Figure 2
Example output from Poly analysis of homopolymer tract frequency. The representation, R, (fAobs/fAexp) of poly dA tracts is plotted as a function of their length (NA). Filled circles represent the data obtained from an analysis of up to 2000 bases of sense-strand sequence upstream of open reading frames in P. falciparum. Unfilled circles represent data obtained from randomised shuffled sequences of the same length and nucleotide content. Nmaxobs is the longest length of poly dA observed (where a minimum of four tracts of this length were observed) in the upstream sequence, with Nmaxexp representing the maximal length of poly dA tract expected from random base sequence of the same length and nucleotide content. The threshold is the length of poly dA tract which is observed to be over-represented (using the R > 0.5 criterion) in the upstream sequences. The slope of overrepresentation (slope R ) is determined from data between the threshold and Nmaxobs points.