name: bio-entrez-fetch description: Retrieve records from NCBI databases using Biopython Bio.Entrez. Use when downloading sequences, fetching GenBank records, getting document summaries, or parsing NCBI data into Biopython objects. tool_type: python primary_tool: Bio.Entrez measurable_outcome: Execute skill workflow successfully with valid output within 15 minutes. allowed-tools:

• read_file
• run_shell_command

Entrez Fetch

Retrieve records from NCBI databases using Biopython's Entrez module (EFetch, ESummary utilities).

Required Setup

from Bio import Entrez

Entrez.email = 'your.email@example.com'  # Required by NCBI
Entrez.api_key = 'your_api_key'          # Optional, raises rate limit 3->10 req/sec

Core Functions

Entrez.efetch() - Retrieve Full Records

Fetch complete records in various formats from any NCBI database.

# Fetch GenBank record by ID
handle = Entrez.efetch(db='nucleotide', id='NM_007294', rettype='gb', retmode='text')
genbank_text = handle.read()
handle.close()

# Fetch FASTA sequence
handle = Entrez.efetch(db='nucleotide', id='NM_007294', rettype='fasta', retmode='text')
fasta_text = handle.read()
handle.close()

# Fetch multiple records
handle = Entrez.efetch(db='nucleotide', id='NM_007294,NM_000059', rettype='fasta', retmode='text')

Key Parameters:

db

'nucleotide'

'protein'

'pubmed'

id

'NM_007294'

'123,456,789'

rettype

'fasta'

'gb'

'abstract'

retmode

'text'

'xml'

retstart

0

retmax

20

WebEnv

query_key

Common Return Types by Database

Nucleotide/Protein:

'fasta'

'text'

'gb'

'text'

'gp'

'text'

'gbwithparts'

'text'

'seqid'

'text'

'acc'

'text'

PubMed:

'abstract'

'text'

'medline'

'text'

'xml'

'xml'

Gene:

'gene_table'

'text'

'xml'

'xml'

Entrez.esummary() - Document Summaries

Get brief summaries without downloading full records. Faster than efetch.

# Get summary for nucleotide record
handle = Entrez.esummary(db='nucleotide', id='NM_007294')
record = Entrez.read(handle)
handle.close()

summary = record[0]  # First (only) record
print(f"Title: {summary['Title']}")
print(f"Length: {summary['Length']}")
print(f"Organism: {summary['Organism']}")

Common Summary Fields:

# Nucleotide/Protein
summary['Title']          # Record title/description
summary['Caption']        # Short identifier
summary['Length']         # Sequence length
summary['Organism']       # Source organism
summary['TaxId']          # Taxonomy ID
summary['AccessionVersion']  # Full accession.version

# PubMed
summary['Title']          # Article title
summary['AuthorList']     # Authors
summary['Source']         # Journal
summary['PubDate']        # Publication date
summary['DOI']            # Digital Object Identifier

Parsing with Biopython

Parse into SeqRecord Objects

from Bio import Entrez, SeqIO

Entrez.email = 'your.email@example.com'

# Parse GenBank into SeqRecord
handle = Entrez.efetch(db='nucleotide', id='NM_007294', rettype='gb', retmode='text')
record = SeqIO.read(handle, 'genbank')
handle.close()

print(f"ID: {record.id}")
print(f"Length: {len(record.seq)}")
print(f"Features: {len(record.features)}")

# Parse FASTA into SeqRecord
handle = Entrez.efetch(db='nucleotide', id='NM_007294', rettype='fasta', retmode='text')
record = SeqIO.read(handle, 'fasta')
handle.close()

Parse Multiple Records

# Fetch multiple as FASTA
handle = Entrez.efetch(db='nucleotide', id='NM_007294,NM_000059,NM_000546', rettype='fasta', retmode='text')
records = list(SeqIO.parse(handle, 'fasta'))
handle.close()

for record in records:
    print(f"{record.id}: {len(record.seq)} bp")

Parse XML with Entrez.read()

# For structured data, use XML mode
handle = Entrez.efetch(db='gene', id='672', retmode='xml')
records = Entrez.read(handle)
handle.close()

# Navigate nested structure
gene = records[0]
print(f"Gene: {gene['Entrezgene_gene']['Gene-ref']['Gene-ref_locus']}")

Code Patterns

Fetch Sequence by Accession

from Bio import Entrez, SeqIO

Entrez.email = 'your.email@example.com'

def fetch_sequence(accession, db='nucleotide'):
    handle = Entrez.efetch(db=db, id=accession, rettype='fasta', retmode='text')
    record = SeqIO.read(handle, 'fasta')
    handle.close()
    return record

seq = fetch_sequence('NM_007294')
print(f"{seq.id}: {seq.seq[:50]}...")

