Tutorial: Exploring the Artemisia Database

Welcome to the Artemisia Database, a comprehensive platform designed to facilitate research on Artemisia annua gene expression, transcription factors, functional annotations, and more. This tutorial will walk you through the database’s key features, showing you how to navigate its interface and utilize its tools effectively. Whether you’re a biologist, bioinformatician, or researcher, this guide will help you get started.

---

Table of Contents

1. Getting Started
   • Accessing the Database
2. Home Page Overview
3. Gene Expression
   • Global Expression Viewer
   • Median Expression Per Tissue
   • Artemisinin Pathway Genes
   • Overall Category Distribution
4. Transcription Factors
   • PlantTFDB
   • Pfam
   • Tissue-specific TFs
5. Gene-Metabolite Correlation Analysis
6. Functional Annotation
7. Gene Editing and Epigenetics
   • CRISPR sgRNA Designer
   • Methylation
8. Tools
   • JBrowse Genomic Browser
   • BLAST
   • Co-expression Analysis
   • GO Enrichment Analysis
   • Get Gene Sequences
9. Download
   • Download Data by Tissue
   • Download Data by Project
   • Download Reference Files

---

Getting Started

Accessing the Database

• Navigate to the Portal: Open your web browser and go to the Artemisia Database.
• Interactive Interface: The platform is built using R Shiny, providing a highly responsive and intuitive environment for data exploration.
• Guidance & Documentation: At the top of every page, you will find a dropdown menu. Use this menu to access detailed information regarding the section’s purpose and specific step-by-step instructions.

---

Home Page Overview

Upon accessing the database, you will land on the Home tab, which serves as the central hub for the platform:

• Welcome & Overview: This section provides a high-level introduction to the Artemisia-DB project. It outlines the platform’s core objective: facilitating the exploration and functional analysis of Artemisia annua gene expression and multi-omics data.
• Getting Started: Use this page to familiarize yourself with the database’s scope before diving into the specific analysis modules.

---

Gene Expression

The Gene Expression menu contains four powerful tools for visualizing expression data.

Global Expression Viewer

Purpose

The Global Expression Viewer allows you to visualize high-dimensional gene expression data through an interactive t-SNE plot. This tool is designed to help you identify expression clusters across different tissues and access the underlying metadata for each sample.

---

Step-by-Step Instructions

1. Navigate: Go to Gene Expression > Global Expression Viewer in the main menu.
2. Explore the Plot: Interact with the t-SNE plot (powered by Plotly) to see how samples cluster.
3. Inspect Metadata: Review the Filtered t-SNE Metadata table below the plot for comprehensive sample details.

---

Key Interactive Features

• Dynamic Hover: Hover over any point in the t-SNE plot to identify the specific tissue type (e.g., “Leaf,” “Root,” or “Trichome”).
• Bidirectional Filtering: Click on a point in the plot to instantly filter the metadata table to that specific sample.
• Literature Links: If a sample originates from a published study, the Pubmed_ID column will provide a direct hyperlink to the research paper on PubMed.

---

Visual Overview

Figure 1: The t-SNE plot displays gene expression data colored by plant tissue. Interactive tools allow for zooming and point-specific identification.

Figure 2: The metadata table dynamically updates based on your plot selections, offering deep-dives into sample attributes and publication IDs.

---

Median Expression Per Tissue

Purpose

This tool allows you to aggregate and visualize the median expression levels of specific genes across various plant tissues. It is particularly useful for comparing the expression profiles of gene families or sets of co-expressed genes.

---

Step-by-Step Instructions

1. Navigate: Go to Gene Expression > Median Expression Per Part.
2. Input Gene IDs: * Enter up to 100 Gene IDs (one per line) into the text area.
   • Alternatively, upload a .txt or .csv file containing your IDs.
   • Quick Start: Use the Example Gene ID buttons to automatically populate the field with a sample set.
3. Select Tissues: Use the checkboxes to choose the specific plant parts (e.g., “Root”, “Leaf”, “Trichome”) you wish to include in the analysis.
4. Process Data: Click the Calculate Median Expression button.
5. Analyze & Save: * Review the generated heatmap, which displays log2-transformed expression values.
   • To save the visualization, click the Camera Icon in the top-right corner of the plot to download it as a PNG.

