初次提交

2025-05-09 14:16:33 +08:00
commit 3a50afeec4
56 changed files with 9224 additions and 0 deletions
--- a/test_tools/chemistry/test_pubchem.py
+++ b/test_tools/chemistry/test_pubchem.py
@@ -0,0 +1,123 @@
+"""
+Test script for PubChem tools
+
+This script tests the search_pubchem_advanced function from the chemistry_mcp module.
+"""
+import sys
+sys.path.append('/home/ubuntu/sas0/lzy/multi_mcp_server')
+import asyncio
+
+from sci_mcp.chemistry_mcp.pubchem_tools.pubchem_tools import _search_by_formula
+
+
+from sci_mcp.chemistry_mcp import search_pubchem_advanced
+
+async def test_search_by_name():
+    """Test searching compounds by name"""
+    print("\n=== Testing search by name ===")
+    result = await search_pubchem_advanced(name="Aspirin")
+    print(result)
+
+async def test_search_by_smiles():
+    """Test searching compounds by SMILES notation"""
+    print("\n=== Testing search by SMILES ===")
+    # SMILES for Caffeine
+    result = await search_pubchem_advanced(smiles="CN1C=NC2=C1C(=O)N(C(=O)N2C)C")
+    print(result)
+
+async def test_search_by_formula():
+    """Test searching compounds by molecular formula"""
+    print("\n=== Testing search by formula ===")
+    # Formula for Aspirin
+    result = await search_pubchem_advanced(formula="C9H8O4", max_results=2)
+    print(result)
+
+async def test_complex_formula():
+    """Test searching with a more complex formula that might cause timeout"""
+    print("\n=== Testing complex formula search ===")
+    # A more complex formula that might return many results
+    result = await search_pubchem_advanced(
+        formula="C6H12O6",  # Glucose and isomers
+        max_results=5
+    )
+    print(result)
+
+async def test_complex_molecules():
+    """Test searching for complex molecules with rich molecular features"""
+    print("\n=== Testing complex molecules with rich features ===")
+
+    # 1. Paclitaxel (Taxol) - Complex anticancer drug with many rotatable bonds and H-bond donors/acceptors
+    print("\n--- Testing Paclitaxel (anticancer drug) ---")
+    result = await search_pubchem_advanced(name="Paclitaxel")
+    print(result)
+
+    # 2. Vancomycin - Complex antibiotic with many H-bond donors/acceptors
+    print("\n--- Testing Vancomycin (antibiotic) ---")
+    result = await search_pubchem_advanced(name="Vancomycin")
+    print(result)
+
+    # 3. Cholesterol - Steroid with complex ring structure
+    print("\n--- Testing Cholesterol (steroid) ---")
+    result = await search_pubchem_advanced(name="Cholesterol")
+    print(result)
+
+    # 4. Ibuprofen - Common NSAID with rotatable bonds
+    print("\n--- Testing Ibuprofen (NSAID) ---")
+    result = await search_pubchem_advanced(name="Ibuprofen")
+    print(result)
+
+    # 5. Amoxicillin - Antibiotic with multiple functional groups
+    print("\n--- Testing Amoxicillin (antibiotic) ---")
+    result = await search_pubchem_advanced(name="Amoxicillin")
+    print(result)
+
+async def test_molecules_by_smiles():
+    """Test searching for complex molecules using SMILES notation"""
+    print("\n=== Testing complex molecules by SMILES ===")
+
+    # 1. Atorvastatin (Lipitor) - Cholesterol-lowering drug with complex structure
+    print("\n--- Testing Atorvastatin (Lipitor) ---")
+    result = await search_pubchem_advanced(
+        smiles="CC(C)C1=C(C(=C(C=C1)C(C)C)C2=CC(=C(C=C2)F)F)C(CC(CC(=O)O)O)NC(=O)C3=CC=C(C=C3)F"
+    )
+    print(result)
+
+    # 2. Morphine - Opioid with multiple rings and H-bond features
+    print("\n--- Testing Morphine ---")
+    result = await search_pubchem_advanced(
+        smiles="CN1CCC23C4C1CC5=C2C(=C(C=C5)O)OC3C(C=C4)O"
+    )
+    print(result)
+
+async def test_invalid_search():
+    """Test searching with invalid parameters"""
+    print("\n=== Testing invalid search ===")
+    # No parameters provided
+    result = await search_pubchem_advanced()
+    print(result)
+
+    # Invalid SMILES
+    print("\n=== Testing invalid SMILES ===")
+    result = await search_pubchem_advanced(smiles="INVALID_SMILES_STRING")
+    print(result)
+
+async def run_all_tests():
+    """Run all test functions"""
+    await test_search_by_name()
+    await test_search_by_smiles()
+    await test_search_by_formula()
+    await test_complex_formula()
+    # await test_complex_molecules()
+    # await test_molecules_by_smiles()
+    #await test_invalid_search()
+    # from sci_mcp.chemistry_mcp.pubchem_tools.pubchem_tools import _search_by_name
+    # compounds=await _search_by_formula('C6H12O6')
+    # print(compounds[0])
+if __name__ == "__main__":
+    print("Testing PubChem search tools...")
