Click on "Install Server".
Wait a few minutes for the server to deploy. Once ready, it will show a "Started" state.
In the chat, type @ followed by the MCP server name and your instructions, e.g., "@BioNext-mcp Analyze pbmc3k.h5ad: quality control, clustering, UMAP, and marker gene identification"
That's it! The server will respond to your query, and you can continue using it as needed.
Here is a step-by-step guide with screenshots.
The simplest way to perform bioinformatics analysis through Claude Desktop - just chat in natural language, no programming required!
δΈζη | English
BioNext-MCP allows you to perform complex bioinformatics analysis through natural language conversations with Claude Desktop, without writing any code!
Simply put:
π£οΈ Tell Claude what data you want to analyze in plain English
π€ Claude automatically generates professional Python analysis scripts
β‘ System automatically executes scripts and displays results
π Get beautiful HTML reports and visualization charts
Related MCP server: BioOpenMCP
Single-cell RNA sequencing (scRNA-seq) - Cell clustering, differential expression, trajectory analysis
Genomics - Variant analysis, annotation, functional enrichment
Transcriptomics - Differential expression, pathway analysis, co-expression networks
Proteomics - Protein identification, quantitative analysis
Multi-omics integration - Data fusion, correlation analysis
Automatic environment setup - Detects Python, auto-installs required packages (pandas, numpy, matplotlib, etc.)
UTF-8 encoding support - Perfect support for international characters
Visualization-first - Automatically generates charts and displays them in HTML reports
Quality assurance - Focuses on code completeness and analysis accuracy
Error handling - Smart diagnosis of issues with solution suggestions
Download Python 3.9 or higher
Make sure to check "Add Python to PATH" during installation
Open command prompt and type:
python --versionIf you see version information, installation was successful!
Download Project
git clone https://github.com/your-username/BioNext-mcp.git
cd BioNext-mcpInstall Dependencies
npm install
npm run buildFind Configuration File
Windows: %APPDATA%\Claude\claude_desktop_config.json
Add Configuration
{
"mcpServers": {
"bioinformatics-workflow": {
"command": "node",
"args": ["D:\\path\\to\\BioNext-mcp\\dist\\index.js"],
"cwd": "D:\\path\\to\\BioNext-mcp",
"env": {
"PROJECT_PATH": "D:\\path\\to\\your\\analysis\\directory"
}
}
}
}Important:
Replace paths with your actual installation paths
Set analysis directory to where you want results saved
Restart Claude Desktop
Describe Your Analysis Needs
I have a single-cell RNA sequencing data file data.h5ad, and I want to perform cell clustering analysis and differential expression analysisClaude will generate analysis scripts and execute them automatically
Get detailed HTML reports including:
Execution results and statistics
Generated charts and visualizations
Complete analysis logs
Please help me analyze this scRNA-seq data:
- File: C:\data\pbmc3k.h5ad
- Need: quality control, normalization, clustering, marker gene identification
- Output: UMAP plot, clustering heatmap, differential expression gene listI have RNA-seq expression matrices from two groups:
- Control group: control_samples.csv
- Treatment group: treatment_samples.csv
- Analysis: differential expression, GO enrichment, KEGG pathway analysis
- Visualization: volcano plot, heatmap, pathway diagramsHelp me explore this gene expression dataset:
- File: gene_expression.csv
- Need: data overview, correlation analysis, PCA analysis
- Generate: statistical summary, correlation heatmap, PCA plotπ Visualization Gallery - Automatically detects and displays generated images
π Interactive Viewing - Click images to zoom and view
π Detailed Logs - Complete execution process records
π Statistical Summary - Script execution status and performance metrics
Reports automatically open in browser after analysis completion
If not auto-opened, manually open the generated HTML file
Q: "Python not found" error? A: Ensure Python is installed and added to PATH environment variable
Q: Package installation fails?
A: System will automatically retry, or manually run pip install package_name
Q: Script execution fails? A:
Check if data file paths are correct
Confirm data format meets requirements
Check error logs for detailed information
Q: No HTML report generated? A: HTML reports are only generated when all scripts execute successfully, fix execution errors first
Q: What data formats are supported? A:
CSV, TSV, Excel files
HDF5 format (.h5, .h5ad)
FASTA, FASTQ sequence files
VCF variant files
Other common bioinformatics formats
β
Good description:
"Analyze single-cell data, perform quality control (filter low-quality cells), normalization, dimensionality reduction (PCA+UMAP), clustering (leiden algorithm), find marker genes for each cluster"
β Vague description:
"Analyze this data"β
Use absolute paths:
"C:\Users\username\data\sample.h5ad"
β Relative paths may fail:
"./data/sample.h5ad"β
Clear output:
"Generate UMAP plot, heatmap, save results to CSV file"
β Unclear:
"Do some visualization"For complex analyses, break into multiple conversations:
First: Data loading and quality control
Second: Normalization and dimensionality reduction
Third: Clustering and visualization
Fourth: Differential analysis
You're ready now! Open Claude Desktop, tell it what data you want to analyze, and let AI handle the complex bioinformatics analysis for you!