---

Example Workflow

To test the tool, click Example#1 to load IDs like mikado.chr7G1274 and mikado.chr4G1337. Select Leaf and Flower, then click Calculate. The resulting heatmap will provide a direct visual comparison of these genes across the two tissues.

---

Visual Guidance

Figure 3: Use the example buttons for a quick setup, then refine your analysis by selecting specific plant parts.

Figure 4: The interactive heatmap visualizes relative expression levels. Hover over cells to see exact log2 values.

---

Artemisinin Pathway Genes

Purpose

This module provides a focused environment for exploring genes specifically linked to artemisinin biosynthesis. It integrates functional data, expression profiles, and sequence information through a centralized, multi-tab interface.

---

Step-by-Step Instructions

1. Access the Module

Navigate to Gene Expression > Artemisinin Pathway Genes. The “Main Table” tab will load by default, listing all relevant pathway genes.

2. Search and Filter

Use the search panel to isolate specific genes of interest:

• By Gene ID: Enter IDs (one per line) and click Search by Gene ID.
• By Gene Name: Enter keywords (e.g., ADS, CYP71AV1) and click Search by Gene Name.

3. Select and Interact

Once you have located your target genes, click on their rows in the Main Table to highlight them. This enables the action buttons at the bottom left:

• Copy Gene ID(s): Copies the selected IDs to your clipboard for use in other modules (like the Median Expression Viewer).
• Check Gene Expression: Automatically switches you to the Expression tab and carries over your selected IDs.

4. Analyze Expression & Sequences

• Expression Tab: Select your desired plant parts (e.g., Leaf, Root, Trichome) and click Calculate Median Expression to generate a custom heatmap.
• Sequence Tab: View the nucleotide or protein sequences for your selected genes. You can download these directly as a .txt file for local use.

---

Workflow Overview

Feature	Description
Main Table	Provides functional annotations and literature references.
Expression	Generates dynamic heatmaps based on tissue-specific data.
Sequence	Offers high-throughput access to FASTA-formatted sequences.

---

Overall Category Distribution

Purpose

This module provides a high-level classification of the Artemisia annua transcriptome. Using an interactive donut chart, you can explore gene categories (such as “Tissue-Specific” vs. “Broadly Expressed”) and drill down into the functional data and sequences associated with each group.

---

Step-by-Step Instructions

1. Navigate: Go to Gene Expression > Overall Category Distribution.
2. Explore the Donut Chart: * The chart displays the proportion of genes within different expression categories.
   • Click a segment (e.g., “Specific”) to instantly filter the data table below to show only genes belonging to that category.
3. Search Specific Genes: Alternatively, enter one or more Gene IDs in the search box to identify their specific classifications and view their details in the table.
4. Select & Action: Click on one or more rows in the Main Table to enable the following tools at the bottom left:
   • Copy Gene ID(s): Saves the selected IDs to your clipboard.
   • Check Gene Expression: Redirects you to the Expression tab with your selected genes pre-loaded.
5. Analyze & Download:
   • Expression Tab: Select plant parts and click Calculate Median Expression to visualize a heatmap of your selected category.
   • Sequence Tab: View and download the nucleotide or protein sequences for your selected genes.

---

Visual Guidance

Figure 5: The interactive donut chart allows you to filter the entire dataset by clicking on specific expression categories, such as tissue-specific or constitutive genes.

---

Pro-Tip

This module is the fastest way to identify “specialist” genes. By clicking the Specific segment and then using the Check Gene Expression button, you can quickly verify which tissue (e.g., Trichome or Root) those genes are primarily active in.

---

Transcription Factors

The Transcription Factors menu provides specialized tools for identifying and analyzing regulatory proteins within the Artemisia annua genome, primarily categorized by the PlantTFDB framework.

PlantTFDB Module

Purpose

This tool allows you to explore transcription factor (TF) families, visualize their distribution, and analyze their tissue-specific expression patterns.

---

Step-by-Step Instructions

1. Navigate: Go to Transcription Factors > PlantTFDB.
2. Filter by Family: * Interact with the TF Family Distribution Plot.
   • Click a bar (e.g., “bZIP”) to instantly filter the “Main Table” to that specific family.
3. Search & Select:
   • In the Main Table, search for specific Gene IDs or browse the filtered list.
   • Select rows to activate the action buttons at the bottom left.
4. Execute Actions:
   • Copy Gene ID(s): Copies your selection to the clipboard.
   • Check Gene Expression: Seamlessly transfers your selected TFs to the Expression tab.
5. Analyze Results:
   • Expression Tab: Select target tissues (e.g., “Root”, “Leaf”) and click Calculate Median Expression to generate a heatmap.
   • Sequence Tab: View and download the transcript sequences for your selected TFs.

