RNA-Integrate (RNA-i): Integrative Analysis of Small RNA and mRNA High-Throughput Sequencing Data

RNA-Integrate (RNA-i) is a reproducible workflow for differential expression analysis of small RNA and mRNA sequencing data using DESeq2, with optional integrative analysis to identify small RNA–mRNA target relationships.

The workflow supports both standalone differential expression analysis (mRNA or small RNA independently) and integrative analysis, where differential expression results are combined with a user-supplied small RNA–mRNA target table to generate integrative visualizations, statistical summaries, and target-pair results tables.

RNA-i can be run through either:

• a Shiny application, providing a user-friendly and interactive interface for first-time users, or
• a highly configurable R Markdown workflow, designed for rapid, reproducible, and customizable execution.

Input files include sequencing count tables, experimental metadata, and (for integrative analyses) a small RNA–mRNA target table. Outputs include differential expression tables, quality-control plots, expression visualizations, and integrative analyses of small RNA–mRNA relationships.

The workflow can be run individually for mRNA or small RNA differential expression analysis or in integrative mode. The quick start guides below illustrate the basic execution process for each version of the workflow.

Workflow Modes

Differential Expression Analysis (mRNA)

mRNA differential expression analysis includes:

• Differential expression results tables
• PCA plots
• Intra-condition replicate scatterplots
• Mean reads scatter plots
• MA plots
• Heatmaps

Supported mRNA Input Formats

• HTSeq
• Counts Matrix
• RSEM
• Salmon
• Kallisto

Differential Expression Analysis (Small RNA)

Small RNA differential expression analysis includes all standard DESeq2 visualizations and optional class-based customization.

Supported Small RNA Input Formats

• Counts Matrix
• tinyRNA output tables

Small RNA analyses support automatic handling of annotation columns, feature classes, and common-name extraction for visualization and integration.

Integrative Analysis

Integrative analysis combines differential expression results from small RNA and mRNA datasets to evaluate small RNA–mRNA target relationships.

Integrative Outputs

• Integrative results tables
• Cosmic plots
• Slope plots
• Statistical analyses

Shiny App Pipeline (Under Construction!)

1. Prepare count files for mRNA or small RNA experiments
2. (Optional) Prepare a Gene Table CSV for common-name conversion and class-based customization
3. Launch App from RStudio or the command line
4. Generate or import a Metadata CSV to specify experimental design
5. (Optional) Import a Gene Table CSV within the app
6. (Optional) Generate or import a Class Parameters CSV to customize mean reads scatter plots by gene class
7. Run analysis (5–10 minute runtime) to produce, save, and interact with tables and plots

See the Shiny App Directory for more details.

Markdown Pipeline

1. Prepare sequencing count data
2. Prepare a Metadata CSV to specify experimental design
3. (Optional) Prepare a Gene Table CSV for common names conversion and gene-class annotations
4. (Optional) Prepare a Class Parameters CSV to customize mean reads scatter plots by gene class
5. Prepare a Parameters YAML to specify analysis settings and output options
6. Execute Markdown Script from RStudio or the command line (3–5 minute runtime) to produce and save a variety of tables and plots, including optional interactive HTML files

See the Markdown Pipeline Directory for more details.

Integrative Pipeline

1. Prepare mRNA count data (HTSeq, Counts Matrix, RSEM, Salmon, or Kallisto)
2. Prepare small RNA count data (Counts Matrix or tinyRNA format)
3. Prepare a Gene Table CSV containing small RNA–mRNA target relationships
4. Prepare Metadata CSV files for mRNA and small RNA experiments to specify experimental design
5. (Optional) Prepare Class Parameters CSV files for mRNA and/or small RNA experiments to customize mean reads scatter plots by gene class
6. Prepare Integrative Parameters YAML to specify comparisons and output settings
7. Execute Markdown Script from RStudio or the command line (10–15 minute runtime) to produce and save a variety of tables and plots, including interactive HTML files

See the Integrative Pipeline Directory for more details.

Authors

• Taiowa Montgomery — 05/2021–present — Colorado State University — taimontgomery
• Spencer Kuhn — 06/2021–present — Colorado State University — smcguirekuhn
• Paige Lillibridge — 08/2025–present — Colorado State University — paigelily