Prediction of ribosomal frameshift signals of user-defined models

Abstract

Programmed ribosomal frameshifts are used frequently by RNA viruses to synthesize a single fusion protein from two or more overlapping open reading frames. Depending on the number of nucleotides shifted and the direction of shifting, frameshifts are classified into -n and +n frameshifts, n being the number of nucleotides shifted. Computational identification of frameshift sites is in general very difficult since the sequences of frameshifting cassettes are diverse and highly dependent on the organism. Most current computational methods focus on predicting -1 frameshift sites only, and cannot handle other types of frameshift sites. We previously developed a program called FSFinder for predicting -1 and +1 frameshifts. As an extension of FSFinder, we now present FSFinder2, which is capable of predicting frameshift sites of general type, including user-defined models. We believe FSFinder2 is the first program capable of predicting frameshift signals of general type, and that it is a powerful and flexible tool for predicting genes that utilize alternative decoding, and for analyzing frameshift sites. © Springer-Verlag Berlin Heidelberg 2005.