---

Example Workflow: Analyzing bZIP TFs

To visualize the expression of bZIP transcription factors:

1. Click the bZIP bar in the distribution plot.
2. Select two or more genes from the Main Table.
3. Click Check Gene Expression.
4. On the Expression tab, select Leaf and Stem, then click Calculate.
5. Use the Camera Icon or the Download Plot button to save your heatmap as a PNG, JPEG, or PDF.

---

Visual Guidance

Figure 5: Interactive bar plot showing TF family distribution. Clicking “bZIP” filters the dataset automatically.

Figure 6: The Main Table updates based on your plot selection. Use the buttons at the bottom left to transition to expression analysis.

Figure 7: Generate and export high-resolution heatmaps for your selected transcription factors across various plant parts.

Pfam

Purpose

The Pfam module allows you to explore transcription factors through the lens of conserved protein domains. This approach is ideal for identifying regulatory genes that share specific structural motifs—such as Zinc fingers or Leucine zippers—even if they are not fully categorized into traditional plant TF families.

---

Step-by-Step Instructions

1. Navigate: Go to Transcription Factors > Pfam.
2. Filter by Domain: View the Pfam Domain Distribution plot. Click on any bar (e.g., “PF00096”) to filter the Main Table below to show only genes containing that specific domain.
3. Detailed Exploration: Within the Details box, utilize the tabs to analyze your selection:
   • Main Table: Search and select specific Gene IDs. Highlight rows to enable the Check Gene Expression or Copy Gene ID(s) buttons at the bottom left.
   • Expression: Select target tissues and click Calculate Median Expression to generate a heatmap for the domain-specific genes.
   • Sequence: View or download transcript sequences for your selected genes as a .txt file.

---

Tissue-specific TFs

Purpose

This module is designed to pinpoint transcription factors that exhibit localized activity. By focusing on genes with high expression in specific tissues (e.g., Leaf, Root, or Trichome), you can identify the primary regulators governing tissue-specific development and specialized metabolism.

---

Step-by-Step Instructions

1. Navigate: Go to Transcription Factors > Tissue-specific TFs.
2. Interact with the Heatmap: The overview heatmap displays TF types across different tissues. Click a specific cell (e.g., the intersection of CO-like TF and Leaf) to instantly filter the Main Table in the “Details” box below.
3. Select & Analyze:
   • Main Table: Highlight the rows of the TFs you wish to investigate.
   • Check Gene Expression: Click this button to carry your selected genes over to the Expression tab automatically.
4. Validate Specificity: * On the Expression tab, select a variety of plant parts and click Calculate Median Expression.
   • This allows you to visually confirm whether the TF is exclusively expressed in your tissue of interest.

---

Example Workflow

To investigate stem-specific regulators, click the ERF TF cell in the Stem row of the overview heatmap. Select the resulting genes in the Main Table and click Check Gene Expression. By selecting both Leaf and Root on the expression tab, you can generate a heatmap that clearly demonstrates the gene’s preferential activity in leaf tissue.

---

Visual Guidance

Figure 8: The interactive heatmap allows for rapid filtering. Clicking a cell instantly isolates the corresponding transcription factors in the data table below.

---

Gene-Metabolite Correlation Analysis

Purpose

The Gene-Metabolite Correlation Analysis module is a powerful multi-omics integration tool. It allows you to discover functional relationships between gene expression (transcriptomics) and metabolite abundance (metabolomics) across different experimental conditions and tissues.

---

Step-by-Step Instructions

1. Configure Analysis Controls

Use the sidebar on the left to set up your correlation parameters:

• Select Metabolite: Search for a specific compound (e.g., Artemisinin, Artemisinic acid).
• Analysis Mode: * Discovery Mode: Automatically identifies the top 30 genes most strongly correlated with your selected metabolite.
  • Hypothesis Mode: Allows you to input a custom list of Gene IDs to test specific correlations.
• Correlation Filter: Adjust the slider to set a minimum absolute correlation threshold (\(|r|\)). Only pairs meeting this value will be displayed.
• Run Analysis: Click the green button to process the data.

