Fig. 1.

Antisense transcription in P. pacificus and C. elegans. a Examples of different antisense overlaps. The plots show assembled transcripts for sense and antisense transcripts and the coverage profiles on logarithmic scale. b Distribution of different classes of antisense overlaps. Intronic ASTs represent by far the most abundant class. c Simulated and observed number of BLASTX hits against available proteomic data. As our transcriptome data has directional information, we filtered out BLASTX hits that referred to the reverse strand of the query transcripts. ASTs (red line) show more BLASTX hits in proteomic data than expected d Simulated and observed number of BLASTX hits against predicted genes in P. pacificus. e Correlation between expression of ASTs and their host genes in developmental transcriptomes of P. pacificus. Under the assumption that ASTs could have a regulatory effect on their host genes, a subset of antisense gene pairs (blue lines) should show a trend towards negative or positive correlation. However, the expression correlation of antisense gene pairs is within the range of randomly chosen gene pairs suggestion no regulalatory effect. f Correlation between expression of ASTs and their host genes in 50 transcriptomic samples of C. elegans. g Size distribution of conserved syntenic blocks between C. elegans and P. pacificus and sizes of conserved antisense gene pairs. In C. elegans, conserved antisense gene pairs tend to be larger than other conserved syntenic blocks (P=0.03, Wilcoxon-test). h Size distribution of conserved syntenic blocks and antisense gene pairs in P. pacificus