+    asyncio.run(run_all_tests())
+    print("\nAll tests completed.")
+    # import pubchempy
+    # compunnds = pubchempy.get_compounds('Aspirin', 'name')
+    # print(compunnds[0].to_dict())
+
--- a/test_tools/chemistry/test_rdkit.py
+++ b/test_tools/chemistry/test_rdkit.py
@@ -0,0 +1,159 @@
+"""
+Test script for RDKit tools.
+
+This script tests the functionality of the RDKit tools implemented in the
+sci_mcp/chemistry_mcp/rdkit_tools module.
+"""
+import sys
+import os
+
+# Add the project root directory to the Python path
+sys.path.append('/home/ubuntu/sas0/lzy/multi_mcp_server')
+
+from sci_mcp.chemistry_mcp.rdkit_tools.rdkit_tools import (
+    calculate_molecular_properties,
+    calculate_drug_likeness,
+    calculate_topological_descriptors,
+    generate_molecular_fingerprints,
+    calculate_molecular_similarity,
+    analyze_molecular_structure,
+    generate_molecular_conformer,
+    identify_scaffolds,
+    convert_between_chemical_formats,
+    standardize_molecule,
+    enumerate_stereoisomers,
+    perform_substructure_search
+)
+
+def test_molecular_properties():
+    """Test the calculation of molecular properties."""
+    print("Testing calculate_molecular_properties...")
+    # Aspirin
+    smiles = "CC(=O)OC1=CC=CC=C1C(=O)O"
+    result = calculate_molecular_properties(smiles)
+    print(result)
+    print("-" * 80)
+
+def test_drug_likeness():
+    """Test the calculation of drug-likeness properties."""
+    print("Testing calculate_drug_likeness...")
+    # Aspirin
+    smiles = "CC(=O)OC1=CC=CC=C1C(=O)O"
+    result = calculate_drug_likeness(smiles)
+    print(result)
+    print("-" * 80)
+
+def test_topological_descriptors():
+    """Test the calculation of topological descriptors."""
+    print("Testing calculate_topological_descriptors...")
+    # Aspirin
+    smiles = "CC(=O)OC1=CC=CC=C1C(=O)O"
+    result = calculate_topological_descriptors(smiles)
+    print(result)
+    print("-" * 80)
+
+def test_molecular_fingerprints():
+    """Test the generation of molecular fingerprints."""
+    print("Testing generate_molecular_fingerprints...")
+    # Aspirin
+    smiles = "CC(=O)OC1=CC=CC=C1C(=O)O"
+    result = generate_molecular_fingerprints(smiles, fingerprint_type="morgan")
+    print(result)
+    print("-" * 80)
+
+def test_molecular_similarity():
+    """Test the calculation of molecular similarity."""
+    print("Testing calculate_molecular_similarity...")
+    # Aspirin and Ibuprofen
+    smiles1 = "CC(=O)OC1=CC=CC=C1C(=O)O"  # Aspirin
+    smiles2 = "CC(C)CC1=CC=C(C=C1)C(C)C(=O)O"  # Ibuprofen
+    result = calculate_molecular_similarity(smiles1, smiles2)
+    print(result)
+    print("-" * 80)
+
+def test_molecular_structure():
+    """Test the analysis of molecular structure."""
