Bio entrez search

<!-- # COPYRIGHT NOTICE # This file is part of the "Universal Biomedical Skills" project. # Copyright (c) 2026 MD BABU MIA, PhD <[email protected]> # All Rights Reserved. # # This code is proprietary and confidential. # Unauthorized copying of this file, via any medium is strictly prohibited. # # Provenance: Authenticated by MD BABU MIA -->

Entrez Search

Search NCBI databases using Biopython's Entrez module (ESearch, EInfo, EGQuery utilities).

Required Setup

from Bio import Entrez

Entrez.email = '[email protected]'  # Required by NCBI
Entrez.api_key = 'your_api_key'          # Optional, raises rate limit 3->10 req/sec

Core Functions

Entrez.esearch() - Search a Database

Search any NCBI database and get matching record IDs.

handle = Entrez.esearch(db='nucleotide', term='human[orgn] AND BRCA1[gene]')
record = Entrez.read(handle)
handle.close()

print(f"Found {record['Count']} records")
print(f"IDs: {record['IdList']}")  # First 20 IDs by default

Key Parameters:

Parameter	Description	Default
db	Database to search	Required
term	Search query	Required
retmax	Max IDs to return	20
retstart	Starting index (pagination)	0
usehistory	Store results on server	'n'
sort	Sort order	database-specific
datetype	Date field to search	'pdat'
reldate	Records from last N days	None
mindate	Start date (YYYY/MM/DD)	None
maxdate	End date (YYYY/MM/DD)	None

ESearch Result Fields:

record['Count']        # Total matching records (string)
record['IdList']       # List of record IDs
record['RetMax']       # Number of IDs returned
record['RetStart']     # Starting index
record['QueryKey']     # For history server (if usehistory='y')
record['WebEnv']       # For history server (if usehistory='y')
record['TranslationSet']  # Query translations applied
record['QueryTranslation']  # Final translated query

Entrez.einfo() - Database Information

Get information about available databases or specific database fields.

# List all available databases
handle = Entrez.einfo()
record = Entrez.read(handle)
handle.close()
print(record['DbList'])  # ['pubmed', 'protein', 'nucleotide', ...]

# Get info about specific database
handle = Entrez.einfo(db='nucleotide')
record = Entrez.read(handle)
handle.close()

print(f"Description: {record['DbInfo']['Description']}")
print(f"Record count: {record['DbInfo']['Count']}")

# List searchable fields
for field in record['DbInfo']['FieldList']:
    print(f"{field['Name']}: {field['Description']}")

Database Info Fields:

record['DbInfo']['DbName']       # Database name
record['DbInfo']['Description']  # Database description
record['DbInfo']['Count']        # Total records in database
record['DbInfo']['LastUpdate']   # Last update date
record['DbInfo']['FieldList']    # Searchable fields
record['DbInfo']['LinkList']     # Available links to other databases

Entrez.egquery() - Global Query

Search across all NCBI databases simultaneously.

handle = Entrez.egquery(term='CRISPR')
record = Entrez.read(handle)
handle.close()

for result in record['eGQueryResult']:
    if int(result['Count']) > 0:
        print(f"{result['DbName']}: {result['Count']} records")

Search Query Syntax

NCBI uses a specific query syntax:

Field Tags

# Search specific fields using [field_name]
term = 'BRCA1[gene]'                    # Gene name field
term = 'human[orgn]'                    # Organism field
term = 'Homo sapiens[ORGN]'             # Full organism name
term = 'NM_007294[accn]'                # Accession number
term = 'Smith J[auth]'                  # Author (PubMed)
term = 'Nature[jour]'                   # Journal (PubMed)
term = '1000:5000[slen]'                # Sequence length range
term = 'mRNA[fkey]'                     # Feature key

Boolean Operators

term = 'BRCA1 AND human'                # Both terms
term = 'cancer OR tumor'                # Either term
term = 'human NOT mouse'                # Exclude term
term = '(BRCA1 OR BRCA2) AND human'     # Grouping

Date Ranges

# Using date parameters
handle = Entrez.esearch(
    db='pubmed',
    term='CRISPR',
    datetype='pdat',     # Publication date
    mindate='2023/01/01',
    maxdate='2024/12/31'
)