2. Explore the Correlation Table

Once processed, the Correlation Table tab provides a searchable list of results:

• Table Data: View Gene IDs, Correlation coefficients (\(r\)), and functional descriptions.
• Row Selection: Click a single row to highlight a specific gene-metabolite pair. This action updates the Trend Visualization and Gene Context tabs.
• Download: Use the Download CSV button to save your filtered results for offline analysis.

3. Visualize the Correlation Matrix

Switch to the Correlation Matrix tab to see a high-level heatmap of the relationships:

• Red Cells: Represent positive correlations (gene and metabolite levels increase together).
• Blue Cells: Represent negative correlations (as one increases, the other decreases).
• Interactivity: Hover over any cell to see the exact correlation value and gene description.

4. Analyze Trends (Dual-Axis Plot)

The Trend Visualization tab offers a detailed look at how a selected pair behaves across four experimental conditions (WT Young Leaf, WT Mature Leaf, Mutant Young Leaf, and Mutant Mature Leaf):

• Bars (Left Y-Axis): Represent the metabolite abundance/intensity.
• Line (Right Y-Axis): Represents the gene expression levels in TPM.
• Co-regulation: If the bars and lines follow the same pattern, it strongly suggests the gene is involved in the metabolite’s biosynthetic pathway.

5. Gene Functional Context

The Gene Context tab provides a deep dive into the selected gene, showing its Pfam domains and a summary of other metabolites it may be correlated with.

---

Experimental Conditions Reference

The data is derived from four specific states to help identify developmental and mutational effects:

1. WT_YL: Wild-type Young Leaf
2. WT_ML: Wild-type Mature Leaf
3. MUT_YL: tdd1 Mutant Young Leaf
4. MUT_ML: tdd1 Mutant Mature Leaf

---

Visual Guidance

Figure 9: Use the sidebar to switch between Discovery and Hypothesis modes and set your correlation thresholds.

Figure 10: The Trend Analysis plot allows for direct comparison of metabolite intensity (bars) and gene expression (line) across wild-type and mutant conditions.

---

Functional Annotation

Purpose

The Functional Annotation module provides a centralized interface to retrieve biological context for Artemisia annua genes. By integrating multiple databases, you can identify gene functions, metabolic pathways, and protein domains through three flexible query methods.

---

Step-by-Step Instructions

1. Navigate: Go to the Functional Annotation section in the main menu.

2. Select Your Query Method:

   • Annotation Table: * Choose a specific database from the dropdown menu (e.g., GO, KEGG, Pfam, or SwissProt).

     • Input corresponding IDs (e.g., GO:0003674).
     • Tip: Use the default example IDs provided in the text area to ensure your format matches the selected database.

   • Gene-Based Queries: * Input specific Gene IDs (one per line) from the latest A. annua gene model.

     • This is the most efficient way to see all biological roles assigned to a particular gene across every available database.

   • Functional Descriptions: * Perform a keyword search (e.g., “RNA polymerase” or “transferase”).

     • The system will scan all functional descriptions to find matches, helping you discover genes related to specific biological activities.

3. Execute Search: Click the Search button corresponding to your chosen method.

4. Explore Multi-Tab Results: The results are displayed in a dynamic table with tabs representing different databases (e.g., GO, Pfam, KEGG, EggNOG, etc.). This allows you to toggle between different layers of functional information for the same set of genes.

---

Example Workflow

To find information for a specific gene:

1. Enter mikado.Super-Scaffold_100038G100 into the Gene-Based Queries field.
2. Click Search Gene IDs.
3. Navigate through the generated tabs (GO, Pfam, KEGG) to view the associated Gene Ontology terms, protein domains, and metabolic pathways for that gene.

---

Visual Guidance

Figure 11: Searching by Gene ID generates a consolidated view. Use the tabs to switch between database-specific annotations like KEGG pathways or GO terms.

---

Gene Editing and Epigenetics

CRISPR sgRNA Designer

Purpose

The CRISPR sgRNA Designer is a specialized tool for planning genome editing experiments in Artemisia annua. Using the crisprDesign methodology and a custom-built LQ9v1 BSgenome, this tool identifies optimal single-guide RNA (sgRNA) sequences for various CRISPR nucleases while considering coding sequences (CDS) and potential quality constraints.

---

Step-by-Step Instructions

1. Verify System Status

Before starting, check the Status Check badges at the top of the page.

