Osta-Docs-Local Skill

Comprehensive assistance with spatial transcriptomics analysis using Bioconductor's OSTA (Orchestrating Spatial Transcriptomics Analysis) framework.

When to Use This Skill

This skill should be triggered when:

• Working with spatial omics data analysis in R/Bioconductor
• Exploring the Bioconductor spatial ecosystem and finding relevant packages
• Learning spatial transcriptomics workflows including sequencing-based and imaging-based platforms
• Implementing spatial data analysis steps like quality control, normalization, clustering
• Finding Bioconductor packages for specific spatial analysis tasks (clustering, visualization, etc.)
• Understanding package ecosystem metrics and growth trends in spatial omics
• Debugging spatial analysis code or troubleshooting Bioconductor package issues
• Learning best practices for spatial transcriptomics data analysis

Quick Reference

Essential Bioconductor Package Discovery

Example 1: Find all spatial omics packages

library(BiocPkgTools)
df <- biocPkgList()
.f <- \(x, y=df) vapply(y$biocViews, \(.) all(x %in% .), logical(1))
spatial_packages <- df$Package[.f("Spatial")]
print(spatial_packages)

Example 2: Find packages for specific spatial analysis tasks

# Find spatial clustering packages
spatial_clustering <- df$Package[.f(c("Spatial", "Clustering"))]
print(spatial_clustering)

Example 3: Interactive package exploration

library(BiocPkgTools)
biocExplore()

Key Spatial Analysis Libraries

Example 4: Core spatial packages to load

library(ggspavis)          # Spatial visualization
library(OSTA.data)         # OSTA example datasets
library(SpatialExperiment) # Spatial data structures
library(SpatialExperimentIO) # Data import/export

Example 5: Analyze Bioconductor ecosystem growth

library(dplyr)
library(ggplot2)

# Get package growth metrics over time
ids <- c("SingleCell", "Spatial")
gg <- lapply(ids, \(id) {
  nm <- df$Package[.f(id)]
  ys <- BiocPkgTools:::getPkgYearsInBioc(nm) |>
    mutate(first=first_version_release_date) |>
    mutate(last=last_version_release_date) |>
    filter(!is.na(first))
  # Complete months between first/last dates
  lapply(split(ys, ys$package), \(.) {
    data.frame(package=.$package, date=seq(.$first, .$last))
  }) |> do.call(what=rbind) |> group_by(date) |> count()
}) |> bind_rows(.id="biocViews")

Example 6: Visualize package ecosystem trends

ggplot(gg, aes(date, n, col=biocViews)) +
  geom_line(linewidth=0.8) +
  geom_smooth(data=filter(gg, n >= 5),
      method="lm", se=FALSE, linewidth=1) +
  scale_x_date(date_breaks = "1 year", date_labels = "%Y") +
  labs(x=NULL, y="# packages") +
  theme_bw()

Example 7: Find packages by specific analysis categories

# Common spatial analysis categories
categories <- c("BatchEffect", "Normalization", "QualityControl",
               "Visualization", "Clustering", "DimensionReduction",
               "FeatureExtraction", "DifferentialExpression",
               "GeneSetEnrichment")

# Count packages for each category
package_counts <- lapply(categories, \(cat) {
  n <- sum(.f(c("Spatial", cat)))
  data.frame(category=cat, count=n)
}) |> do.call(what=rbind)

Key Concepts

Spatial Omics Data Types

• Sequencing-based platforms: Visium, Visium HD, Visium DLPFC, Visium CRC
• Imaging-based platforms: Xenium, CODEX, MIBI
• Data structures: SpatialExperiment, SpatialFeatureExperiment

Bioconductor Ecosystem

• biocViews: Hierarchical categorization system for packages
• Package discovery: Using BiocPkgTools for ecosystem exploration
• Growth metrics: ~0.48 spatial packages added per year on average
• Package lifetime: Average 4 years in Bioconductor

Analysis Workflow Steps

1. Data import: Reading spatial data from various platforms
2. Quality control: Filtering and assessment of data quality
3. Normalization: Standardizing expression values
4. Dimensionality reduction: PCA, UMAP, t-SNE for spatial data
5. Clustering: Identifying spatial domains or cell types
6. Visualization: ggspavis for spatial plots
7. Statistical testing: Differential expression, spatial statistics

Reference Files

This skill includes comprehensive documentation in references/:

analysis.md

• Overview: Introduction to OSTA (Orchestrating Spatial Transcriptomics Analysis)
• Content: Complete book structure with 39 pages covering spatial transcriptomics workflows
• Key sections: Sequencing-based and imaging-based platforms, platform-independent analyses, cross-platform analyses
• Target audience: Users learning spatial transcriptomics with Bioconductor

pages.md

• Detailed documentation: Chapter 4 "Ecosystem" with comprehensive package exploration
• Interactive examples: R code for discovering and analyzing Bioconductor spatial packages
• Statistical analysis: Package growth trends, lifetime analysis, co-occurrence patterns
• Practical tools: Code snippets for package discovery and ecosystem metrics

Working with This Skill

For Beginners

1. Start with core concepts: Read the analysis.md reference to understand OSTA framework
2. Explore package ecosystem: Use Example 1-3 to discover relevant spatial packages
3. Learn basic workflows: Focus on sequencing-based or imaging-based platform sections
4. Practice with examples: Run the R code examples to get hands-on experience

For Intermediate Users

1. Advanced package discovery: Use Example 7 to find packages for specific analysis tasks
2. Ecosystem analysis: Apply Example 5-6 to understand package trends and metrics
3. Multi-platform workflows: Explore cross-platform analysis sections
4. Integration with Python: Check sections on Python interoperability

For Advanced Users

1. Custom analysis pipelines: Combine multiple packages for specialized workflows
2. Ecosystem research: Use the metrics and trend analysis tools for research
3. Package development: Understand ecosystem patterns for new package creation
4. Large-scale analyses: Apply multi-sample and differential analysis methods

Navigation Tips

• Search by platform: Look for specific platform names (Visium, Xenium) in reference docs
• Find by analysis type: Use keywords like "clustering", "normalization", "visualization"
• Cross-reference examples: Code examples are linked to specific analysis steps
• Follow workflows: Each platform has dedicated workflow sections with complete examples

Resources

references/

• analysis.md: Complete OSTA book overview and structure
• pages.md: Detailed ecosystem chapter with interactive R examples
• Both files contain executable R code with proper language annotations
• Examples are extracted from official documentation and tested

External Resources

• Bioconductor BiocViews: Online package browsing
• OSTA book: Complete online resource for spatial transcriptomics
• Related book: OSCA (Orchestrating Single-Cell Analysis) for single-cell foundations

Notes

• This skill focuses on spatial transcriptomics using Bioconductor in R
• BiocPkgTools is essential for package discovery and ecosystem analysis
• Code examples emphasize reproducible workflows with real datasets
• The skill bridges sequencing-based and imaging-based spatial platforms
• Python interoperability is supported for integrated analyses

Updating

To refresh this skill with updated documentation:

1. Re-run the OSTA documentation scraper
2. Update package lists and examples as Bioconductor evolves
3. Add new platform workflows as they become available
4. Maintain currency with Bioconductor releases (every 6 months)