GitHub Issues: Report problems or suggest improvements
Documentation: View detailed usage documentation
Examples: Reference example analysis cases
Remember: Describe your analysis needs in natural language, Claude will handle all the technical details for you! π
Analyze user intent and create a bioinformatics workflow plan. This tool helps understand your analysis goals and prepares the workflow structure. After this, ask Claude to generate Python scripts (β€100 lines each), then use execute_claude_script to run them.
| Name | Required | Description | Default |
|---|---|---|---|
| user_request | Yes | The user's bioinformatics analysis request in natural language | |
| data_files | Yes | List of input data file paths (e.g., ["C:\Users\username\data.h5ad", "data2.csv"]). Include full file paths. | |
| additional_context | No | Any additional context or specific requirements |
Does the description disclose side effects, auth requirements, rate limits, or destructive behavior?
No annotations are provided, so the description must disclose behavioral traits. It states the tool 'analyzes' and 'creates a plan', implying read-only, but does not specify side effects, permissions, or output format. The instruction to ask Claude for scripts is meta-guidance, not behavioral transparency.
Agents need to know what a tool does to the world before calling it. Descriptions should go beyond structured annotations to explain consequences.
Is the description appropriately sized, front-loaded, and free of redundancy?
The description is extremely concise: two sentences covering purpose and usage flow. Every sentence earns its place with no redundancy. It is front-loaded with the primary action.
Shorter descriptions cost fewer tokens and are easier for agents to parse. Every sentence should earn its place.
Given the tool's complexity, does the description cover enough for an agent to succeed on first attempt?
Given no output schema and three parameters, the description fails to specify what the tool returns (the workflow plan). The agent lacks information on how to use the tool's output. The mention of script generation is helpful but not sufficient to cover return value expectations.
Complex tools with many parameters or behaviors need more documentation. Simple tools need less. This dimension scales expectations accordingly.
Does the description clarify parameter syntax, constraints, interactions, or defaults beyond what the schema provides?
Schema coverage is 100%, so the description need not repeat parameter details. The description does not add new semantics beyond the schema's descriptions. Baseline of 3 is appropriate as no value is added.
Input schemas describe structure but not intent. Descriptions should explain non-obvious parameter relationships and valid value ranges.
Does the description clearly state what the tool does and how it differs from similar tools?
The description clearly states the tool analyzes user intent and creates a bioinformatics workflow plan. It distinguishes from siblings (debug_workflow, execute_claude_script) by placing itself as the initial planning step. However, it could be more specific about the form of the workflow plan output.
Agents choose between tools based on descriptions. A clear purpose with a specific verb and resource helps agents select the right tool.
Does the description explain when to use this tool, when not to, or what alternatives exist?
The description provides clear usage context: use this tool first to understand goals and prepare workflow, then generate scripts (β€100 lines), then execute them. It implies the tool is for planning before script generation/execution. Explicit 'when not to use' or alternatives are missing, but the flow is well-defined.
Agents often have multiple tools that could apply. Explicit usage guidance like "use X instead of Y when Z" prevents misuse.
Analyze workflow execution results and provide debugging insights
| Name | Required | Description | Default |
|---|---|---|---|
| workflow_id | Yes | The workflow ID to debug | |
| error_context | No | Additional context about the error or issue |
Does the description disclose side effects, auth requirements, rate limits, or destructive behavior?
With no annotations provided, the description bears full responsibility for disclosing behavioral traits. It indicates the tool is analytical ('analyze', 'provide debugging insights'), implying a read-only operation, but does not confirm lack of side effects, required permissions, rate limits, or output format. The minimal disclosure leaves significant ambiguity about its behavior.
Agents need to know what a tool does to the world before calling it. Descriptions should go beyond structured annotations to explain consequences.
Is the description appropriately sized, front-loaded, and free of redundancy?
The description is a single sentence that immediately conveys the tool's purpose without any extraneous words. It is front-loaded and efficient, earning its place with no waste.
Shorter descriptions cost fewer tokens and are easier for agents to parse. Every sentence should earn its place.
Given the tool's complexity, does the description cover enough for an agent to succeed on first attempt?
Given the tool has no output schema and no annotations, the description should compensate by explaining what the debugging insights include (e.g., error logs, step traces) or any other behavioral details. The current description is too brief to cover the tool's complexity, leaving users uninformed about the results and operational context.
Complex tools with many parameters or behaviors need more documentation. Simple tools need less. This dimension scales expectations accordingly.
Does the description clarify parameter syntax, constraints, interactions, or defaults beyond what the schema provides?
The input schema has 100% description coverage, so both parameters (workflow_id, error_context) are already documented. The description does not add any parameter-specific meaning beyond what the schema provides. Per guidelines, baseline 3 is appropriate since schema handles the burden.
