Similarity-Detection and Localization

Abstract

The detection of similarities between long DNA and protein sequences is studied using concepts of statistical physics. It is shown that mutual similarities can be detected by sequence alignment methods only if their amount exceeds a threshold value. The onset of detection is a continuous phase transition which can be viewed as a localization-delocalization transition. The ``fidelity'' of the alignment is the order parameter of that transition; it leads to criteria for the selection of optimal alignment parameters.