rotifer.db.ncbi.entrez.SequenceCursor#

class rotifer.db.ncbi.entrez.SequenceCursor(database='nucleotide', progress=True, tries=3, sleep_between_tries=1, batch_size=20, threads=10, *args, **kwargs)[source]#

Bases: SequenceCursor, SimpleParallelProcessCursor

Fetch annotated sequences from NCBI with the E-utilities.

Sequences are downloaded with EFetch and parsed from GenBank format, the most richly annotated format NCBI provides.

Parameters:

databasestr, default ‘nucleotide’: A valid NCBI sequence database name, such as protein or nucleotide.
progressbool, default True: Whether to print a progress bar.
triesint, default 3: Number of attempts to download data.
sleep_between_triesint, default 1: Number of seconds to wait between download attempts.
batch_sizeint, default 20: Number of accessions per batch. The default may be changed by the module configuration.
threadsint, default 10: Number of simultaneous threads. Capped at 3 without an NCBI API key and at 10 with one.

See also

FastaCursor: faster download without annotations
rotifer.db.ncbi.SequenceCursor: delegator that combines backends

Examples

Fetch a protein sequence:

>>> from rotifer.db.ncbi import entrez
>>> sc = entrez.SequenceCursor(database="protein")
>>> seqrec = sc.fetchall("YP_009724395.1")

Fetch several nucleotide entries:

>>> import sys
>>> from Bio import SeqIO
>>> query = ['CP084314.1', 'NC_019757.1', 'AAHROG010000026.1']
>>> sc = entrez.SequenceCursor(database="nucleotide")
>>> for seqrec in sc.fetchone(query):
...     SeqIO.write(seqrec, sys.stdout, "genbank")

__getitem__(accession)[source]#

Download data for one batch of entries, dictionary style.

Entries that fail with a recoverable error are retried once in a second pass.

Parameters:

accessionstr or iterable of str: NCBI database entry identifiers. A string may contain several accessions separated by commas.

Returns:

Bio.SeqRecord.SeqRecord or list of Bio.SeqRecord.SeqRecord: A single record when a single accession yields a single record, otherwise a list.

fetcher(accession)[source]#

Run EFetch and return the response stream.

Parameters:

accessionstr or iterable of str: Database entry identifiers.

Returns:

file-like: The open EFetch response.

parser(stream, accession)[source]#

Parse an EFetch stream into sequence records.

Parameters:

streamfile-like: Open stream returned by fetcher().
accessionstr or iterable of str: Database entry identifiers, kept for interface compatibility.

Returns:

list of Bio.SeqRecord.SeqRecord: The parsed records.