Fetch GenBank with Features

def fetch_genbank(accession):
    handle = Entrez.efetch(db='nucleotide', id=accession, rettype='gb', retmode='text')
    record = SeqIO.read(handle, 'genbank')
    handle.close()
    return record

gb = fetch_genbank('NM_007294')
for feature in gb.features:
    if feature.type == 'CDS':
        print(f"CDS: {feature.location}")
        print(f"Product: {feature.qualifiers.get('product', ['?'])[0]}")

Fetch PubMed Abstract

def fetch_abstract(pmid):
    handle = Entrez.efetch(db='pubmed', id=pmid, rettype='abstract', retmode='text')
    abstract = handle.read()
    handle.close()
    return abstract

abstract = fetch_abstract('35412348')
print(abstract)

Get Record Summaries

def get_summaries(db, ids):
    if isinstance(ids, list):
        ids = ','.join(ids)
    handle = Entrez.esummary(db=db, id=ids)
    records = Entrez.read(handle)
    handle.close()
    return records

summaries = get_summaries('nucleotide', ['NM_007294', 'NM_000059'])
for s in summaries:
    print(f"{s['Caption']}: {s['Title'][:50]}... ({s['Length']} bp)")

Search Then Fetch

# Search for records
handle = Entrez.esearch(db='nucleotide', term='human[orgn] AND insulin[gene] AND mRNA[fkey]', retmax=5)
search_results = Entrez.read(handle)
handle.close()

ids = search_results['IdList']

# Fetch the sequences
handle = Entrez.efetch(db='nucleotide', id=','.join(ids), rettype='fasta', retmode='text')
records = list(SeqIO.parse(handle, 'fasta'))
handle.close()

for record in records:
    print(f"{record.id}: {len(record.seq)} bp")

Fetch Protein by Gene ID

# Search gene database
handle = Entrez.esearch(db='gene', term='BRCA1[sym] AND human[orgn]')
result = Entrez.read(handle)
handle.close()
gene_id = result['IdList'][0]

# Get linked protein IDs
handle = Entrez.elink(dbfrom='gene', db='protein', id=gene_id)
links = Entrez.read(handle)
handle.close()

protein_ids = [link['Id'] for link in links[0]['LinkSetDb'][0]['Link'][:3]]

# Fetch proteins
handle = Entrez.efetch(db='protein', id=','.join(protein_ids), rettype='fasta', retmode='text')
proteins = list(SeqIO.parse(handle, 'fasta'))
handle.close()

Save Fetched Records to File

def download_sequences(ids, output_file, db='nucleotide', format='fasta'):
    handle = Entrez.efetch(db=db, id=','.join(ids), rettype=format, retmode='text')
    with open(output_file, 'w') as out:
        out.write(handle.read())
    handle.close()

download_sequences(['NM_007294', 'NM_000059'], 'brca_genes.fasta')

Common Errors

HTTPError 400

HTTPError 429

ValueError

ExpatError

retmode='text'

Decision Tree

Need to retrieve NCBI records?
├── Need full sequence?
│   └── Use efetch with rettype='fasta'
├── Need sequence + annotations?
│   └── Use efetch with rettype='gb' (GenBank)
├── Just need metadata (length, organism)?
│   └── Use esummary (faster)
├── Need PubMed abstract?
│   └── Use efetch with rettype='abstract'
├── Need structured data for parsing?
│   └── Use efetch with retmode='xml' + Entrez.read()
├── Downloading many records?
│   └── See batch-downloads skill
└── Need records from multiple databases?
    └── See entrez-link skill first

Related Skills

• entrez-search - Find record IDs before fetching
• entrez-link - Find related records in other databases
• batch-downloads - Download large numbers of records efficiently
• sequence-io/read-sequences - Parse downloaded sequences with SeqIO