• Both the BSgenome and Annotation must show as loaded (green).
• If they are not loaded, use the Force Load Genome or Force Load Annotation buttons to initialize the reference data.

2. Input Target Genes

In the Design Parameters sidebar:

• Target Gene ID(s): Enter one or more Mikado Gene IDs (e.g., mikado.chr1G1335) into the text area (one per line).
• Target Region: * CDS: Recommended for precision knockouts (targets coding regions only).
  • Gene ± 500bp: Includes the promoter and terminator regions.
  • Custom: Allows you to define specific upstream/downstream boundaries.

3. Select Nuclease and Filters

• CRISPR Nuclease: Choose your enzyme system. The tool supports SpCas9 (NGG PAM), AsCas12a (TTTV PAM), enCas12a, and SaCas9.
• Design Filters: Apply quality control filters:
  • GC Content: Restricted to 20-80% for optimal stability.
  • Poly-T stretches: Removes sequences that may terminate RNA Polymerase III transcription.
  • Restriction Sites: Avoids sequences that might interfere with cloning.

4. Run and Analyze

Click the Design sgRNAs button. Once processing is complete, explore the results across the following tabs:

• Guide RNAs: A detailed table listing the spacer sequence, PAM, genomic coordinates, and strand.

---

Understanding the Results

Column/Metric	Description
Spacer	The 20bp (for Cas9) or 21bp (for Cas12a) targeting sequence.
Composite Score	A normalized score (0–1) incorporating GC content, on-target affinity (CRISPRater), and sequence penalties.
Cut Site	The exact genomic coordinate where the double-strand break is predicted to occur.
Poly-T	Marked “Yes” if the guide contains a terminator sequence (avoid these for U6-driven vectors).

---

Visual Guidance

Figure 12: Configure your CRISPR experiment by selecting the appropriate nuclease and target region (CDS or full gene).

Figure 13: The Quality Scores tab allows you to visualize and select the highest-performing guides based on composite modeling.

---

Data Export

You can download your designs in multiple formats:

• CSV: Full data for spreadsheets.

Methylation

Purpose

The Methylation module is designed to investigate the epigenetic landscape of Artemisia annua. It allows you to explore genes involved in three primary regulatory mechanisms: RNA methylation (m6A), DNA methylation (5-methyladenosine), and histone modification (Histone-H3).

---

Step-by-Step Instructions

1. Choose Your Exploration Method

You can navigate the epigenetic data using two distinct approaches:

• Method A: Filter by Type:
  • Select a category from the Select Methylation Type dropdown.
  • Use the dynamic Sub-filter to narrow results by functional role (e.g., Writer, Eraser, Reader for m6A) or specific Pfam domains (e.g., SET or PHD domains for Histone-H3).
• Method B: Search by Gene ID:
  • Enter a custom list of up to 100 Gene IDs in the Search by Gene ID(s) box to see if they have known epigenetic associations.

2. Interact with the Main Table

The results are displayed in a comprehensive data table:

• Functional Links: Click on any Pfam ID to open the corresponding entry in the InterPro/Pfam database for detailed protein domain information.
• Selection: Click on one or more rows to highlight genes for further analysis.
• Action Buttons: Once rows are selected, use the buttons at the bottom:
  • Copy Gene ID(s): Saves the selected IDs to your clipboard.
  • Check Gene Expression: Automatically transfers the selected genes to the Expression tab for tissue-pattern analysis.

3. Analyze Tissue Expression Patterns

In the Expression tab:

• Select Parts: Choose the plant tissues you wish to compare (Root, Stem, Leaf, Flower, Petiole).
• Calculate: Click Calculate Median Expression.
• Heatmap Visualization: The system generates a clustered heatmap using log2-transformed data (\(log_2(TPM + 1)\)).
• Download: Export the high-resolution heatmap as a PNG, JPEG, or PDF.

4. Retrieve Sequences

Switch to the Sequence tab to view detailed transcript information:

• Review the Transcript ID, CDS information, and the full nucleotide sequence.
• The sequence is formatted for readability (100 characters per line).
• Click Download Sequences (Text) to save the data in a FASTA-like format.