+    print("Testing analyze_molecular_structure...")
+    # Aspirin
+    smiles = "CC(=O)OC1=CC=CC=C1C(=O)O"
+    result = analyze_molecular_structure(smiles)
+    print(result)
+    print("-" * 80)
+
+def test_molecular_conformer():
+    """Test the generation of molecular conformers."""
+    print("Testing generate_molecular_conformer...")
+    # Aspirin
+    smiles = "CC(=O)OC1=CC=CC=C1C(=O)O"
+    result = generate_molecular_conformer(smiles)
+    print(result)
+    print("-" * 80)
+
+def test_scaffolds():
+    """Test the identification of molecular scaffolds."""
+    print("Testing identify_scaffolds...")
+    # Aspirin
+    smiles = "CC(=O)OC1=CC=CC=C1C(=O)O"
+    result = identify_scaffolds(smiles)
+    print(result)
+    print("-" * 80)
+
+def test_format_conversion():
+    """Test the conversion between chemical formats."""
+    print("Testing convert_between_chemical_formats...")
+    # Aspirin
+    smiles = "CC(=O)OC1=CC=CC=C1C(=O)O"
+    result = convert_between_chemical_formats(smiles, "smiles", "inchi")
+    print(result)
+    print("-" * 80)
+
+def test_standardize_molecule():
+    """Test the standardization of molecules."""
+    print("Testing standardize_molecule...")
+    # Betaine with charges
+    smiles = "C[N+](C)(C)CC(=O)[O-]"
+    result = standardize_molecule(smiles)
+    print(result)
+    print("-" * 80)
+
+def test_stereoisomers():
+    """Test the enumeration of stereoisomers."""
+    print("Testing enumerate_stereoisomers...")
+    # 3-penten-2-ol (has both a stereocenter and a stereobond)
+    smiles = "CC(O)C=CC"
+    result = enumerate_stereoisomers(smiles)
+    print(result)
+    print("-" * 80)
+
+def test_substructure_search():
+    """Test the substructure search."""
+    print("Testing perform_substructure_search...")
+    # Aspirin, search for carboxylic acid group
+    smiles = "CC(=O)OC1=CC=CC=C1C(=O)O"
+    pattern = "C(=O)O"
+    result = perform_substructure_search(smiles, pattern)
+    print(result)
+    print("-" * 80)
+
+def main():
+    """Run all tests."""
+    print("Testing RDKit tools...\n")
+
+    # Uncomment the tests you want to run
+    test_molecular_properties()
+    test_drug_likeness()
+    test_topological_descriptors()
+    test_molecular_fingerprints()
+    test_molecular_similarity()
+    test_molecular_structure()
+    test_molecular_conformer()
+    test_scaffolds()
+    test_format_conversion()
+    test_standardize_molecule()
+    test_stereoisomers()
+    test_substructure_search()
+
+    print("All tests completed.")
+
+if __name__ == "__main__":
+    main()
--- a/test_tools/chemistry/test_rxn.py
+++ b/test_tools/chemistry/test_rxn.py
@@ -0,0 +1,289 @@
+"""
+测试RXN工具函数模块
+
+此模块包含用于测试rxn_tools模块中化学反应预测和分析工具函数的测试用例。
+"""
+import asyncio
+import sys
+import os
+from pathlib import Path
+from rich.console import Console
+
+# 添加项目根目录到Python路径
+sys.path.append('/home/ubuntu/sas0/lzy/multi_mcp_server')
+
+from sci_mcp.chemistry_mcp.rxn_tools.rxn_tools import (
+    predict_reaction_outcome_rxn,
+    predict_reaction_topn_rxn,
+    predict_reaction_properties_rxn,
+    extract_reaction_actions_rxn
+)
+
+# 创建控制台对象用于格式化输出
+console = Console()
+
+
+async def test_predict_reaction_outcome():
+    """测试反应结果预测功能"""
+    console.print("[bold cyan]测试反应结果预测功能[/bold cyan]")
+
+    # 使用固定参数：溴和蒽的反应
+    fixed_reactants = "BrBr.c1ccc2cc3ccccc3cc2c1"
+    console.print(f"固定反应物: {fixed_reactants}")
+
+    try:
+        result = await predict_reaction_outcome(fixed_reactants)
+        console.print("[green]结果:[/green]")
+        console.print(result)
+    except Exception as e:
+        console.print(f"[bold red]错误:[/bold red] {str(e)}")
+
+
+async def test_predict_reaction_topn():
+    """测试多产物预测功能"""
+    console.print("\n[bold cyan]测试多产物预测功能[/bold cyan]")
+
+    # 测试1：单个反应（字符串格式）
+    fixed_reactants = "C=CC=O.CN"  # 丙烯醛和甲胺
+    fixed_topn = 2
+    console.print(f"测试1 - 单个反应（字符串格式）")
+    console.print(f"固定反应物: {fixed_reactants}")
+    console.print(f"固定预测产物数量: {fixed_topn}")
+
+    try:
+        result = await predict_reaction_topn(fixed_reactants, fixed_topn)
+        console.print("[green]结果:[/green]")
+        console.print(result)
+    except Exception as e:
+        console.print(f"[bold red]错误:[/bold red] {str(e)}")
+
+    # 测试2：单个反应（列表格式）
+    fixed_reactants_list = ["BrBr", "c1ccc2cc3ccccc3cc2c1"]  # 溴和蒽
+    console.print(f"\n测试2 - 单个反应（列表格式）")
+    console.print(f"固定反应物: {fixed_reactants_list}")
+    console.print(f"固定预测产物数量: {fixed_topn}")
+
+    try:
+        result = await predict_reaction_topn(fixed_reactants_list, fixed_topn)
+        console.print("[green]结果:[/green]")
+        console.print(result)
+    except Exception as e:
+        console.print(f"[bold red]错误:[/bold red] {str(e)}")
+
+    # 测试3：多个反应（列表的列表格式）
+    fixed_reactants_batch = [
+        ["BrBr", "c1ccc2cc3ccccc3cc2c1"],  # 溴和蒽
+        ["BrBr", "c1ccc2cc3ccccc3cc2c1CCO"]  # 溴和修饰的蒽
+    ]
+    console.print(f"\n测试3 - 多个反应（列表的列表格式）")
+    console.print(f"固定反应物批量: {fixed_reactants_batch}")
+    console.print(f"固定预测产物数量: {fixed_topn}")
+
+    try:
+        result = await predict_reaction_topn(fixed_reactants_batch, fixed_topn)
+        console.print("[green]结果:[/green]")
+        console.print(result)
+    except Exception as e:
+        console.print(f"[bold red]错误:[/bold red] {str(e)}")
+        console.print(f"[bold red]错误:[/bold red] {str(e)}")
+
+
+async def test_predict_retrosynthesis():
+    """测试逆合成分析功能"""
+    console.print("\n[bold cyan]测试逆合成分析功能[/bold cyan]")
+
+    # 使用固定参数：阿司匹林的逆合成分析
+    fixed_target_molecule = "CC(=O)OC1=CC=CC=C1C(=O)O"  # 阿司匹林
+    fixed_max_steps = 1
+    console.print(f"固定目标分子: {fixed_target_molecule}")
+    console.print(f"固定最大步骤数: {fixed_max_steps}")
+
+    try:
+        result = await predict_retrosynthesis(fixed_target_molecule, fixed_max_steps)
+        console.print("[green]结果:[/green]")
+        console.print(result)
+    except Exception as e:
+        console.print(f"[bold red]错误:[/bold red] {str(e)}")
+
+
+async def test_predict_biocatalytic_retrosynthesis():
+    """测试生物催化逆合成分析功能"""
+    console.print("\n[bold cyan]测试生物催化逆合成分析功能[/bold cyan]")
+
+    # 使用固定参数：一个可能适合酶催化的分子
+    fixed_target_molecule = "OC1C(O)C=C(Br)C=C1"
+    console.print(f"固定目标分子: {fixed_target_molecule}")
+
+    try:
+        result = await predict_biocatalytic_retrosynthesis(fixed_target_molecule)
+        console.print("[green]结果:[/green]")
+        console.print(result)
+    except Exception as e:
+        console.print(f"[bold red]错误:[/bold red] {str(e)}")
+
+
+async def test_predict_reaction_properties():
+    """测试反应属性预测功能"""
+    console.print("\n[bold cyan]测试反应属性预测功能[/bold cyan]")
+
+    # 使用固定参数：原子映射
+    fixed_reaction = "CC(C)S.CN(C)C=O.Fc1cccnc1F.O=C([O-])[O-].[K+].[K+]>>CC(C)Sc1ncccc1F"
+    fixed_property_type = "atom-mapping"
+    console.print(f"固定反应: {fixed_reaction}")
+    console.print(f"固定属性类型: {fixed_property_type}")