# Or in query string
term = 'CRISPR AND 2024[pdat]'
term = 'CRISPR AND 2023:2024[pdat]'

Wildcards and Phrases

term = 'immun*'                         # Wildcard
term = '"breast cancer"[title]'         # Exact phrase

Common Databases

Database	db value	Common Fields
PubMed	pubmed	[auth], [title], [jour], [pdat]
Nucleotide	nucleotide	[orgn], [gene], [accn], [slen]
Protein	protein	[orgn], [gene], [accn], [molwt]
Gene	gene	[orgn], [sym], [chr]
SRA	sra	[orgn], [platform], [strategy]
Taxonomy	taxonomy	[scin], [comn], [rank]
Assembly	assembly	[orgn], [level], [refseq]

Code Patterns

Basic Search with Pagination

from Bio import Entrez

Entrez.email = '[email protected]'

def search_ncbi(db, term, max_results=100):
    handle = Entrez.esearch(db=db, term=term, retmax=max_results)
    record = Entrez.read(handle)
    handle.close()
    return record['IdList'], int(record['Count'])

ids, total = search_ncbi('nucleotide', 'human[orgn] AND insulin[gene]')
print(f'Retrieved {len(ids)} of {total} total records')

Paginated Search for Large Results

def search_all_ids(db, term, batch_size=10000):
    all_ids = []
    handle = Entrez.esearch(db=db, term=term, retmax=0)
    record = Entrez.read(handle)
    handle.close()
    total = int(record['Count'])

    for start in range(0, total, batch_size):
        handle = Entrez.esearch(db=db, term=term, retstart=start, retmax=batch_size)
        record = Entrez.read(handle)
        handle.close()
        all_ids.extend(record['IdList'])

    return all_ids

Search with History Server (for Large Results)

# Store results on NCBI server for subsequent fetching
handle = Entrez.esearch(db='nucleotide', term='human[orgn] AND mRNA[fkey]', usehistory='y')
record = Entrez.read(handle)
handle.close()

webenv = record['WebEnv']
query_key = record['QueryKey']
total = int(record['Count'])

# Use webenv and query_key with efetch for batch downloads
# See batch-downloads skill for details

Recent Records Only

# Records from last 30 days
handle = Entrez.esearch(db='pubmed', term='CRISPR', reldate=30, datetype='pdat')
record = Entrez.read(handle)
handle.close()

Get Available Fields for a Database

def get_search_fields(db):
    handle = Entrez.einfo(db=db)
    record = Entrez.read(handle)
    handle.close()
    return [(f['Name'], f['Description']) for f in record['DbInfo']['FieldList']]

fields = get_search_fields('nucleotide')
for name, desc in fields[:10]:
    print(f'{name}: {desc}')

Check Query Translation

handle = Entrez.esearch(db='nucleotide', term='human BRCA1')
record = Entrez.read(handle)
handle.close()

# See how NCBI interpreted your query
print(f"Your query was translated to: {record['QueryTranslation']}")
# e.g., '"homo sapiens"[Organism] AND BRCA1[All Fields]'

Common Errors

Error	Cause	Solution
HTTPError 429	Rate limit exceeded	Add delays or use API key
HTTPError 400	Invalid query syntax	Check field names and operators
Empty IdList	No matches or typo	Check QueryTranslation field
RuntimeError	Missing email	Set Entrez.email

Decision Tree

Need to search NCBI?
├── Finding records in one database?
│   └── Use Entrez.esearch()
├── Search across all databases?
│   └── Use Entrez.egquery()
├── Need database field names?
│   └── Use Entrez.einfo(db='database')
├── List all available databases?
│   └── Use Entrez.einfo() (no db argument)
├── Results > 10,000 records?
│   └── Use usehistory='y', then batch fetch
└── Need to fetch actual records?
    └── See entrez-fetch skill

Related Skills

• entrez-fetch - Retrieve full records after searching
• entrez-link - Find related records in other databases
• batch-downloads - Download large result sets efficiently
• geo-data - Search GEO expression datasets (specialized search)
• blast-searches - Search by sequence similarity instead of keywords

<!-- AUTHOR_SIGNATURE: 9a7f3c2e-MD-BABU-MIA-2026-MSSM-SECURE -->