我是Python新手。我正在尝试将包含8个染色体序列的基因组fasta文件作为输入,针对查询序列进行爆破并提取前50个命中。 Hre是我的代码:
from Bio import SeqIO
from Bio.Seq import Seq
from Bio.Blast import NCBIXML
from Bio.Blast.Applications import NcbiblastnCommandline
from Bio.SeqRecord import SeqRecord
from Bio.Alphabet import IUPAC
import os
db = list(SeqIO.parse('genome.fasta', 'fasta'))
for i in range(len(db)):
print "Chromosome_"+str(i+1)
print('Doing the BLAST and retrieving the results...')
output_handle = open("dbn.fasta", "w")
nseq=db[i]
SeqIO.write(nseq, output_handle, "fasta")
output_handle.close()
os.system("makeblastdb -in dbn.fasta -dbtype nucl -out dbn")
result_handle= NcbiblastnCommandline(query="sequence.fasta", db = "dbn", outfmt= 5, out="my_blast_"+str(i)+".xml", evalue = 0.00001, task = "megablast")
stderr, stdout = result_handle()
E_VALUE_THRESH = 1e-100
c=0
print "Extracting hits for Chromosome"+str(i+1)+"in another file"
for record in NCBIXML.parse(open("my_blast_"+str(i)+".xml")):
if record.alignments:
for align in record.alignments:
for hsp in align.hsps:
if hsp.expect < E_VALUE_THRESH:
if c>50: break
start=hsp.sbjct_start
end=hsp.sbjct_end
newSeq = db[i].seq[:end]
newSeq = newSeq[start:]
ids=db[i].id+str(c)
myseq[c]=SeqRecord(Seq(newSeq,IUPAC.DNA))
myseq[c].id=str(ids)
c=c+1
output_handler = open("example_"+str(i)+".fasta", "w")
SeqIO.write(myseq, output_handler, "fasta")
output_handler.close()
它在第33行给出了一个错误:
AttributeError: 'module' object has no attribute 'DNA'
我试图删除IUPAC.DNA,制作一行:
myseq[c]=SeqRecord(Seq(newSeq)
它给我一个类型错误:
TypeError: The sequence data given to a Seq object should be a string (not another Seq object etc)
有没有办法解决这个问题,有没有其他方法可以创建一个多快速列表变量?
答案 0 :(得分:0)
也许你必须出于某种原因解析python中的fasta,但如果没有,为什么不直接保留文件并从命令行调用blast?然后解析python中的结果。
import subprocess
subprocess.call('makeblastdb -dbtype nucl -in dbn.fasta -out dbn.DATABASE')
subprocess.call('blastn -db dbn.DATABASE -query genome.fasta -outfmt 7 -out blast_results')
for line in open('blast_results') # outfmt 7 is tabular output
if not line.startswith('#'):
# Get top 50 hits (per chromosome I assume)
答案 1 :(得分:0)
感谢大家的建议。我能够从染色体中提取序列,并将它们直接写入另一个文件,以&#39;&gt;&#39;
开头ids=db[i]+"_"+str(hsp.sbjct_start)+"_"+str(hsp.sbjct_end)
f.write(">"+str(ids) + "\n")
f.write(str(newSeq) + "\n")