brenda-database

BRENDA 酶数据库

概述

BRENDA（BRaunschweig ENzyme DAtabase，布伦瑞克酶数据库）是全球最全面的酶信息系统，包含 80,000 多条酶条目，涵盖所有已分类的酶（EC 编号）。数据库收录了来自 200,000 多篇文献的 700 多万个实验测定动力学参数（Km、Vmax、kcat、Ki、抑制常数）、底物特异性数据、辅因子需求、组织表达信息，以及特定生物的酶变体。可通过基于 SOAP 的 Web 服务（Python zeep 库）以编程方式访问，学术用途可免费注册。

适用场景

• 获取特定酶与底物组合的动力学参数（Km、kcat、Vmax、Ki）
• 比较不同生物或酶突变体的动力学参数
• 查找某个 EC 编号对应的天然底物、抑制剂和辅因子
• 构建需要 Michaelis-Menten 参数的代谢模拟动力学模型
• 查找特定酶的结构数据（推荐 pH 值、最佳温度）
• 交叉查询 EC 编号与 UniProt 登录号及生物分类信息
• 代谢网络模拟请使用 cobrapy-metabolic-modeling；代谢物结构请使用 hmdb-database

前提条件

• Python 软件包：zeep（SOAP 客户端）、pandas、requests
• 数据要求：EC 编号（例如 1.1.1.1）、酶名称或生物名称
• 环境：需要互联网连接；在 https://www.brenda-enzymes.org/register.php 免费注册学术账号
• 速率限制：未明确说明限制；避免批量自动查询；请求之间应加入休眠间隔

pip install zeep pandas requests
# Register at https://www.brenda-enzymes.org/register.php to obtain API credentials

快速入门

from zeep import Client

WSDL = "https://www.brenda-enzymes.org/soap/brenda_zeep.wsdl"
client = Client(WSDL)

EMAIL = "[email protected]"
PASSWORD_SHA256 = "your_sha256_hashed_password"  # Use hashlib.sha256

# Get Km values for lactate dehydrogenase (EC 1.1.1.27) and pyruvate
ec_number = "1.1.1.27"
params = (EMAIL, PASSWORD_SHA256,
          f"ecNumber*{ec_number}", "substrate*pyruvate", "", "", "", "", "")
result = client.service.getKmValue(*params)
print(f"Km values for LDH with pyruvate: {len(result)} records")
for r in result[:3]:
    print(f"  Km={r.kmValue} {r.kmValueMaximum or ''} mM | org: {r.organism} | PMID: {r.literature}")

核心 API

查询 1：酶-底物组合的 Km 值

获取特定酶和底物的 Michaelis 常数（Km）值。

from zeep import Client
import hashlib, pandas as pd

WSDL = "https://www.brenda-enzymes.org/soap/brenda_zeep.wsdl"
client = Client(WSDL)

EMAIL = "[email protected]"
PASSWORD = "your_password"
PASSWORD_SHA256 = hashlib.sha256(PASSWORD.encode()).hexdigest()

def get_km_values(ec_number, substrate=""):
    """Retrieve Km values for an EC number, optionally filtered by substrate."""
    substrate_param = f"substrate*{substrate}" if substrate else ""
    params = (EMAIL, PASSWORD_SHA256,
              f"ecNumber*{ec_number}", substrate_param, "", "", "", "", "")
    return client.service.getKmValue(*params)

# Km for glucokinase (EC 2.7.1.2) with glucose
results = get_km_values("2.7.1.2", substrate="glucose")
print(f"Km (glucose, glucokinase): {len(results)} measurements")

rows = []
for r in results[:10]:
    rows.append({
        "km_value": r.kmValue,
        "km_max": r.kmValueMaximum,
        "unit": "mM",
        "organism": r.organism,
        "commentary": r.commentary[:80] if r.commentary else "",
        "pmid": r.literature,
    })
df = pd.DataFrame(rows)
print(df.to_string(index=False))

