Python 利用Entrez库筛选下载PubMed文献摘要的示例

 import numpy as np
 from Bio import Medline, Entrez # 一般是通过BioPython的Bio.Entrez模块访问Entrez
 from collections import Counter
 
 Entrez.email = "（此处写你自己在官网注册的邮箱账号）" # 应用自己的账号访问NCBI数据库
 
 # 此处需将服务器协议指定为1.0，否则会出现报错。http.client.IncompleteRead: IncompleteRead(0 bytes read)
 # 服务器http协议1.0，而python的是1.1，解决办法就是指定客户端http协议版本
 import http.client
 http.client.HTTPConnection._http_vsn = 10
 http.client.HTTPConnection._http_vsn_str = 'HTTP/1.0'
 
 """
 Entrez 是一个检索系统，可以用其访问NCBI数据库，比如说PubMed，GenBank，GEO等。
 获得有关 global PBDE 的所有文献的PubMed IDs
 """
 # handle_0 = Entrez.esearch(db="pubmed", term="drug therapy[Subheading] AND adverse effects[Subheading] AND humans[MeSH Terms]", retmax=306431)
 handle_0 = Entrez.esearch(db="pubmed", term="drug therapy[MeSH Subheading] AND adverse effects[MeSH Subheading] AND humans[MeSH Terms] AND (2000/01/01[Date - Publication] : 2021/12/31[Date - Publication])",
              ptyp="Review", usehistory="y", retmax=306431)
 record = Entrez.read(handle_0) # 获取检索条件的所有文献
 idlist = record["IdList"] # 提取出文献id
 print ("Total: ", record["Count"])
 No_Papers = len(idlist)  # 共306431篇文献 2000-01-01:2021-12-31
 webenv = record['WebEnv']
 query_key = record['QueryKey']
 
 total = No_Papers
 step = 1300
 print("Result items:", total)
 with open("./Data_PubMed/PBDE1.txt", 'w') as f:
   for start in range(0, total, step):
     print("Download record %i to %i" % (start + 1, int(start + step)))
     handle_1 = Entrez.efetch(db="pubmed", retstart=start, rettype="medline", retmode="text",
                 retmax=step, webenv=webenv, query_key=query_key) # 获取上述所有文献的PubMed IDs
     records = Medline.parse(handle_1)
     records = list(records) # 将迭代器转换至列表（list）
     for index in np.arange(len(records)):
       id = records[index].get("PMID", "")
       title = records[index].get("TI", "")
       title = title.replace('[', '').replace('].', '') # 若提取的标题出现[].符号，则去除
       abstract = records[index].get("AB", "")
       f.write(id)
       f.write("\n")
       f.write(title)
       f.write("\n")
       f.write(abstract)
       f.write("\n")