RAX2: genome-wide detection of condition-associated transcription variation

Abstract

AbstractAlmost all of mammalian genes have mRNA variants due to alternative promoters, alternative splice sites, and alternative cleavage and polyadenylation sites. In most cases, change in transcript due to choosing alternative cleavage and polyadenylation sites does not lead to change in protein sequence, while selection of alternative promoters and alternative splice sites would alter protein sequence. Nevertheless, all these alternations would give rise to different RNA isoforms. Selection of alternative RNA isoforms has been found to be associated with change in condition. For example, many studies have revealed that alternative cleavage and polyadenylation (poly(A)) are correlated to proliferation, differentiation, and cellular transformation. Thus, unlike gene expression in microarray, change in usage of splice sites or poly(A) sites associated with conditions does not involve differential expression and hence cannot be detected by differential analysis methods but can be done by association methods. Traditional association methods such as Pearson chi-square test and Fisher Exact test are single test methods and do not work on the RNA count data derived from replicate libraries. For this reason, we here developed a large-scale association method, called ranking analysis of chi-squares (RAX2). Simulations demonstrated that RAX2 worked well for finding association of changes in usage of poly(A) sites with condition change. We applied our RAX2 to our primary T-cell transcriptomic data of over 9899 tags scattered in 3812 genes and found that 1610 (16.3%) tags were associated in transcription with immune stimulation at FDR < 0.05 and most of these tags associated with stimulation also had differential expression. Analysis of two and three tags within genes revealed that under immune stimulation, short RNA isoforms were significantly preferably used. Like cell proliferation and division, short RNA isoforms are highly prioritized to be used for cell growth.