# Get ALL Km values (all substrates) for an EC number
all_km = get_km_values("1.1.1.1")  # Alcohol dehydrogenase
print(f"\nAlcohol dehydrogenase - total Km records: {len(all_km)}")
substrate_counts = {}
for r in all_km:
    sub = r.substrate or "unknown"
    substrate_counts[sub] = substrate_counts.get(sub, 0) + 1
top_substrates = sorted(substrate_counts.items(), key=lambda x: -x[1])[:5]
print("Top substrates by measurement count:")
for sub, cnt in top_substrates:
    print(f"  {sub}: {cnt} measurements")

查询 2：kcat（周转数）值

获取酶的催化速率常数（kcat）。

from zeep import Client
import hashlib, pandas as pd

WSDL = "https://www.brenda-enzymes.org/soap/brenda_zeep.wsdl"
client = Client(WSDL)

EMAIL = "[email protected]"
PASSWORD_SHA256 = hashlib.sha256("your_password".encode()).hexdigest()

def get_kcat_values(ec_number, substrate=""):
    substrate_param = f"substrate*{substrate}" if substrate else ""
    params = (EMAIL, PASSWORD_SHA256,
              f"ecNumber*{ec_number}", substrate_param, "", "", "", "", "")
    return client.service.getTurnoverNumber(*params)

results = get_kcat_values("1.1.1.27")  # Lactate dehydrogenase
print(f"kcat records for LDH: {len(results)}")

rows = []
for r in results[:10]:
    rows.append({
        "kcat": r.turnoverNumber,
        "unit": "1/s",
        "substrate": r.substrate,
        "organism": r.organism,
    })
df = pd.DataFrame(rows)
print(df.head())

查询 3：底物和产物

获取酶的天然底物和产物。

from zeep import Client
import hashlib, pandas as pd

WSDL = "https://www.brenda-enzymes.org/soap/brenda_zeep.wsdl"
client = Client(WSDL)

EMAIL = "[email protected]"
PASSWORD_SHA256 = hashlib.sha256("your_password".encode()).hexdigest()

def get_substrates_products(ec_number):
    params = (EMAIL, PASSWORD_SHA256,
              f"ecNumber*{ec_number}", "", "", "", "", "", "")
    return client.service.getSubstrates(*params)

results = get_substrates_products("4.2.1.1")  # Carbonic anhydrase
print(f"Substrates for carbonic anhydrase (EC 4.2.1.1):")
substrates_seen = set()
for r in results[:10]:
    if r.substrate not in substrates_seen:
        print(f"  {r.substrate} | organism: {r.organism}")
        substrates_seen.add(r.substrate)

# Get inhibitors
def get_inhibitors(ec_number):
    params = (EMAIL, PASSWORD_SHA256,
              f"ecNumber*{ec_number}", "", "", "", "", "", "")
    return client.service.getInhibitors(*params)

inhibitors = get_inhibitors("4.2.1.1")
print(f"\nInhibitors of carbonic anhydrase: {len(inhibitors)} records")
inhib_names = list(set(r.inhibitor for r in inhibitors if r.inhibitor))
print("Sample inhibitors:", inhib_names[:8])

查询 4：特定生物的酶数据

按生物筛选并查询动力学参数。

from zeep import Client
import hashlib

WSDL = "https://www.brenda-enzymes.org/soap/brenda_zeep.wsdl"
client = Client(WSDL)

EMAIL = "[email protected]"
PASSWORD_SHA256 = hashlib.sha256("your_password".encode()).hexdigest()

def get_km_by_organism(ec_number, organism):
    params = (EMAIL, PASSWORD_SHA256,
              f"ecNumber*{ec_number}", "", f"organism*{organism}", "", "", "", "")
    return client.service.getKmValue(*params)

# Human GAPDH Km values
human_km = get_km_by_organism("1.2.1.12", "Homo sapiens")
print(f"Human GAPDH (EC 1.2.1.12) Km values: {len(human_km)} records")
for r in human_km[:5]:
    print(f"  Substrate: {r.substrate:30s} Km={r.kmValue} mM")

