Talk:SeqIO

From Biopython
(Difference between revisions)
Jump to: navigation, search
(How to specify SeqRecord Alphabet: Bug 2443 is fixed now)
 
(11 intermediate revisions by 3 users not shown)
Line 1: Line 1:
 +
Hello everyone,
 +
 +
* If you want help on using Bio.SeqIO, please join the discussion mailing list (see [[mailing lists]]).
 +
* If you think you've found a bug, please report it on [http://bugzilla.open-bio.org/ bugzilla].
 +
* If you have a nice example you'd like to add, I would encourage you to join the discussion mailing list, but posting the code here should be fine.
 +
 +
Thanks,
 +
 +
[[User:Maubp|Peter]]
 +
 +
== New example, Random Fragments ==
 +
 +
I added a example into the output section, but now I wonder if this is the best place to put it. Maybe it should be under "examples". --[[User:Sbassi|Sbassi]] 01:20, 27 August 2007 (EDT)
 +
 +
[[User:Maubp|Peter]] Nice idea - and yes, I would have put it under examples too.  I've moved it and edited it too - I wanted the style to match the rest of the page, and also I think there was a possible problem in your randomising code where a sequence might run over the end of the record. I opted to simplify the example - I hope you're still happy with it.
 +
 +
== EMBL Problem? ==
 +
 
Hi, I am a newbie of Biopython.
 
Hi, I am a newbie of Biopython.
 
I tested Bio.SeqIO.parse on EMBL formated miRNA.dat from the microRNA Registry. But I got an error. Is the EMBL format supported fully?
 
I tested Bio.SeqIO.parse on EMBL formated miRNA.dat from the microRNA Registry. But I got an error. Is the EMBL format supported fully?
 
As:
 
As:
  
 +
<python>
 
>>>from Bio import SeqIO
 
>>>from Bio import SeqIO
 
>>>handle = open('/home/liang/Desktop/miRNA.dat','rU')
 
>>>handle = open('/home/liang/Desktop/miRNA.dat','rU')
Line 8: Line 27:
 
>>>first_record=record_iterator.next()
 
>>>first_record=record_iterator.next()
 
Traceback (most recent call last):
 
Traceback (most recent call last):
  File "<stdin>", line 1, in <module>
+
File "<stdin>", line 1, in <module>
  File "/usr/lib/python2.5/site-packages/Bio/GenBank/Scanner.py", line 410, in parse_records
+
File "/usr/lib/python2.5/site-packages/Bio/GenBank/Scanner.py", line 410, in parse_records
    record = self.parse(handle)
+
  record = self.parse(handle)
  File "/usr/lib/python2.5/site-packages/Bio/GenBank/Scanner.py", line 393, in parse
+
File "/usr/lib/python2.5/site-packages/Bio/GenBank/Scanner.py", line 393, in parse
    if self.feed(handle, consumer) :
+
  if self.feed(handle, consumer) :
  File "/usr/lib/python2.5/site-packages/Bio/GenBank/Scanner.py", line 360, in feed
+
File "/usr/lib/python2.5/site-packages/Bio/GenBank/Scanner.py", line 360, in feed
    self._feed_first_line(consumer, self.line)
+
  self._feed_first_line(consumer, self.line)
  File "/usr/lib/python2.5/site-packages/Bio/GenBank/Scanner.py", line 540, in _feed_first_line
+
File "/usr/lib/python2.5/site-packages/Bio/GenBank/Scanner.py", line 540, in _feed_first_line
    assert len(fields) == 7
+
  assert len(fields) == 7
AssertionError
+
AssertionError</python>
  
 
* Questions like this will probably be answered faster if you post them to the BioPython [[Mailing lists|mail list]]. --[[User:Cjfields|Cjfields]] 16:03, 23 May 2007 (EDT)
 
* Questions like this will probably be answered faster if you post them to the BioPython [[Mailing lists|mail list]]. --[[User:Cjfields|Cjfields]] 16:03, 23 May 2007 (EDT)
  
== SeqIO.to_dict() how to specify SeqRecord Alphabet ==
+
* [[User:Maubp|Peter]] as Chris suggested, your question would have been noticed much earlier on the mailing list - but filing a bug would have been an better idea (see [http://bugzilla.open-bio.org/ bugzilla]). Could you do that with a bit more information on where the example file came from?  Thanks.
 +
 
 +
== How to specify SeqRecord Alphabet ==
  
 
Hi,
 
Hi,
Line 28: Line 49:
 
The to_dict() method creates a dictionary with biopython SeqRecord objects.
 
The to_dict() method creates a dictionary with biopython SeqRecord objects.
  
record_dict = SeqIO.to_dict(SeqIO.parse(handle, "fasta"))
+
<python>record_dict = SeqIO.to_dict(SeqIO.parse(handle, "fasta"))</python>
  
 
How can I specify which Alphabet do they must have?
 
