A ‘while’ loop inside a if clause with a couple of flags should to the trick.

HTH

Paulo

Here is the result I need:
Fasta1
$value1= AAGCCCCCCCCCCCCCCCCCCCCCCCCCCCCGAGAT
$value2 = 1
$value3 = 80
$value4 = 16.30
$value 5 =(((((()))))>>>…….()))))))))
$value6 = 0.999
Fasta1
…..
fast1
—-
I know how to handle regular expression and passing the above value to database. I only need the loop to itrate

Thank you,
Dereje

Don’t worry about commenting and please continue doing so. And don’t worry about using my time or annoying the readers. It is very important to receive comments, suggestions, ideas and criticism.

Please continue posting. Or send me an email when needed, it is a pleasure to interact.

I am posting your code to the topic this afternoon.

Paulo

Now I see what you mean! Somehow my indents don’t show on the comments posts! They should show on the email I sent, but I’m using a lot of your time, and this must be rather boring to any readers of the comments. So I’m going to post less and read more for a while.

Thanks for the advice on not keeping all the data in memory.

Joe

With the whitespace issues its hard for me to be sure, but it appears that you solution does add some useful checks. However I’d like to caution you on one aspect. It appears you are trying to keep all of the sequences in memory, before you finally print them. That is fine for small files, but fasta files are often quite huge (especially the ones you’d want to split). If so, your code will hit a pretty hard wall once the input file is nearly as big as ram.

I sent you an email to your netscape address (the one I have registered). Check your spam box otherwise my email is

nuin at genedrift dot org

Cheers
Paulo

I’m embarrassed to say that I can’t find the email address on your blog to send the source to. If you don’t mind sending same to me or telling me where to find it, I’ll be glad to send the source. Thanks for your patience.

Joe

Would you mind sending me source in an email? The comments are not formatted for Python.

Thanks
Paulo

It’s been fascinating reading your blog. And very instructive. I’m a beginner in Python and bioinformatics, but have been able to follow it (really, that’s not a boast but a compliment).

The way to split a FASTA file with Python that Michael Cariaso sent in is short, efficient and understandable (maybe that’s what Pythonic means). I’m sending a little modification (actually two small additions) to it. I’m pretty sure that they’d be obvious additions to you and to him, but, if correct, they might be useful and instructive to others who are “beginning Python for bioinformatics” like me. It seems to work with the fasta.seq file you supplied in you blog, and some permutations and multiplications of it that I invented.

At any rate, if you have the time to examine and correct it, I’d be grateful.

Joe

# f6smod.py 10-26-07 jto
# purpose: modify program by Michael Cariaso, given in Paolo Nuin’s blog:
# http://python.genedrift.org
# archive: kept by joeoettinger@netscape.net

import sys

fileobj = open(‘fasta.seq’)
ignore = fileobj.read(1)
# The above line removes the first char from the file object.
# It could also be used to check that the first char is a ‘>’

if ignore != ‘>’:
print “The first character in the supposed FASTA file is: ” + ignore
print “not ‘>’, so sys.exit() is being invoked.”
sys.exit()

text = fileobj.read()
records = text.split(‘>’)

# Here, rather than use the for loop to just print out the sequences, it
# is used to store them in a list. After that they can be printed out, or
# stored in separate files, or be further split into header line and sequence
# (using the carriage return at the end of the header file).
seqlist = []
listcount = 0

# store each header-sequence in a list
for i in records:
i = ‘>’ + i
seqlist.append(i)
listcount += 1
# Just to show it’s working right, print the list
for seq in seqlist:
print seq
# Split into header line and sequence, and make the sequence a single string.
for seq in seqlist:
splitCR = seq.split(‘\n’)
print “header: ” + splitCR[0]
sequence = ”.join(splitCR[1:])
print “sequence: ”
print sequence

I will add your code to the post (referencing it of course). I will work on a sed/awk solution this weekend.

Thanks a lot for your input.

http://mike.pbwiki.com/pythonfasta

# This magic variable makes perl read lines
# that end with >
# instead of a newline \n
$/=”>”;

while () { # foreach line in the input files
if (/^>(\w+)/) { # grab the first word of text
open(F,”>$1″) || # open a file named that word
warn “$1 write failed:$!\n”;
chomp; # strip off the > at the end
print F “>”, $_; # print >text to the file
}
}

It as well as the python scripts above suffer from an extra ‘>’ at the beginning of the output.

An easy fix is

fileobj = open(“myfile.fasta”)
ignore = fileobj.read(1)
text = fileobj.read()
records = text.split(“>”)
for i in records:
print ‘>’ + i

But I consider this a better approach

def eachfasta(fileobj):
sofar = fileobj.readline()
for line in fileobj:
if ‘>’ == line[0]:
yield sofar
sofar = line
else:
sofar += line

yield sofar

fileobj = open(“myfile.fasta”)

for i in eachfasta(fileobj):
print ‘**[%s]**’ % i

Thanks for the Perl explanation. I am researching a way to do it with sed and awk and I post it if no one posts before.

I also expect that Andrew would know the Pythonic way! I will probably write him and ask.

Cheers
Paulo

the old perl motto was
Tim Toady. In that spirit, I was unaware of cplit and am still unlikely to use it. I probably would have kicked this one off in perl, and assuming someone will find it useful, I will try to explain the perl above. I expect there is also a sed/awk solution, and I’d be curious to hear those if anyone can enlighten me.

Incidentally I use biopython daily, but mainly for the ncbi eutils interfaces. I certainly don’t know the biopythonic way to split fasta. I expect Andrew Dalke might.

# This magic variable makes perl read ‘lines’
# that end with ‘>’
# instead of a newline ‘\n’
$/=”>”

while () { # foreach ‘line’ in the input files

if (/^\s*(\S+)/) { # grab the first ‘word’ of text
open(F,”>$1″) || # open a file named that ‘word’
warn “$1 write failed:$!\n”;
chomp; # strip off the “>” at the end
print F “>”, $_; # print ‘>text’ to the file
}

}

I have the intention of include some BioPython basic usage in the future. I am not an experienced user of BioPython, but I am currently playing with it and hope to but something soon here.

The idea of the site was to create a basic guide of Python and Bioinformatics, especially for people that don’t know that much Python and/or a lot of biology. Most of the scripts presented here are very simple and short; the scripts can be used as a primer for more advanced programming. And I hope to do that with BioPython too.

I would like to include some basic tutorials of other Python packages too (bioinfo or not) and even some basic steps to build interfaces with wxPython.

If there is an specific topic you would like to see it covered, please let me know. And thanks for your support.

Cheers
Paulo