---

Understanding the Methylation Types

Category	Description	Sub-filters / Roles
m6A	N6-methyladenosine RNA methylation	Writers, Erasers, Readers
5-methyl	DNA methylation enzymes	DNA_methylase, DNMT1-RFD, TP_methylase
Histone-H3	Proteins modifying Histone H3	PHD and SET domains

---

Visual Guidance

Figure 14: Use the dynamic filters to isolate specific epigenetic regulators like m6A ‘Writers’ or Histone-H3 ‘SET’ domain proteins.

Figure 15: The Expression tab allows you to visualize if specific epigenetic regulators are tissue-specific (e.g., active primarily in the Root or Trichome-rich Leaf).

---

JBrowse Genomic Browser

Purpose

The JBrowse tool provides a high-performance, interactive environment for visualizing the Artemisia annua genome. Based on the latest genome assembly and Mikado annotations developed for this project, it allows you to explore gene structures, regulatory regions, and sequence features in their precise genomic context.

---

Step-by-Step Instructions

1. Launch the Browser: Navigate to Tools > JBrowse and click the Open button to initialize the browser interface.
2. Select Tracks: * Open the Track Selector in the sidebar.
   • Enable the Artemisia Annotation (GFF3) and Genome Sequence (FASTA) tracks to populate the linear genome view.
3. Navigate & Zoom: Use the search bar to jump to a specific coordinate or Gene ID. Zoom in until individual gene features—such as Exons (thick blocks), Introns (thin lines), and UTRs—become visible.
4. Inspect Features: Click on any gene or CDS feature to open a detailed information window.
5. Extract Sequences:
   • In the feature details window, click Show Feature Sequence.
   • Custom Flanking Regions: By default, the tool allows you to extract the sequence with 1000 bp upstream and downstream.
   • Modify Range: Click the wheel icon next to the extraction dropdown to adjust these boundaries (e.g., to 500 bp for promoter analysis or 2000 bp for broader context).

---

Example Workflow: Promoter Extraction

To extract a promoter sequence for a gene of interest:

1. Search for the gene ID and zoom in until the gene structure is clear.
2. Click the gene feature to open the details panel.
3. Select Show Feature Sequence.
4. Adjust the upstream extraction range to 2000 bp via the wheel icon to capture the full promoter region.
5. Copy the resulting sequence for downstream motif analysis or primer design.

---

Visual Guidance

Figure 16: Exploring genomic features in JBrowse. (a) Tracks loaded in the linear view. (b) Feature details window. (c) Sequence extraction tools. (d) Configuration of flanking regions.

---

BLAST

Purpose

The BLAST module is essential for cross-referencing external data with our database. It allows you to perform sequence similarity searches to identify corresponding Artemisia-DB Gene IDs. This is particularly useful if you have a sequence from a different species or are working with IDs from older Artemisia annua genome versions and need to find their updated counterparts in our latest annotation.

---

Step-by-Step Instructions

1. Input Your Query: Navigate to Tools > BLAST. You can provide your sequence in two ways:
   • Text Input: Paste a FASTA-formatted sequence (e.g., >My_Gene\nATCG...) directly into the text area.
   • File Upload: Upload a sequence file (up to 2MB) in .fasta or .txt format.
2. Configure Search Parameters: Adjust the thresholds to control the stringency of your search:
   • Minimum Identity (%): The minimum percentage of identical residues required for a match (Default: 80%).
   • E-value: The Expectation value threshold for reporting matches. A lower E-value indicate a more statistically significant hit (Default: 1e-5).
3. Execute: Click the Run BLAST button. The system will align your query against the Artemisia annua genome and the latest Mikado gene models.
4. Review Results: * The results table will display the matching Artemisia-DB Gene IDs, alignment scores, and coverage.
   • Use these Gene IDs to jump to the Expression, Annotation, or CRISPR modules.
   • Click Download CSV to save the alignment summary.

---

Example Workflow: Migrating Old Gene IDs

If you have a sequence associated with a gene ID from a previous publication or older version of the genome:

1. Paste the sequence into the BLAST text box.
2. Set the E-value to 1e-10 for high-stringency matching.
3. Identify the top hit in the results table (e.g., mikado.chr7G1274).
4. Copy this new ID and use it in the Global Expression Viewer to see how that gene is expressed across the six developmental stages of Salvia miltiorrhiza or Artemisia annua.

---

Visual Guidance

Figure 17: The BLAST interface facilitates the translation of external sequences into database-specific Gene IDs, enabling a full multi-omics deep dive.