Input schemas describe structure but not intent. Descriptions should explain non-obvious parameter relationships and valid value ranges.
Does the description clearly state what the tool does and how it differs from similar tools?
The description clearly states the tool's purpose: 'Analyze workflow execution results and provide debugging insights'. It specifies a verb ('analyze') and a resource ('workflow execution results'), making the core function understandable. However, it does not explicitly distinguish this tool from its siblings (analyze_bioinformatics_task, execute_claude_script), though the names imply different domains.
Agents choose between tools based on descriptions. A clear purpose with a specific verb and resource helps agents select the right tool.
Does the description explain when to use this tool, when not to, or what alternatives exist?
The description provides no guidance on when to use this tool versus alternatives, no prerequisites, and no exclusion criteria. It simply states what the tool does without context for decision-making, leaving the agent without direction on when this tool is appropriate.
Agents often have multiple tools that could apply. Explicit usage guidance like "use X instead of Y when Z" prevents misuse.
π MAIN TOOL: Automatically detect and execute Python scripts generated by Claude LLM for bioinformatics tasks. Features: β Auto-detects Python installation β Provides detailed installation guide if Python missing β Auto-installs required packages (pandas, numpy, biopython, etc.) β Full execution logging and error handling β Script length monitoring (recommends β€100 lines) β HTML report generation with auto-browser opening
| Name | Required | Description | Default |
|---|---|---|---|
| claude_response | Yes | The full response from Claude LLM that may contain Python scripts | |
| workflow_id | No | Optional workflow ID to associate with this execution | |
| execution_context | No | Context about what the script is supposed to do |
Does the description disclose side effects, auth requirements, rate limits, or destructive behavior?
Description discloses several behavioral aspects: auto-detection, installation, package management, logging, script length monitoring, and report generation. However, it does not address potential destructive effects of script execution.
Agents need to know what a tool does to the world before calling it. Descriptions should go beyond structured annotations to explain consequences.
Is the description appropriately sized, front-loaded, and free of redundancy?
The description uses bullet points and emojis which are clear but somewhat verbose. Could be more succinct.
Shorter descriptions cost fewer tokens and are easier for agents to parse. Every sentence should earn its place.
Given the tool's complexity, does the description cover enough for an agent to succeed on first attempt?
The tool lacks an output schema, and the description does not explain what the tool returns to the agent (e.g., execution result, path to report). Missing this information for agent decision-making.
Complex tools with many parameters or behaviors need more documentation. Simple tools need less. This dimension scales expectations accordingly.
Does the description clarify parameter syntax, constraints, interactions, or defaults beyond what the schema provides?
Schema already provides descriptions for all three parameters. The tool description does not add further parameter semantics beyond what is in the schema.
Input schemas describe structure but not intent. Descriptions should explain non-obvious parameter relationships and valid value ranges.
Does the description clearly state what the tool does and how it differs from similar tools?
Description clearly states the tool detects and executes Python scripts for bioinformatics tasks, distinguishing it from sibling tools that analyze tasks or debug workflows.
Agents choose between tools based on descriptions. A clear purpose with a specific verb and resource helps agents select the right tool.
Does the description explain when to use this tool, when not to, or what alternatives exist?
Despite listing features, the description does not provide explicit guidance on when to use this tool versus siblings. Usage context is implied but not stated.
Agents often have multiple tools that could apply. Explicit usage guidance like "use X instead of Y when Z" prevents misuse.
Recent tool additions, removals, and schema changes observed during successful MCP inspections. Dates show when Glama detected each change.
v2.1.0analyze_bioinformatics_taskdebug_workflowexecute_claude_scriptEach tool has a clearly distinct purpose: planning analysis (analyze_bioinformatics_task), executing scripts (execute_claude_script), and debugging (debug_workflow). No functional overlap.
All tool names follow a consistent verb_noun snake_case pattern (e.g., analyze_bioinformatics_task, debug_workflow, execute_claude_script), making them predictable.
With only 3 tools, the set is well-scoped for its bioinformatics workflow automation purposeβeach tool is essential and not excessive.
The tools cover planning, execution, and debuggingβthe core workflow. Minor gap: no explicit data retrieval or result analysis tool, but execution report generation partially addresses this.
Unclaimed servers have limited discoverability.
If you are the server author, to access and configure the admin panel.
Connect Claude to Fathom meeting recordings, transcripts, and summaries
WHOOP recovery, strain, sleep and workouts in Claude via official WHOOP OAuth. Free, open source.
Connect Claude to your Platform7n workspaces β chat, links, and tasks. One-click OAuth.
Connect Claude to your Intervals.icu watch data for fitness, workout review, and plan writing.
We provide all the information about MCP servers via our MCP API.
curl -X GET 'https://glama.ai/api/mcp/v1/servers/Cherine0205/BioNext-mcp'
If you have feedback or need assistance with the MCP directory API, please join our Discord server
