Skip to content

Instantly share code, notes, and snippets.

@philippbayer
Created January 17, 2012 13:34
Show Gist options
  • Save philippbayer/1626642 to your computer and use it in GitHub Desktop.
Save philippbayer/1626642 to your computer and use it in GitHub Desktop.
openSNP-parser for all files in Python 3
#!/usr/bin/python3
# This Script parses a couple of genotyping-files downloaded from openSNP.
# It supports deCODEme, FamilyTreeDNA and 23andme-files.
# If you don't like the Object Oriented Magic just comment out the lines with allSnps.append
# and do your own magic.
# Example-Useage is name of the script followed by as many filenames as you like.
# ./megaparser.py 200.23andme.12 92.decodeme.1 [...]
# or:
# python3 megaparser.py 200.23andme.12 92.decodeme.1 [...]
#
# License is Public Domain. Made by Philipp Bayer in 2012
import argparse # to parse the arguments, argparse is the "new" optparse
# global array containing all SNPs
allSnps = []
class Snp:
# Each SNP is one line in the original file
def __init__(self, name, variation, chromosome, position, strand, uservariation):
self.name = name # e.g. RS1234
# only deCODEme lists possible variations and strand
if variation is None:
self.variation = "Unknown"
else:
self.variation = variation # e.g. A/T
self.chromosome = chromosome # 1,2,3...
self.position = position # e.g. 213412
if strand is None:
self.strand = "Unknown"
else:
self.strand = strand # +,-
self.uservariation = uservariation # e.g. TT
def printargs(self):
print("Name: " + self.name)
print("Possible variations: " + self.variation)
print("Chromosome: " + self.chromosome)
print("Position: " + self.position)
print("Strand: " + self.strand)
print("Uservariation: " + self.uservariation)
print()
def parseDecodeme(filein):
for line in filein:
# skip the first line
if "Name,Variation,Chromosome,Position,Strand,YourCode" in line:
continue
# Make an array containing all the elements of the line
lineArray = line.replace("\n","").replace(" ","").split(",")
# Make an object based on that array and append it to the global array of all SNPs
allSnps.append(Snp(lineArray[0], lineArray[1], lineArray[2], lineArray[3], lineArray[4], lineArray[5]))
def parse23andme(filein):
for line in filein:
# skip comments
if "#" in line:
continue
lineArray = line.replace(" ", "").replace("\n","").split("\t")
allSnps.append(Snp(lineArray[0], None, lineArray[1], lineArray[2], None, lineArray[3]))
def parseFTDna(filein):
for line in filein:
# skip first line
if "RSID,CHROMOSOME,POSITION,RESULT" in line:
continue
lineArray = line.replace('"','').replace("\n","").split(",")
allSnps.append(Snp(lineArray[0], None, lineArray[1], lineArray[2], None, lineArray[3]))
def parseAllFiles(args):
# go through all files, and call the relevant subroutines
for element in args.filenames:
# use openSNP's style of file-naming to see what's in the file
if "23andme" not in element and "decodeme" not in element and "ftdna" not in element:
print("File called " + element + "doesn't seem to be a proper file.")
continue
filein = open(element)
if "23andme" in element:
parse23andme(filein)
elif "decodeme" in element:
parseDecodeme(filein)
elif "ftdna" in element:
parseFTDna(filein)
filein.close()
def printAllSnps():
# Dummy function that just prints all SNPs to the command-line
for element in allSnps:
element.printargs()
# Define the parser for command-line arguments and add arguments
parser = argparse.ArgumentParser(description="Parses raw genotyping files from 23andme, deCODEme and FamilyTreeDNA based on the filenames openSNP provides.")
# There is an unlimited number of files, so nargs is +
parser.add_argument("filenames", metavar="filename", type=str, nargs="+", help="Name of the file(s) you want to parse.")
args = parser.parse_args()
# Start the main-program
parseAllFiles(args)
# Just print out all SNPs. Do your own thing here :)
printAllSnps()
@superbobry
Copy link

Hello, I've re-worked the code a bit and created a tiny library on top of it, can you please mention it in the Genotype section, in case someone needs a library, instead of a CLI tool?

@philippbayer
Copy link
Author

Very cool! I linked to your library from the page: http://opensnp.org/genotypes

@superbobry
Copy link

Thank you! :)

Sign up for free to join this conversation on GitHub. Already have an account? Sign in to comment