---

Co-expression Analysis

Purpose

The Co-expression Analysis tool identifies genes that exhibit similar expression patterns across various experimental conditions in Artemisia annua. This module is essential for discovering functional modules, potential regulatory networks, and genes that may be co-regulated within specific metabolic pathways, such as artemisinin biosynthesis.

---

Step-by-Step Instructions

1. Define Your Query:

   • Target Genes: Enter one or more Gene IDs (e.g., mikado.chr1G1335) into the search box to find their co-expressed partners.
   • Correlation Method: Choose between Pearson (linear relationship) or Spearman (rank-based relationship) from the dropdown menu.

2. Filter the Dataset:

   • Bioprojects: Select specific BioProjects or choose ALL to perform a global analysis across the entire database.
   • Plant Parts: Choose specific tissues (e.g., Leaf, Root, Trichome). This list dynamically updates based on your BioProject selection.
   • Sample Metadata: The table on the right will update automatically to show which samples are included in your current filter.

3. Set Statistical Thresholds:

   • Minimum Absolute Correlation: Use the slider (0 to 1) to define the strength of the relationship (\(|r|\)).
   • FDR Rate Threshold: Set the False Discovery Rate to ensure statistical significance.

4. Analyze Results: Click Run Co-expression to generate data across three interactive tabs:

   • Table: A comprehensive list of co-expressed genes, including their correlation values (\(r\)) and significance (FDR).
   • Network Graph: A visual representation of the top 10 co-expressed genes. The nodes represent genes, and the edges represent the strength of their co-expression.
   • GO Enrichment: * Dot Plot: Visualizes significantly enriched Gene Ontology (GO) terms for the identified gene cluster. Hover over the plot and click the camera icon to save it.
     • Enrichment Table: Provides a detailed breakdown of enriched terms. You can refine this list using the Root Node filter or the Max P.adjust slider.

---

Example Workflow: Identifying Pathway Partners

If you want to find genes co-regulated with a key enzyme like ADS (Artemisinic acid synthase):

1. Enter the Gene ID for ADS.
2. Select Spearman correlation and set the threshold to 0.85.
3. Choose Leaf and Trichome tissues, as these are the primary sites of artemisinin production.
4. Run the analysis and check the Network Graph to see the top partners.
5. Review the GO Enrichment tab to see if the co-expressed cluster is enriched for “terpenoid biosynthetic processes.”

---

Visual Guidance

Figure 18: Configure your network analysis by selecting correlation methods and specific tissue types to refine the biological context.

Figure 19: Visualize gene relationships through the Network Graph and interpret the biological significance of the co-expressed cluster via the GO Enrichment dot plot.

---

GO Enrichment Analysis

Purpose

The GO Enrichment Analysis tool allows you to identify overrepresented biological themes within a list of genes. By comparing your target gene set against the entire Artemisia annua genome “universe,” the tool determines which Biological Processes (BP), Molecular Functions (MF), and Cellular Components (CC) are statistically significant.

---

Step-by-Step Instructions

1. Input Your Gene List

Navigate to the GO Enrichment tab. You can provide your gene IDs in two ways:

• Manual Entry: Paste your Mikado Gene IDs (one per line) into the text area.
• File Upload: Upload a .txt file containing your gene list (one ID per line).
• Note: You can combine both methods; the tool will automatically merge the lists and remove duplicates.

2. Execute the Analysis

Click the Run GO Enrichment button. The system will process your list using the Benjamini-Hochberg (FDR) correction method. You can monitor the progress via the Analysis Status box, which will report how many genes were successfully mapped and how many GO terms were identified.

3. Visualize Results (Dot Plot)

An interactive Dot Plot (powered by Plotly) will appear once the analysis is complete:

• Dot Size: Represents the number of genes associated with the GO term.
• Color Gradient: Represents the significance level (\(p.adjust\)). Redder dots indicate higher significance.
• Interactivity: Hover over any dot to see the full GO description and exact statistics.

4. Filter and Refine

Use the controls below the plot to clean up your results:

• Filter by Root Node: Isolate specific categories such as Biological Process, Molecular Function, or Cellular Component.
• Max P.adjust Slider: Filter out less significant results by lowering the \(p\)-value threshold (e.g., set to \(0.05\) or \(0.01\)).

5. Inspect and Download the Data Table

The Results Table at the bottom provides a deep dive into every enriched term:

• Gene Column: View the specific IDs from your input list that matched each GO term.
• Download: Click Download Results Table to save the filtered data as a CSV file for your publication or further research.

