Calculate mean only for match strings - python-3.x

Question

I have this assignment in which I have a file that contains alot of chromosed that I need to calculate for each one of them the mutation level.
The problem is that each chromosome can appear several times and I need to find the mean for all the mutation levels of this chromosome. and on top of that i need that the mutation will be in same nucleotides (T-->C or G-->A).
The mutation level is calculate by DP4 under INFO which contains four numbers that represented as [ref+,ref-,alt+,alt-]
Example of the file:
#CHROM POS ID REF ALT QUAL FILTER INFO FORMAT Aligned.sortedByCoord.out.bam
chr1 143755378 . T C 62 . DP=550;VDB=0;SGB=-0.693147;RPB=1.63509e-10;MQB=1;BQB=0.861856;MQ0F=0;AC=2;AN=2;DP4=0,108,0,440;MQ=20 GT:PL:DP 1/1:89,179,0:548
chr3 57644487 . T C 16.4448 . DP=300;VDB=0;SGB=-0.693147;RPB=0.993846;MQB=1;BQB=0.316525;MQ0F=0;ICB=1;HOB=0.5;AC=1;AN=2;DP4=0,166,0,134;MQ=20 GT:PL:DP 0/1:49,0,63:300
chr3 80706912 . T C 212 . DP=298;VDB=0;SGB=-0.693147;RPB=0.635135;MQB=1;MQSB=1;BQB=0.609797;MQ0F=0;AC=2;AN=2;DP4=1,1,256,40;MQ=20 GT:PL:DP 1/1:239,255,0:298
So this what I did until now and Im kinda stuck not really knowing how to continue from that point:
def vcf(file):
with open(file, "r+") as my_file:
"""First I wanted to clear the headline"""
for columns in my_file:
if columns.startswith("#"):
continue
"""Then I split the file into columns"""
for columns in my_file:
columns=columns.rstrip('\n').split('\t')
"""This is the info column"""
for row in columns[7]:
row = columns[7].split(";")
"""Using slicing I extracted the DP4 part and removed the str DP4"""
DP4 = [row[-2]]
new_DP4 = [x.replace("DP4=","") for x in DP4]
"""Then I took all the int outs and put them under the categories"""
for x in new_DP4:
xyz = x.split(",")
ref_plus = int(xyz[0])
ref_minus = int(xyz[1])
alt_plus = int(xyz[2])
alt_minus = int(xyz[3])
"""calculated the mean for each one"""
formula = ((alt_minus+alt_plus)/(alt_minus+alt_plus+ref_minus+ref_plus))
"""made a list of the chromosomes and their means"""
chr_form = [columns[0] , columns[3], columns[4], (formula)]
so basically I thought that now that I have all the data in list I can sort out somehow the same chr and do the means but I cant figure out how to do it.. I tried to use regex as well but im not that familiar with that
this is my current output for chr_form:
['chr3', 'T', 'C', 0.44666666666666666]
['chr3', 'T', 'C', 0.9932885906040269]
['chr5', 'A', 'G', 0.42073170731707316]
['chr5', 'A', 'G', 0.5772870662460567]
['chr6', 'A', 'G', 0.5153061224489796]
['chr6', 'A', 'G', 0.8934010152284264]
and so on..
but the output I want to get in the end is this:
{1: {‘T->C’: 0.802}, 3: {‘T->C’:0.446}}
I'll be happy to get an idea or example how to calculate the mean for each chr,

Answer 1

You have lots of unnecessary for loops. The only loop you need is for the lines in the file, you don't need to loop over the characters in fields when you're splitting them or removing something from the whole field.
At the end, you should be adding the result of the calculation to a dictionary.
def vcf(file):
chromosomes = {}
with open(file, "r+") as my_file:
# First I wanted to clear the headline
for line in my_file:
if line.startswith("#"): # skip comment lines.
continue
line=line.rstrip('\n').split('\t')
# This is the info column
info = line[7].split(";")
# Using slicing I extracted the DP4 part and removed the str DP4
DP4 = info[-2].replace("DP4=","")
# Then I took all the int outs and put them under the categories
ref_plus, ref_minus, alt_plus, alt_minus = map(int, DP4.split(','))
# calculated the mean for each one
formula = ((alt_minus+alt_plus)/(alt_minus+alt_plus+ref_minus+ref_plus))
# Get chromosome number from first field
chr_num = int(line[0].replace('chr', ''))
chromosomes[chr_num] = {f'{line[3]}->{line[4]}': formula}
return chromosomes