Python snippets for biology

Requires BioPython to be installed

Open common formats:

.gbk / .gbf / .gb
import SeqIO
with open('/my/path/','r') as file_handle:
    record_dict = SeqIO.to_dict(SeqIO.parse(file_handle, 'gb'))
gbkFile = record_dict[list(record_dict.keys())[0]]
# the above is if there is only 1 record in the GBK file
# if multiple files are in the record, such as a genome&plas,
# this will only extract the first record

.fa / .fasta / .fna

import SeqIO
fastaList = list(SeqIO.parse("path/file.fasta", "fasta"))

Write common formats:

.gbk / .gbf / .gb

import SeqIO
SeqIO.write(seqRecordObj_or_list,'/my/path/file.gbk', 'gb')

.fa / .fasta / .fna

import SeqIO
SeqIO.write(seqRecordObj_or_list,'/my/path/file.fasta', 'fasta')

local BLASTing from Python

Also requires NCBI BLAST command line software, a local BLAST database, and Pandas

from Bio.Seq import Seq
from Bio import SeqIO
from Bio.SeqRecord import SeqRecord
import subprocess
from tempfile import NamedTemporaryFile
import pandas as pd

def BLAST(seq, db = 'nr_db', type = "p"):
    # 'seq' is a sequence (as a str) of a protein or nucleotide sequence
    # 'db' points to location of local BLAST database
    # 'type' specifies the type of BLAST (e.g. 'n', 'p', 'x', etc)

    query = NamedTemporaryFile()
    tmp = NamedTemporaryFile()
    SeqIO.write(SeqRecord(Seq(seq), id="temp"),, "fasta")

    flags = 'qstart qend sseqid sframe pident slen sseq length sstart send qlen'
    # 'flags' specifies the specific outputs for 'output format 6' in the BLAST CL software
    extras = "-max_target_seqs 20000 -culling_limit 10 -perc_identity 75"
    # 'extras' are further flags that can be called on the CL #the actual CL BLAST
            (f'blast{type} -query {} -out {} '
            f'-db {db} {extras} -word_size {str(wordsize)} -outfmt "6 {flags}"'),

    with open(, "r") as file_handle:  #opens BLAST file
        align = file_handle.readlines()


    df = pd.DataFrame([ele.split() for ele in align], columns = flags.split())
    df = df.apply(pd.to_numeric, errors='ignore')
    # puts the output of BLAST into a tidy Pandas dataframe

    return df

-- Matt McGuffie - 2020-12-03


Edit | Attach | Watch | Print version | History: r2 < r1 | Backlinks | Raw View | More topic actions...

 Barrick Lab  >  ComputationList  >  ProtocolsPythonSnippets

Contributors to this topic Edit topic JeffreyBarrick, MattMcGuffie
Topic revision: r1 - 2020-12-03 - 21:40:14 - Main.MattMcGuffie
This site is powered by the TWiki collaboration platform Powered by Perl This site is powered by the TWiki collaboration platformCopyright ©2021 Barrick Lab contributing authors. Ideas, requests, problems? Send feedback