I have a word representing a DNA sequence and each letter of that word is a nucleotide base (A,T,G,C). Each letter has a phred quality score. The formula used for the phred quality score is $Q=-10\log_{10} P$ where $Q$ is the phred score and $P$ the probability of that letter to be incorrect( for example, there was a problem in DNA sequencing).
My question is:
what is the best way to give the entire word a quality score, so that it takes into account the phred score of each letter?
One thing I though about was to take the average and another was to add the phred scores.
