Groups | Search | Server Info | Keyboard shortcuts | Login | Register [http] [https] [nntp] [nntps]

Groups > comp.lang.python > #46994 > unrolled thread

Issue values dictionary

Back to article view | Back to comp.lang.python

Contents

  Issue values dictionary claire morandin <claire.morandin@gmail.com> - 2013-06-04 19:41 -0700
    Re: Issue values dictionary alex23 <wuwei23@gmail.com> - 2013-06-04 20:17 -0700
      Re: Issue values dictionary Peter Otten <__peter__@web.de> - 2013-06-05 09:43 +0200
        Re: Issue values dictionary alex23 <wuwei23@gmail.com> - 2013-06-05 02:46 -0700
    Re: Issue values dictionary claire morandin <claire.morandin@gmail.com> - 2013-06-04 21:05 -0700

#46994 — Issue values dictionary

 I have two text file with a bunch of transcript name and their corresponding length, it looks like this:
ERCC.txt
ERCC-00002	1061
ERCC-00003	1023
ERCC-00004	523
ERCC-00009	984
ERCC-00012	994
ERCC-00013	808
ERCC-00014	1957
ERCC-00016	844
ERCC-00017	1136
ERCC-00019	644
blast.tx
ERCC-00002	1058
ERCC-00003	1017
ERCC-00004	519
ERCC-00009	977
ERCC-00019	638
ERCC-00022	746
ERCC-00024	134
ERCC-00024	126
ERCC-00024	98
ERCC-00025	445

I want to compare the length of the transcript and see if the length in blast.txt is at least 90% of the length in ERCC.txt for the corresponding transcript name ( I hope I am clear!) 
So I wrote the following script:
ercctranscript_size = {}
for line in open('ERCC.txt'):
    columns = line.strip().split()
    transcript = columns[0]
    size = columns[1]
    ercctranscript_size[transcript] = int(size)

unknown_transcript = open('Not_sequenced_ERCC_transcript.txt', 'w')
blast_file = open('blast.txt')
out_file = open ('out.txt', 'w')

blast_transcript = {}
blast_file.readline()
for line in blast_file:
    blasttranscript = columns[0].strip()
    blastsize = columns[1].strip()
    blast_transcript[blasttranscript] = int(blastsize)
    
blastsize = blast_transcript[blasttranscript]    
size = ercctranscript_size[transcript]
print size 
if transcript not in blast_transcript:
    unknown_transcript.write('{0}\n'.format(transcript))
else:
    size = ercctranscript_size[transcript]
    if blastsize >= 0.9*size:
        print >> out_file, transcript, True
    else:
        print >> out_file, transcript, False

But I have a problem storing all size length to the value size as it is always comes back with the last entry. 
Could anyone explain to me what I am doing wrong and how I should set the values for each dictionary? I am really new to python and this is my first script 

Thanks for your help everybody!

[toc] | [next] | [standalone]

#46996

On Jun 5, 12:41 pm, claire morandin <claire.moran...@gmail.com> wrote:
> But I have a problem storing all size length to the value size as it is always comes back with the last entry.
> Could anyone explain to me what I am doing wrong and how I should set the values for each dictionary?

Your code has two for loops, one that reads ERCC.txt into a dict, and
one that reads blast.txt into a dict. The first assigns to
`transcript`, the second to `blasttranscript`. When the loops are
finished, you're using the _last_ value set for both `transcript` and
`blasttranscript`. So, really, you want _three_ loops: two to load the
files into dicts, then another to compare the two of them. If the
transcripts in blast.txt are guaranteed to be a subset of ERCC.txt,
then you could get away with two loops:

# convenience function for splitting lines into values
    def get_transcript_and_size(line):
        columns = line.strip().split()
        return columns[0].strip(), int(columns[1].strip())

    # read in blast_file
    blast_transcripts = {}
    with open('transcript_blast.txt') as blast_file:
        # this is a context manager, it'll close the file when it's
finished
        for line in blast_file:
            blasttranscript, blastsize = get_transcript_and_size(line)
            blast_transcripts[blasttranscript] = blastsize

    # read in ERCC and compare to blast
    with open('transcript_ERCC.txt') as ercc_file, \
         open('Not_sequenced_ERCC_transcript.txt', 'w') as
unknown_transcript, \
         open('transcript_out.txt', 'w') as out_file:
        # this is called a _nested_ context manager, and requires 2.7+
or 3.1+
        for line in ercc_file:
            ercctranscript, erccsize = get_transcript_and_size(line)
            if ercctranscript not in blast_transcripts:
                print >> unknown_transcript, ercctranscript
            else:
                is_ninety_percent = blast_transcripts[ercctranscript]
>= 0.9*erccsize
                print >> out_file, ercctranscript, is_ninety_percent

I've cleaned up your code a bit, using more similar naming schemes and
the same open/write procedures for all file access. Generally, any
time you're repeating code, you should stick it into a function and
use that instead, like the `get_transcript_and_size` func. If the
columns in your two files are separated by tabs, or always by the same
number of spaces, you can simplify this even further by using the csv
module: http://docs.python.org/2/library/csv.html

Hope this helps.

[toc] | [prev] | [next] | [standalone]

#47031

alex23 wrote:

>     def get_transcript_and_size(line):
>         columns = line.strip().split()
>         return columns[0].strip(), int(columns[1].strip())
 
You can remove all strip() methods here as split() already strips off any 
whitespace from the columns.

Not really important, but the nitpicker in me keeps nagging ;)

[toc] | [prev] | [next] | [standalone]

#47054

On Jun 5, 5:43 pm, Peter Otten <__pete...@web.de> wrote:
> You can remove all strip() methods here as split() already strips off any
> whitespace from the columns.
>
> Not really important, but the nitpicker in me keeps nagging ;)

Thanks, I really should have checked but just pushed the OPs code into
a function, I didn't want to startle them with completely different
code :)

As I mentioned, I would've used the csv module for this anyway, which
is why I never remember the split/strip behaviour.

Nitpickery can be a virtue in this field :)

[toc] | [prev] | [next] | [standalone]

#47004

@alex23 I can't thank you enough this really helped me so much, not only fixing my issue but also understanding where was my original error

Thanks a lot

[toc] | [prev] | [standalone]

Back to top | Article view | comp.lang.python

csiph-web