OHMind-Chem MCP Server

Comprehensive chemistry tools for molecular informatics, SMILES operations, property calculations, and chemistry-aware web search.

Table of Contents

• Overview
• Server Configuration
• Tools Reference
• Usage Examples
• Troubleshooting
• See Also

Overview

The OHMind-Chem MCP server provides 17 tools for molecular informatics operations. It handles SMILES notation, molecular property calculations, name conversions, and chemistry-aware web search.

Server Details

Property	Value
Server Name	OHMind-Chem
Entry Point	python -m OHMind_agent.MCP.Chem.server
Tool Count	17
Dependencies	RDKit, Transformers, Requests

Capabilities

• SMILES validation and canonicalization
• Molecular property calculations
• Chemical name conversions (IUPAC, common names, CAS)
• Functional group identification
• Molecular similarity analysis
• Structure visualization
• Chemistry-aware web search

Server Configuration

Starting the Server

# stdio transport (default)
python -m OHMind_agent.MCP.Chem.server --transport stdio

# HTTP transport
python -m OHMind_agent.MCP.Chem.server --transport streamable-http --port 8101

Environment Variables

Variable	Purpose	Required
PYTHONPATH	Path to OHMind project	Yes
TAVILY_API_KEY	API key for web search	For WebSearch
RDKIT_QUIET	Suppress RDKit warnings	No

mcp.json Configuration

{
  "OHMind-Chem": {
    "command": "python",
    "args": ["-m", "OHMind_agent.MCP.Chem.server", "--transport", "stdio"],
    "env": {
      "PYTHONPATH": "/path/to/OHMind",
      "TAVILY_API_KEY": "tvly-your-api-key",
      "RDKIT_QUIET": "1"
    }
  }
}

Tools Reference

SMILES Operations

MoleculeSmilesCheck

Validate a molecular SMILES string and explain syntax issues.

Parameters: | Name | Type | Required | Description | |——|——|———-|————-| | smiles | string | Yes | SMILES string to validate |

Returns: Validation result with error details if invalid

Example:

{
  "smiles": "CC(C)(C)c1ccc(cc1)C(O)c2ccccc2"
}

SmilesCanonicalization

Canonicalize a SMILES string to standard form.

Parameters: | Name | Type | Required | Description | |——|——|———-|————-| | smiles | string | Yes | Input SMILES string | | isomeric | boolean | No | Include stereochemistry (default: true) | | atom_maps | boolean | No | Preserve atom mapping (default: false) |

Returns: Canonical SMILES string

ReactionSmilesCheck

Validate reaction SMILES format (reactants>reagents>products).

Parameters: | Name | Type | Required | Description | |——|——|———-|————-| | reaction_smiles | string | Yes | Reaction SMILES string |

Returns: Validation result

Molecular Properties

MoleculeWeight

Compute exact molecular weight from SMILES.

Parameters: | Name | Type | Required | Description | |——|——|———-|————-| | smiles | string | Yes | Input SMILES string |

Returns: Molecular weight in g/mol

Example:

{
  "smiles": "CC(=O)OC1=CC=CC=C1C(=O)O"
}
// Returns: 180.16 g/mol (Aspirin)

MoleculeAtomCount

Count atoms of each element in a molecule.

Parameters: | Name | Type | Required | Description | |——|——|———-|————-| | smiles | string | Yes | Input SMILES string |

Returns: Dictionary of element counts

Smiles2Formula

Get molecular formula from SMILES.

Parameters: | Name | Type | Required | Description | |——|——|———-|————-| | smiles | string | Yes | Input SMILES string |

Returns: Molecular formula (e.g., C9H8O4)

FunctionalGroups

Detect common functional groups from SMILES.

Parameters: | Name | Type | Required | Description | |——|——|———-|————-| | smiles | string | Yes | Input SMILES string |

Returns: List of identified functional groups

Detected Groups:

• Alcohol, Aldehyde, Ketone
• Carboxylic acid, Ester, Amide
• Amine (primary, secondary, tertiary)
• Ether, Sulfonate, Phosphate
• Aromatic rings, Halides

MoleculeSimilarity

Calculate Tanimoto similarity between two molecules.

Parameters: | Name | Type | Required | Description | |——|——|———-|————-| | smiles1 | string | Yes | First molecule SMILES | | smiles2 | string | Yes | Second molecule SMILES |

Returns: Similarity score (0-1) and qualitative description

Name Conversions

Iupac2Smiles

Convert IUPAC name to SMILES (via PubChem).

