If your genome is not too long you can try the pairwise alignment from BioPython http://biopython.org/DIST/docs/api/Bio.pairwise2-module.html

Otherwise you'll have to align this sequence using a proper alignment software like BWA or Bowtie2

Edit : Lactococcus lactis, 2.5M bases

You can give a try to BioPython and check the running time

Transcription factors bind to motifs which can have some variation in nucleotide composition. Better use a dedicated tool such as fimo from the MEME suite for this. You pattern matching approach would require 100% sequence identity which is simply not how transcription factor binding works.

BLAST would be the obvious choice, but expect lots of hits so you'll have to do some filtering a posteriori.

(You needn't use BLAST via python, but you could if you wanted to).

you can use seqkit locate @ PaSua

Example input:

$ cat test.fa 
>a
TGTAAACCTTTTCATACTEAAGATTTGTAAACCTTTTCATGACCGTAGTGTAAACCTTTTCA
>b
ATCGATGCGATTGTAAACCTTTTCAATGCGATGACTGTAAACCTTTTCA

output:

$ seqkit locate -idp "TGTAAACCTTTTCA" test.fa

seqID   patternName pattern strand  start   end matched
a   TGTAAACCTTTTCA  TGTAAACCTTTTCA  +   1   14  TGTAAACCTTTTCA
a   TGTAAACCTTTTCA  TGTAAACCTTTTCA  +   26  39  TGTAAACCTTTTCA
a   TGTAAACCTTTTCA  TGTAAACCTTTTCA  +   49  62  TGTAAACCTTTTCA
b   TGTAAACCTTTTCA  TGTAAACCTTTTCA  +   12  25  TGTAAACCTTTTCA