Intron distribution in ancient paralogs supports random insertion and not random loss

Abstract

The intron positions of ten different protein families were examined to determine (the statistical likelihood of) whether spliceosomal introns are the result of random insertion events into previously intronless genes, on the one hand, or the result of random loss from common ancestral introns, on the other. The number of expected matches for the alternative scenarios was calculated for a binomial distribution by considering currently observed introns relative to all possible locations for insertion or loss. Introns occurring at approximately the same location (hereafter called a "match") were tallied for each of the paired proteins. Matches were identified by their positions in the multiple alignment and were defined as any two introns occurring within a window of 11 possible nucleotide positions, thereby allowing for possible alignment errors and "intron sliding." Matches were tallied from the raw data and compared with the expected number of matches for the two different scenarios. The results suggest that the distribution of introns in genes encoding proteins is due to random insertion and not random loss.