Parameters: | Name | Type | Required | Description | |——|——|———-|————-| | iupac_name | string | Yes | IUPAC chemical name |

Returns: SMILES string

Smiles2Iupac

Convert SMILES to IUPAC name.

Parameters: | Name | Type | Required | Description | |——|——|———-|————-| | smiles | string | Yes | Input SMILES string |

Returns: IUPAC name

Name2Smiles

Convert common/brand/chemical name to SMILES.

Parameters: | Name | Type | Required | Description | |——|——|———-|————-| | name | string | Yes | Chemical name |

Returns: SMILES string

Smiles2Cas

Look up CAS number from SMILES (via PubChem).

Parameters: | Name | Type | Required | Description | |——|——|———-|————-| | smiles | string | Yes | Input SMILES string |

Returns: CAS registry number

Format Conversions

Selfies2Smiles

Convert SELFIES to SMILES.

Parameters: | Name | Type | Required | Description | |——|——|———-|————-| | selfies | string | Yes | SELFIES string |

Returns: SMILES string

Smiles2Selfies

Convert SMILES to SELFIES.

Parameters: | Name | Type | Required | Description | |——|——|———-|————-| | smiles | string | Yes | Input SMILES string |

Returns: SELFIES string

Visualization

Smiles2Image

Render a 2D structure image from SMILES.

Parameters: | Name | Type | Required | Description | |——|——|———-|————-| | smiles | string | Yes | Input SMILES string | | width | integer | No | Image width in pixels (default: 300) | | height | integer | No | Image height in pixels (default: 300) |

Returns: Path to generated image file

AI-Powered Tools

MoleculeCaptioner

Generate natural-language description of a molecule using MolT5 model.

Parameters: | Name | Type | Required | Description | |——|——|———-|————-| | smiles | string | Yes | Input SMILES string |

Returns: Natural language description

Note: Requires Transformers library and model download on first use.

Web Search

WebSearch

Chemistry-aware web search using Tavily API.

Parameters: | Name | Type | Required | Description | |——|——|———-|————-| | query | string | Yes | Search query |

Returns: Search results with relevant chemistry information

Note: Requires TAVILY_API_KEY environment variable.

Usage Examples

Property and Functional Group Analysis

Using your OHMind-Chem MCP tools, analyze this molecule with SMILES 
`CC(=O)OC1=CC=CC=C1C(=O)O`.

1) Check that the SMILES is valid.
2) Report its molecular weight and formula.
3) List the functional groups and briefly explain what each group 
   typically does chemically.

Name to SMILES to Image Workflow

With your OHMind-Chem tools, convert the IUPAC name "2-propanol" to SMILES, 
verify the SMILES is valid, and then generate a small 300×300 structure image.

Return the SMILES, a short caption describing the molecule, and a link or 
reference to the generated image file.

Similarity and RAG-Assisted Design

Use your Chem MCP tools to compare lidocaine and procaine by SMILES.

1) Compute their structural similarity.
2) Use your web-search tool to summarize key differences in their 
   pharmacological profiles.
3) Propose a new candidate local anesthetic scaffold and justify it.

Molecular Weight Calculation

What is the molecular weight of the compound with SMILES CN+(C)C1=CC=CC=C1?

Troubleshooting

Common Issues

RDKit Not Found

ImportError: No module named 'rdkit'

Solution: Install RDKit via conda:

conda install -c conda-forge rdkit

Transformers Not Found

ImportError: No module named 'transformers'

Solution: Install Transformers:

pip install transformers

WebSearch Fails

Error: TAVILY_API_KEY not set

Solution: Set the API key:

export TAVILY_API_KEY="tvly-your-api-key"

Invalid SMILES

Error: Could not parse SMILES

Solution: Use MoleculeSmilesCheck to validate SMILES before other operations.

Debug Mode

Run the server with verbose logging:

PYTHONPATH=/path/to/OHMind \
  python -m OHMind_agent.MCP.Chem.server --transport stdio 2>&1 | tee chem_debug.log

See Also

• MCP Server Reference - Overview of all servers
• Chemistry Agent - Agent documentation
• Configuration Reference - Full configuration

---

Last updated: 2025-12-22 | OHMind v1.0.0