+
+    try:
+        result = await predict_reaction_properties(fixed_reaction, fixed_property_type)
+        console.print("[green]结果:[/green]")
+        console.print(result)
+    except Exception as e:
+        console.print(f"[bold red]错误:[/bold red] {str(e)}")
+
+
+async def test_extract_reaction_actions():
+    """测试从文本提取反应步骤功能"""
+    console.print("\n[bold cyan]测试从文本提取反应步骤功能[/bold cyan]")
+
+    # 使用固定参数：从文本描述中提取反应步骤
+    fixed_reaction_text = """To a stirred solution of 7-(difluoromethylsulfonyl)-4-fluoro-indan-1-one (110 mg, 0.42 mmol) in methanol (4 mL) was added sodium borohydride (24 mg, 0.62 mmol). The reaction mixture was stirred at ambient temperature for 1 hour."""
+    console.print(f"固定反应文本: {fixed_reaction_text}")
+
+    try:
+        result = await extract_reaction_actions(fixed_reaction_text)
+        console.print("[green]结果:[/green]")
+        console.print(result)
+    except Exception as e:
+        console.print(f"[bold red]错误:[/bold red] {str(e)}")
+
+
+async def test_all():
+    """测试所有RXN工具函数"""
+    console.print("[bold magenta]===== 开始测试RXN工具函数 =====[/bold magenta]\n")
+
+    # 测试各个功能
+    await test_predict_reaction_outcome()
+    await test_predict_reaction_topn()
+    await test_predict_retrosynthesis()
+    await test_predict_biocatalytic_retrosynthesis()
+    await test_predict_reaction_properties()
+    await test_extract_reaction_actions()
+
+    console.print("\n[bold magenta]===== RXN工具函数测试完成 =====[/bold magenta]")
+
+
+def get_rxn_tool_questions():
+    """
+    获取为RXN工具函数生成的问题列表
+    这些问题设计为能够引导大模型调用相应的工具函数
+
+    Returns:
+        包含工具名称和对应问题的字典
+    """
+    questions = {
+        "predict_reaction_outcome": [
+            "如果我将溴和蒽混合在一起，会形成什么产物？",
+            "乙酸和乙醇反应会生成什么？",
+            "预测一下丙烯醛和甲胺反应的结果"
+        ],
+
+
+        "predict_reaction_topn": [
+            "丙烯醛和甲胺反应可能生成哪几种主要产物？",
+            "预测溴和蒽反应可能的前3个产物",
+            "乙酸和乙醇反应可能有哪些不同的结果？请给出最可能的几种产物"
+        ],
+
+        "predict_retrosynthesis": [
+            "如何合成阿司匹林？请给出可能的合成路线",
+            "对于分子CC(=O)OC1=CC=CC=C1C(=O)O，有哪些可能的合成路径？",
+            "请分析一下布洛芬的可能合成路线"
+        ],
+
+        "predict_biocatalytic_retrosynthesis": [
+            "有没有可能用酶催化合成OC1C(O)C=C(Br)C=C1这个分子？",
+            "请提供一种使用生物催化方法合成对羟基苯甲醇的路线",
+            "我想用酶催化方法合成一些复杂分子，能否分析一下OC1C(O)C=C(Br)C=C1的可能合成路径？"
+        ],
+
+        "predict_reaction_properties": [
+            "在这个反应中：CC(C)S.CN(C)C=O.Fc1cccnc1F.O=C([O-])[O-].[K+].[K+]>>CC(C)Sc1ncccc1F，原子是如何映射的？",
+            "这个反应的产率可能是多少：Clc1ccccn1.Cc1ccc(N)cc1>>Cc1ccc(Nc2ccccn2)cc1",
+            "能分析一下这个反应中原子的去向吗：CC(=O)O.CCO>>CC(=O)OCC"
+        ],
+
+        "extract_reaction_actions": [
+            "能否将这段实验描述转换为结构化的反应步骤？'To a stirred solution of 7-(difluoromethylsulfonyl)-4-fluoro-indan-1-one (110 mg, 0.42 mmol) in methanol (4 mL) was added sodium borohydride (24 mg, 0.62 mmol). The reaction mixture was stirred at ambient temperature for 1 hour.'",
+            "请从这段文本中提取出具体的反应操作步骤：'A solution of benzoic acid (1.0 g, 8.2 mmol) in thionyl chloride (10 mL) was heated under reflux for 2 hours. The excess thionyl chloride was removed under reduced pressure to give benzoyl chloride as a colorless liquid.'",
+            "帮我解析这个实验步骤，提取出关键操作：'The aldehyde (5 mmol) was dissolved in methanol (20 mL) and sodium borohydride (7.5 mmol) was added portionwise at 0°C. The mixture was allowed to warm to room temperature and stirred for 3 hours.'"