查询 5：pH 和温度最佳值

获取酶的最佳 pH 和温度数据。

from zeep import Client
import hashlib

WSDL = "https://www.brenda-enzymes.org/soap/brenda_zeep.wsdl"
client = Client(WSDL)

EMAIL = "[email protected]"
PASSWORD_SHA256 = hashlib.sha256("your_password".encode()).hexdigest()

def get_ph_optimum(ec_number):
    params = (EMAIL, PASSWORD_SHA256,
              f"ecNumber*{ec_number}", "", "", "", "", "", "")
    return client.service.getPhOptimum(*params)

def get_temp_optimum(ec_number):
    params = (EMAIL, PASSWORD_SHA256,
              f"ecNumber*{ec_number}", "", "", "", "", "", "")
    return client.service.getTemperatureOptimum(*params)

ec = "3.4.21.4"  # Trypsin
ph_data = get_ph_optimum(ec)
temp_data = get_temp_optimum(ec)

print(f"Trypsin (EC {ec}):")
ph_values = [r.phOptimum for r in ph_data[:10] if r.phOptimum]
temp_values = [r.temperatureOptimum for r in temp_data[:10] if r.temperatureOptimum]
if ph_values:
    print(f"  pH optima: {sorted(ph_values)}")
if temp_values:
    print(f"  Temperature optima (°C): {sorted(temp_values)}")

查询 6：EC 编号与 UniProt 交叉引用

将 EC 编号映射到 UniProt 登录号。

from zeep import Client
import hashlib

WSDL = "https://www.brenda-enzymes.org/soap/brenda_zeep.wsdl"
client = Client(WSDL)

EMAIL = "[email protected]"
PASSWORD_SHA256 = hashlib.sha256("your_password".encode()).hexdigest()

def get_uniprot_accessions(ec_number):
    params = (EMAIL, PASSWORD_SHA256,
              f"ecNumber*{ec_number}", "", "", "", "", "", "")
    return client.service.getUniprotAccession(*params)

results = get_uniprot_accessions("1.1.1.27")  # LDH
print(f"UniProt accessions for LDH (EC 1.1.1.27):")
seen = set()
for r in results[:10]:
    acc = r.uniprotAccessionNumber
    org = r.organism
    if acc and acc not in seen:
        print(f"  {acc:12s} ({org})")
        seen.add(acc)

关键概念

SOAP 接口和身份验证

BRENDA 通过 WSDL 定义使用 SOAP（而非 REST）。Python 库 zeep 会解析 WSDL 并生成类型化方法调用。身份验证需要使用 SHA256 哈希后的密码（而非明文）。每个服务方法都接收 (email, password_sha256, param1, param2, ..., "") 参数，其中字段筛选器以竖线分隔。

EC 编号分类

酶学委员会（EC）编号采用 X.X.X.X 格式，每一级都表示反应类别（氧化还原酶=1、转移酶=2、水解酶=3、裂合酶=4、异构酶=5、连接酶=6、转位酶=7）。BRENDA 按 EC 编号整理所有数据。

常见工作流

工作流 1：提取用于代谢建模的动力学参数

目标：针对代谢通路中的一组酶，提取 Km 和 kcat 值，为动力学模型设置参数。

from zeep import Client
import hashlib, pandas as pd, time

WSDL = "https://www.brenda-enzymes.org/soap/brenda_zeep.wsdl"
client = Client(WSDL)

EMAIL = "[email protected]"
PASSWORD_SHA256 = hashlib.sha256("your_password".encode()).hexdigest()

