc=GSE316219

To reproduce the analysis from raw data, use the Lexogen NGS Kangaroo Data Analysis Platform or equivalent pipeline to perform:

• Quality control and trimming
• UMI deduplication
• STAR alignment to the Oryza sativa Nipponbare reference transcriptome (Os-Nipponbare-Reference-IRGSP-1.0.58)
• Transcript count quantification

Alternatively, use the pre-processed gene count matrices provided in this repository.

---

Citation

If you use this code or dataset, please cite the associated publication (DOI and citation to be added upon acceptance).