Approximate statistics of gapped alignments

A heuristic approximation to the score distribution of gapped alignments in the logarithmic domain is presented. The method applies to comparisons bet ween random, unrelated protein sequences, using standard score matrices and arbitrary gap penalties. It is shown that gapped alignment behavior is ess entially governed by a single parameter, alpha, depending on the penalty sc heme and sequence composition. This treatment also predicts the position of the transition point between logarithmic and linear behavior. The approxim ation is tested by simulation and shown to be accurate over a range of comm only used substitution matrices and gap-penalties.