# Glycolysis enzymes
enzymes = {
    "Hexokinase": "2.7.1.1",
    "Phosphoglucose isomerase": "5.3.1.9",
    "Phosphofructokinase": "2.7.1.11",
    "Aldolase": "4.1.2.13",
}

rows = []
for name, ec in enzymes.items():
    params = (EMAIL, PASSWORD_SHA256, f"ecNumber*{ec}", "", "organism*Homo sapiens", "", "", "", "")
    try:
        km_results = client.service.getKmValue(*params)
        kcat_results = client.service.getTurnoverNumber(*params)
        km_vals = [r.kmValue for r in km_results if r.kmValue]
        kcat_vals = [r.turnoverNumber for r in kcat_results if r.turnoverNumber]
        rows.append({
            "enzyme": name,
            "ec": ec,
            "n_km_records": len(km_vals),
            "km_median_mM": pd.Series(km_vals).median() if km_vals else None,
            "n_kcat_records": len(kcat_vals),
            "kcat_median_1_s": pd.Series(kcat_vals).median() if kcat_vals else None,
        })
    except Exception as e:
        rows.append({"enzyme": name, "ec": ec, "error": str(e)})
    time.sleep(0.5)

df = pd.DataFrame(rows)
df.to_csv("glycolysis_kinetics.csv", index=False)
print(df.to_string(index=False))

工作流 2：比较酶家族中的抑制剂

目标：比较一组相关酶的抑制剂概况，以确定药物发现的优先级。

from zeep import Client
import hashlib, pandas as pd, time
from collections import Counter

WSDL = "https://www.brenda-enzymes.org/soap/brenda_zeep.wsdl"
client = Client(WSDL)

EMAIL = "[email protected]"
PASSWORD_SHA256 = hashlib.sha256("your_password".encode()).hexdigest()

# Carbonic anhydrase isoforms
ca_ecs = ["4.2.1.1"]  # All carbonic anhydrases share this EC

rows = []
for ec in ca_ecs:
    params = (EMAIL, PASSWORD_SHA256, f"ecNumber*{ec}", "", "", "", "", "", "")
    try:
        inhib_results = client.service.getInhibitors(*params)
        for r in inhib_results[:30]:
            rows.append({
                "ec": ec,
                "inhibitor": r.inhibitor,
                "organism": r.organism,
                "ic50": r.ic50Value if hasattr(r, "ic50Value") else None,
            })
    except Exception as e:
        print(f"Error for {ec}: {e}")
    time.sleep(0.5)

df = pd.DataFrame(rows)
print(f"Total inhibitor records: {len(df)}")
top_inhib = Counter(df["inhibitor"]).most_common(10)
print("\nMost reported inhibitors:")
for inhib, count in top_inhib:
    print(f"  {inhib}: {count} records")

关键参数

参数	模块	默认值	范围 / 选项	作用
ecNumber*	所有查询	必填	EC 编号字符串	按酶类别筛选
substrate*	Km、kcat	—	底物名称	按底物筛选
organism*	所有查询	—	物种名称	按生物体筛选（例如，"Homo sapiens"）
commentary*	所有查询	—	文本子串	按注释文本筛选
ligandStructureId*	基于化合物的查询	—	BRENDA 结构 ID	按配体 ID 筛选
Password	身份验证	必填	SHA256 哈希值	身份验证（hashlib.sha256）

最佳实践

1. 正确哈希密码：BRENDA 要求提供纯文本密码的 SHA256 哈希值，而不是密码本身。使用 hashlib.sha256("your_password".encode()).hexdigest()。

2. 将凭据存储在环境变量中：切勿将凭据硬编码。使用 os.environ["BRENDA_EMAIL"] 和 os.environ["BRENDA_PASSWORD"] 这类模式。

3. 在查询之间添加 time.sleep()：BRENDA 的 SOAP 服务可能运行缓慢；对于大型批量查询，每次间隔 0.5–1 秒，以避免超时。

4. 建模时按生物体筛选：不同生物体的动力学参数差异可能很大；务必按与你的模型相关的生物体筛选（例如，organism*Homo sapiens）。

5. 使用中位数 / IQR 汇总参数：同一底物的多项文献测量值通常相差一个数量级；使用中位数 + IQR 而非均值来概括分布。

常用示例

示例：获取某个 EC 编号的所有底物

适用场景：了解酶的底物范围，以进行通路分析。

from zeep import Client
import hashlib

WSDL = "https://www.brenda-enzymes.org/soap/brenda_zeep.wsdl"
client = Client(WSDL)

