What is "k-mer"

The term k-mer typically refers to all the possible substrings of length k that are contained in a string. In computational genomics, k-mers refer to all the possible subsequences (of length k) from a read obtained through DNA Sequencing. The amount of k-mers possible given a string of length, L, is L − k + 1 whilst the number of possible k-mers given n possibilities (4 in the case of DNA e.g. ACTG) is n. K-mers are typically used during sequence assembly, but can also be used in sequence alignment. In the context of the human genome, k-mers of various lengths have been used to explain variability in mutation rates.