+        ]
+    }
+
+    return questions
+
+
+def update_agent_test_questions():
+    """
+    更新agent_test.py中的工具问题字典，添加RXN工具函数的问题
+    """
+    try:
+        # 获取agent_test.py文件路径
+        agent_test_path = Path('/home/ubuntu/sas0/lzy/multi_mcp_server/test_tools/agent_test.py')
+
+        # 读取文件内容
+        with open(agent_test_path, 'r') as f:
+            content = f.read()
+
+        # 获取RXN工具函数的问题
+        rxn_questions = get_rxn_tool_questions()
+
+        # 检查文件中是否已包含RXN工具函数的问题
+        rxn_tools_exist = any(tool in content for tool in rxn_questions.keys())
+
+        if not rxn_tools_exist:
+            # 找到questions字典的结束位置
+            dict_end_pos = content.find('    return questions')
+
+            if dict_end_pos != -1:
+                # 构建要插入的RXN工具函数问题
+                rxn_questions_str = ""
+                for tool_name, questions_list in rxn_questions.items():
+                    rxn_questions_str += f'\n        "{tool_name}": [\n'
+                    for q in questions_list:
+                        rxn_questions_str += f'            "{q}",\n'
+                    rxn_questions_str += '        ],'
+
+                # 在字典结束前插入RXN工具函数问题
+                new_content = content[:dict_end_pos] + rxn_questions_str + content[dict_end_pos:]
+
+                # 写回文件
+                with open(agent_test_path, 'w') as f:
+                    f.write(new_content)
+
+                console.print("[green]成功更新agent_test.py，添加了RXN工具函数的测试问题[/green]")
+            else:
+                console.print("[yellow]无法找到questions字典的结束位置，未更新agent_test.py[/yellow]")
+        else:
+            console.print("[yellow]agent_test.py中已包含RXN工具函数的问题，无需更新[/yellow]")
+
+    except Exception as e:
+        console.print(f"[bold red]更新agent_test.py时出错:[/bold red] {str(e)}")
+
+
+if __name__ == "__main__":
+    # 运行所有测试
+    # asyncio.run(test_all())
+
+    # # 更新agent_test.py中的工具问题
+    # update_agent_test_questions()
+
+     api_key = 'apk-8928522a146c2503f30b16d9909222d7583f412ee8f1049f08d32a089ba88d34'
+     from rxn4chemistry import RXN4ChemistryWrapper
+
+     rxn4chemistry_wrapper = RXN4ChemistryWrapper(api_key=api_key)
+     rxn4chemistry_wrapper.create_project('test_wrapper')
+     response = rxn4chemistry_wrapper.predict_automatic_retrosynthesis(
+    'Brc1c2ccccc2c(Br)c2ccccc12')
+     results = rxn4chemistry_wrapper.get_predict_automatic_retrosynthesis_results(response['prediction_id'])
+     print(results['status'])
+    # NOTE: upon 'SUCCESS' you can inspect the predicted retrosynthetic paths.
+     print(results['retrosynthetic_paths'][0])