How can I specify which Alphabet do they must have?
Line 38: Line 59:
 
  Desription: cneo_JEC21:cn-jec21_chr11
 
  Desription: cneo_JEC21:cn-jec21_chr11
 
  Seq('CCCCCTAACCCCCTAACCCCCTAACCCCCTAACCCCCTAACCCCTAACCCCCTAACCCTA ...', SingleLetterAlphabet())
 
  Seq('CCCCCTAACCCCCTAACCCCCTAACCCCCTAACCCCCTAACCCCTAACCCCCTAACCCTA ...', SingleLetterAlphabet())
 +
 +
[[User:Peter]]: The to_dict() method doesn't do anything with alphabets - your question is really about the parse() function.  For Biopython 1.49 we've extending parse() and read() to take an optional alphabet to cover this exact situation where the file format doesn't specify the alphabet (see [http://bugzilla.open-bio.org/show_bug.cgi?id=2443 enhancement bug 2433]).
 +
 +
In the short term, if you need to specify the alphabet, then the best I can suggest is something like this:
 +
 +
<python>Bio.Alphabet import generic_dna
 +
records = list(SeqIO.parse(handle, "fasta"))
 +
for record in records :
 +
    record.seq.alphabet = generic_dna
 +
record_dict = SeqIO.to_dict(records)</python>
 +
 +
Or, perhaps:
 +
 +
<python>Bio.Alphabet import generic_dna
 +
def force_dna(record) :
 +
    record.seq.alphabet = generic_dna
 +
    return record
 +
record_dict = SeqIO.to_dict(map(force_dna, SeqIO.parse(handle, "fasta")))</python>
 +
 +
Or, if you like generators and are using Python 2.4 or later:
 +
 +
<python>Bio.Alphabet import generic_dna
 +
def force_dna(record) :
 +
    record.seq.alphabet = generic_dna
 +
    return record
 +
record_dict = SeqIO.to_dict(force_dna(rec) for rec in SeqIO.parse(handle, "fasta")))</python>
 +
 +
Why not join the discussion mailing list (see [[mailing lists]]) and tell us what you think?

Latest revision as of 09:43, 4 November 2008

Hello everyone,

  • If you want help on using Bio.SeqIO, please join the discussion mailing list (see mailing lists).
  • If you think you've found a bug, please report it on bugzilla.
  • If you have a nice example you'd like to add, I would encourage you to join the discussion mailing list, but posting the code here should be fine.

Thanks,

Peter

New example, Random Fragments

I added a example into the output section, but now I wonder if this is the best place to put it. Maybe it should be under "examples". --Sbassi 01:20, 27 August 2007 (EDT)

Peter Nice idea - and yes, I would have put it under examples too. I've moved it and edited it too - I wanted the style to match the rest of the page, and also I think there was a possible problem in your randomising code where a sequence might run over the end of the record. I opted to simplify the example - I hope you're still happy with it.

EMBL Problem?

Hi, I am a newbie of Biopython. I tested Bio.SeqIO.parse on EMBL formated miRNA.dat from the microRNA Registry. But I got an error. Is the EMBL format supported fully? As:

>>>from Bio import SeqIO
>>>handle = open('/home/liang/Desktop/miRNA.dat','rU')
>>>record_iterator = SeqIO.parse(handle,'embl')
>>>first_record=record_iterator.next()
Traceback (most recent call last):
 File "<stdin>", line 1, in <module>
 File "/usr/lib/python2.5/site-packages/Bio/GenBank/Scanner.py", line 410, in parse_records
   record = self.parse(handle)
 File "/usr/lib/python2.5/site-packages/Bio/GenBank/Scanner.py", line 393, in parse
   if self.feed(handle, consumer) :
 File "/usr/lib/python2.5/site-packages/Bio/GenBank/Scanner.py", line 360, in feed
   self._feed_first_line(consumer, self.line)
 File "/usr/lib/python2.5/site-packages/Bio/GenBank/Scanner.py", line 540, in _feed_first_line
   assert len(fields) == 7
AssertionError
  • Questions like this will probably be answered faster if you post them to the BioPython mail list. --Cjfields 16:03, 23 May 2007 (EDT)
  • Peter as Chris suggested, your question would have been noticed much earlier on the mailing list - but filing a bug would have been an better idea (see bugzilla). Could you do that with a bit more information on where the example file came from? Thanks.

How to specify SeqRecord Alphabet

Hi, nice work with this module!

The to_dict() method creates a dictionary with biopython SeqRecord objects.

record_dict = SeqIO.to_dict(SeqIO.parse(handle, "fasta"))

How can I specify which Alphabet do they must have? By default, they have SingleLetterAlphabet.

Chromosome cneo_JEC21:cn-jec21_chr11:
ID: cneo_JEC21:cn-jec21_chr11
Name: cneo_JEC21:cn-jec21_chr11
Desription: cneo_JEC21:cn-jec21_chr11
Seq('CCCCCTAACCCCCTAACCCCCTAACCCCCTAACCCCCTAACCCCTAACCCCCTAACCCTA ...', SingleLetterAlphabet())

User:Peter: The to_dict() method doesn't do anything with alphabets - your question is really about the parse() function. For Biopython 1.49 we've extending parse() and read() to take an optional alphabet to cover this exact situation where the file format doesn't specify the alphabet (see enhancement bug 2433).

In the short term, if you need to specify the alphabet, then the best I can suggest is something like this:

Bio.Alphabet import generic_dna
records = list(SeqIO.parse(handle, "fasta"))
for record in records :
    record.seq.alphabet = generic_dna
record_dict = SeqIO.to_dict(records)

Or, perhaps:

Bio.Alphabet import generic_dna
def force_dna(record) :
    record.seq.alphabet = generic_dna
    return record
record_dict = SeqIO.to_dict(map(force_dna, SeqIO.parse(handle, "fasta")))

Or, if you like generators and are using Python 2.4 or later:

Bio.Alphabet import generic_dna
def force_dna(record) :
    record.seq.alphabet = generic_dna
    return record
record_dict = SeqIO.to_dict(force_dna(rec) for rec in SeqIO.parse(handle, "fasta")))

Why not join the discussion mailing list (see mailing lists) and tell us what you think?

Personal tools
Namespaces
Variants
Actions
Navigation
Toolbox