Only left with the last key/value pairs added to a dictionary when I go through a loop in Python

Question

Using Python 3 I am trying to get a dictionary of names and counts of the occurrence of certain strings in one long string.
I am sitting here pulling my hair out as this should not be complicated but I have read a lot of answers to this already and I still am not getting it. I'm 5 hours in and definitely not seeing the wood for the trees now.
Hopefully, someone can show me where I am going wrong.
The string is called seq.
seq = 'AAGGTAAGTTTAGAATATAAAAGGTGAGTTAAATAGAATAGGTTAAAATTAAAGGAGATCAGATCAGATCAGATCTATCTATCTATCTATCTATCAGAAAAGAGTAAATAGTTAAAGAGTAAGATATTGAATTAATGGAAAATATTGTTGGGGAAAGGAGGGATAGAAGG'

I have a CSV of words I am looking for and that is in a list called nu
nu = ['AGATC', 'AATG', 'TATC']

The code should use each of the words in nu and get a count of the number of occurrences in seq.
Here is my loop
for i in nu:
    searchstr = {}
    # Line returns a dict of the last value added
    searchstr = dict(key = (i), count = (seq.count(i)))
    print(searchstr)

print(searchstr.keys())
print(searchstr.values())

and the output so I know I'm matching the count correctly with the keys:
{'key': 'AGATC', 'count': 4}
{'key': 'AATG', 'count': 1}
{'key': 'TATC', 'count': 5}
dict_keys(['key', 'count'])
dict_values(['TATC', 5])

I just can't for the life of me get the three dicts into one. I am just left with a dict of ['TATC', 5] as it has overwritten the previous in the list.
I'm still new to this but trying to learn along the way.

alaniwi · Accepted Answer

All you need is to assign elements to the dictionary, not create a new dictionary each time:
searchstr = {}
for i in nu:
    searchstr[i] = seq.count(i)
print(searchstr)

Roberto Caboni · Answer

You declare each at iteration of the loop; that's why you can always see just the last inserted key.
I don't know if it would be an appreciated suggestion, but instead of defining key and count as... key and value I would just use the searched DNA sequence as a key. Something like that:
searchstr = dict()
for i in nu:
    searchstr[i] = seq.count(i)
    
print(searchstr.keys())
print(searchstr.values())
print(searchstr)

print(searchstr['AATG']) #reading a specific result

Output:
dict_keys(['AGATC', 'AATG', 'TATC'])
dict_values ([4, 1, 5])
{'AGATC': 4, 'AATG': 1, 'TATC': 5}
1

As you can see, the dictionary just needs to be declared outside the loop, and in the loop you'll add an element for every searched string.
Please note how it will be easier accessing the specific sequence count.

Kaushal · Answer

searchstr = {}
for i in nu:
    # Line returns a dict of the last value added
    # Earlier the dictionary declaration was here which was overriding the previous value
    searchstr = dict(key = (i), count = (seq.count(i)))
    print(searchstr)

print(searchstr.keys())
print(searchstr.values())

Move the dictionary declaration outside

Raiyan Chowdhury · Answer

I think this is what you want:
searchstr = {}
for i in nu:
    # Line returns a dict of the last value added
    searchstr[i] = seq.count(i)
print(searchstr)

Only left with the last key/value pairs added to a dictionary when I go through a loop in Python

4 Answers

Add your own answers!

Ask a Question