Sort dictionary in descending order

Question

Hi I am looking to sort a list of kmers into descending order. I used KMC to obtain a list of kmers from a fastq file separated by tab (e.g.; kmer count). From there I have written this python script to attempt to sort the kmers and their counts in descending order:

import csv

with open("6mers.tsv") as kmer_file:
    
    tsv_reader = csv.DictReader(kmer_file, delimiter="\t")
    
    for kmer in tsv_reader:
        
        name = kmer["kmer"]
        
        count = kmer["count"]
        
        # section below attempts to sort the dictionary into descending order
        # taken from (https://www.freecodecamp.org/news/sort-dictionary-by-value-in-python/) 
        sorted_kmers = sorted(kmer.items(), key=lambda x:x[1], reverse=True)

        converted_dict = dict(sorted_kmers)
    
        print(converted_dict)

The code above outputs this:

{'kmer': 'TGCCAA', 'count': '2'}
{'kmer': 'TGGACA', 'count': '2'}
{'kmer': 'TGGGAA', 'count': '2'}
{'kmer': 'TGGTAA', 'count': '5'}
{'kmer': 'TGTCAA', 'count': '5'}
{'kmer': 'TGTGAA', 'count': '2'}
{'kmer': 'TTCCAA', 'count': '13'}
{'kmer': 'TTGAAA', 'count': '6'}
{'kmer': 'TTGCAA', 'count': '4'}
{'kmer': 'TTTAAA', 'count': '6'}

Clearly I am creating a separate dictionary for each kmer and it's count - I'm not sure how but I would like a result that looks more like this:

{'TTCCAA', '13'}
{'TTGAAA', '6'}
{'TTTAAA', '6'}
{'TGGTAA', '5'}
{'TGTCAA', '5'}
{'TTGCAA', '4'}
{'TGCCAA', '2'}
{'TGGACA', '2'}
{'TGGGAA', '2'}
{'TGTGAA', '2'}

Can someone help me fix my code so it works?

EDIT:

The .tsv file looks just like this:

TGCCAA  2
TGGACA  2
TGGGAA  2
TGGTAA  5
TGTCAA  5
TGTGAA  2
TTCCAA  13
TTGAAA  6
TTGCAA  4
TTTAAA  6

Answer 1

Just make sure to cast the count to an integer prior to sorting. Assuming you need to do something with the counts, you could use:

import csv

from operator import itemgetter
from pprint import pprint


with open("6mers.tsv") as kmer_file:

    reader = csv.reader(kmer_file, delimiter="\t")
    counts = ((kmer, int(count)) for kmer, count in reader)

    sorted_kmers = sorted(counts, key=itemgetter(1), reverse=True)

    pprint(sorted_kmers)

Results:

[('TTCCAA', 13),
 ('TTGAAA', 6),
 ('TTTAAA', 6),
 ('TGGTAA', 5),
 ('TGTCAA', 5),
 ('TTGCAA', 4),
 ('TGCCAA', 2),
 ('TGGACA', 2),
 ('TGGGAA', 2),
 ('TGTGAA', 2)]

Answer 2

This is untested code ... without seeing to the raw file its tricky. Could you try:

import csv
with open("6mers.tsv") as kmer_file:
    tsv_reader = csv.DictReader(kmer_file, delimiter="\t")
    mydict = {rows[0]:rows[1] for rows in tsv_reader if not rows[0] == 'kmer'}

sorted_kmers = sorted(mydict.items(), key=lambda x:x[1], reverse=True)
print(sorted_kmers)

I'm assuming there is a header called 'kmer'

---

The output didn't work. The rows['kmers'] infers that (somehow) tsv_reader is already a dictionary (that's not how I thought it worked). Thus try,

import csv
with open("6mers.tsv") as kmer_file:
    tsv_reader = csv.DictReader(kmer_file, delimiter="\t")
    mydict = {rows['kmers']:rows['count'] for rows in tsv_reader}

sorted_kmers = sorted(mydict.items(), key=lambda x:x[1], reverse=True)
print(sorted_kmers)

Unless it's loaded as a singular dictionary with Kmer as the key and count as the value the sort line will not work. Once that format is achieved as a dictionary then it will work.

What I don't understand is where it's getting 'kmer' or 'count' from if there are no headers in the csv file. Something isn't right if the above code works.