---

Key Features Summary

Feature	Description
Statistical Method	Hypergeometric testing via the enricher methodology.
P-value Adjustment	Benjamini-Hochberg (FDR) correction to minimize false positives.
Ontology Scope	Comprehensive coverage of BP, MF, and CC root nodes.
Dynamic Filtering	Real-time table and plot updates based on significance thresholds.

---

Example Workflow: Functional Discovery

If you have identified a cluster of 50 genes that are highly expressed in the Root:

1. Paste those 50 IDs into the Enter Genes box.
2. Click Run GO Enrichment.
3. Review the Dot Plot. If you see terms like “secondary metabolite biosynthetic process” or “phenylpropanoid metabolic process,” it suggests this gene set is actively involved in producing specialized compounds.
4. Filter for Molecular Function to see if specific enzyme activities (e.g., “transferase activity”) are driving this process.

---

Visual Guidance

Figure 20: The Dot Plot summarizes the top enriched terms. Use the camera icon in the top-right of the plot to save the visualization as a PNG.

Figure 21: The results table provides the “Gene_Count” and the specific “Genes” involved in each term, allowing for direct functional cross-referencing.

---

Get Gene Sequences

• Purpose: Retrieve and download gene sequences for specified Artemisia annua gene IDs.
• Steps:
  1. Navigate to Tools > Get Gene Sequences.
  2. Enter gene IDs (one per line) in the text area (e.g., mikado.chr1G1813).
  3. If ≤5 gene IDs are entered, view the sequences in the output area below the text input.
  4. Click Download Sequences to save the sequences as a text file. Note: For >5 gene IDs, sequences are not displayed on-screen but can still be downloaded.

Download

Purpose

The Download menu provides a comprehensive repository for retrieving raw and processed Artemisia annua genomic and transcriptomic data. Whether you need expression matrices for specific tissues, bulk data for entire BioProjects, or core reference files (Genome and Annotation), this section facilitates high-throughput data access.

---

1. Download Data by Tissue

Purpose

This module allows you to extract expression data tailored to specific plant organs across all integrated studies.

Steps

1. Navigate: Go to Download > Download Data by Tissue.
2. Select Criteria: * Tissues: Check one or more boxes (e.g., Flower, Leaf, Root).
   • Measure: Choose between TPM (standardized abundance) or Bias-corrected counts (for differential expression tools).
   • Level: Select Gene-level or Transcript-level resolution.
3. Search: Click the Search button to generate a summary.
4. Review Summary: The right-hand panel will display the number of associated projects, total samples, and the estimated Total Download Size.
5. Download: Click Download Selected File(s) to receive a ZIP archive containing the .tsv files.

---

2. Download Data by Project

Purpose

Best for users interested in the original context of a specific study, this tool provides bulk expression matrices.

Steps

1. Navigate: Go to Download > Download Data by Project.
2. Select Project: Browse the table and click on a row to select a specific BioProject (e.g., studies comparing wild-type and tdd1 mutants).
3. Configure Format: In the “Download Options” box, select your preferred data type (TPM or Counts).
4. Process & Save: Click Download Selected Project Data. The system will dynamically pivot the underlying Parquet data into a wide-format .tsv file with genes as rows and samples as columns.

---

3. Download Reference Files

Purpose

Access the foundational reference files required for local bioinformatics pipelines.

Steps

1. Navigate: Go to Download > Download Files.
2. Select File Type: * Annotation File: Download the latest GFF3 file containing the Mikado gene models.
   • Transcriptome Assembly: Download the compiled FASTA sequences of all transcripts.
   • Genome File: Download the LQ-9_phase0 primary genome assembly in FASTA format.
3. Download: Click the respective buttons to save these files directly to your machine.

---

Understanding the Data Formats

File Type	Extension	Recommended Use
TPM Matrix	.tsv	Comparing expression levels across different genes or tissues.
Count Matrix	.tsv	Input for statistical tools like DESeq2 or EdgeR.
Reference	.gff3 / .fasta	Local mapping, BLAST searches, or phylogenomic analysis.

---

Visual Guidance

Figure 22: The Tissue Download interface provides a real-time summary of the data size and file list before you initiate the ZIP download.

Figure 23: Select an individual study from the project table to retrieve its specific expression matrix.