EMAIL = "[email protected]"
PASSWORD_SHA256 = hashlib.sha256("your_password".encode()).hexdigest()

ec = "1.1.1.1"  # Alcohol dehydrogenase
params = (EMAIL, PASSWORD_SHA256, f"ecNumber*{ec}", "", "", "", "", "", "")
results = client.service.getSubstrates(*params)
substrates = list(set(r.substrate for r in results if r.substrate))
print(f"Substrates of EC {ec} ({len(substrates)} unique): {substrates[:10]}")

示例：kcat/Km 效率比

适用场景：根据 BRENDA 数据计算催化效率（kcat/Km）。

import pandas as pd

# After fetching km_results and kcat_results for same ec + substrate
# km_values = [r.kmValue for r in km_results if r.kmValue]  # mM
# kcat_values = [r.turnoverNumber for r in kcat_results if r.turnoverNumber]  # 1/s

km_median = 0.1   # mM (example)
kcat_median = 500  # s^-1 (example)

efficiency = kcat_median / (km_median * 1e-3)  # Convert Km to M
print(f"Catalytic efficiency (kcat/Km): {efficiency:.2e} M^-1 s^-1")
# Diffusion limit ≈ 10^8-10^9 M^-1 s^-1

示例：根据酶名称查找 EC 编号

适用场景：将酶的常用名称解析为 EC 编号，以便查询 BRENDA。

from zeep import Client
import hashlib

WSDL = "https://www.brenda-enzymes.org/soap/brenda_zeep.wsdl"
client = Client(WSDL)

EMAIL = "[email protected]"
PASSWORD_SHA256 = hashlib.sha256("your_password".encode()).hexdigest()

# Search enzymes by name
params = (EMAIL, PASSWORD_SHA256, "recommendedName*lactate dehydrogenase", "", "", "", "", "", "")
results = client.service.getEcNumber(*params)
print(f"EC numbers for 'lactate dehydrogenase':")
for r in results[:5]:
    print(f"  EC {r.ecNumber}: {r.recommendedName}")

故障排除

问题	原因	解决方案
zeep.exceptions.Fault: Authentication failed	密码错误或 SHA256 格式不正确	确保使用 hashlib.sha256(password.encode()).hexdigest()——必须用 hexdigest，而不是 digest
结果列表为空	未找到 EC 编号或底物	检查 EC 格式（X.X.X.X，以点分隔）；先尝试不使用底物筛选
SOAP 超时	查询量大或连接缓慢	使用生物体筛选来缩小结果集；设置 zeep 传输超时时间
结果字段出现 AttributeError	此查询不提供该字段	使用 getattr(r, "field", None) 安全访问可选字段
常见酶响应缓慢	数据集很大（TP53 = 10K+ 条记录）	按生物体和底物筛选，以减少数据传输
zeep.exceptions.TransportError	网络连接问题	检查 VPN，并在 30 秒后重试

相关技能

• cobrapy-metabolic-modeling — 使用 BRENDA 中的 Km/Vmax 作为动力学约束进行基于约束的代谢建模
• hmdb-database — 获取 BRENDA 底物的代谢物结构和生物学背景
• kegg-database — 通过 EC 编号交叉引用，了解 BRENDA 酶的通路背景
• uniprot-protein-database — 获取 BRENDA 中酶的蛋白质序列和结构数据

参考资料

• BRENDA 数据库 — BRENDA 主门户和手动搜索
• BRENDA Web 服务文档 — SOAP API 参考和参数说明
• zeep Python SOAP 客户端 — 用于 SOAP Web 服务的 Python 库
• Chang 等人（2021）BRENDA 更新 — BRENDA 2021 数据库更新论文