LETTERS ERAP2 ERAP1 2 Examples of differer oform events observed. Data is graphed as in Figure 1b. (a)Gene expression level changes of assette exon. (b)Differential 3 UTR change of ERAPI resulting in long and short isoforms with alternative stop codon use. (c) Expression of tw Pu叫 TCL6 transcript isoforms that contain different 5 and 3'ends. (d)Increasing significance and fold hange in expression levels toward the 3 end of he CCT2 gene, suggesting genetic variation associated with mRNA stability 男系己吧器器语导导器寻 88R和NR器888%拓等等等哥 888 388888888888888888888 3, 554 genes 0. Differences in statistical strin- Probe s gency and false discovery rate most likely d CCT explain the higher proportion of SNP tions in their study. However, their set of 10 3, 554 genes was preselected for the most variable expression phenotypes among g original set of >8,000 genes. This restricted set of genes may exclude examples of isoform changes without an accompanying change in whole-gene expression, which we observed in our study. In future expression association studies, comparative meta-analyses across dif- ferent microarray designs may help eliminate platform-specific technical artifacts and allow the elucidation of true isoform and We show that tools such as the exon array, 9 hybridization signals(see Methods ) suboptimal primer design, targeting probes to many regions of the gene, give a more complete o limited sensitivity of our validation methods, and/or noise from the picture of the true complexity of variation in gene expression than 5 microarray. We also validated several differentially spliced exons under previously believed. This variation exists at all levels of transcript 2 a more relaxed stringency below our estimated cutoff, indicating that processing, beginning with initiation of transcription, through pre ao the frequency of genes showing SNP-associated changes is probably mRNA splicing 6,202, to alternative polyadenylation, and it has the a greater than what can be estimated from our current analysis. A recent potential to exert diverse cellular responses and phenotypic effects. o estimate suggests that -21%6 of annotated alternatively spliced genes are associated with SNPs that determine the relative abundances of the alternative transcript isoforms. A recent study used Ilumina arrays to capture gene expression information within the CEU population. The Illumina design, along Expression with many other expression platforms, targets probes to the 3 end of genes and cannot identify specific isoform changes Our present results demonstrate that the nature of the changes is qualitatively different than previously reported for several genes in that study. For example our analysis shows that IRF5, implicated in susceptibility to systemic lupus erythematosus, shows differences in the 3 UTR (Fig. 4), where the A allele of rsl0954213 creates a functional polyadenylation site, shortening its 3 UTR9. This result for IRF5 contrasts the original edicted change at the gene expression level0, I3 and occurs because he Illumina array interrogates IRF5 with a probe in the 3 UTR specific Figure 3 Classification of genes showing expression changes at the exon to the long isoform. Other examples previously classified as expression changes include PTER, which we show to have a variation in the 3 categories depending on the nature of the isoform change occurring expression changes at the whole transcript level (green), transcripti UTR,and CI7orf81(also known as DERP6), which shows alternative initiation changes (yellow), alternative splicing of a cassette exon(blue), splicing of a cassette exon. Another interesting example is ERAP2, transcription termination changes (purple), and complex changes of multipl hich has been reported as having an expression change. Our results event types(red). The percentages shown assume a uniform false-positive onfirm this variation in expression; however, we additionally detect rate for all results To obtain a lower bound for the relative frequency of alternative splice-site use in one of the exons(Fig. 2a). Many platforms isoform variants, we have also recalculated the frequencies of the isoform have been used so far in these population-wide expression analyses, changes(but not whole-gene expression and complex changes)based on our ough there is substantial werlap between the studies, signifi- Thus, we obtained the following ranges for each of the changes: whole gene cant discordance also exists. A recent paper identified 374 gene- expression, 39-44%: initiation, 10-11%; splicing. 24-26%: termination expression phenotypes associated with SNP markers from a study of 16-18%; and complex events, 6-7% NATURE GENETICS VOLUME 40 NUMBER 2 I FEBRUARY 2008hybridization signals19 (see Methods), suboptimal primer design, limited sensitivity of our validation methods, and/or noise from the microarray. We also validated several differentially spliced exons under a more relaxed stringency below our estimated cutoff, indicating that the frequency of genes showing SNP-associated changes is probably greater than what can be estimated from our current analysis. A recent estimate suggests that B21% of annotated alternatively spliced genes are associated with SNPs that determine the relative abundances of the alternative transcript isoforms20. A recent study used Illumina arrays to capture gene expression information within the CEU population13. The Illumina design, along with many other expression platforms, targets probes to the 3¢ end of genes and cannot identify specific isoform changes. Our present results demonstrate that the nature of the changes is qualitatively different than previously reported for several genes in that study. For example, our analysis shows that IRF5, implicated in susceptibility to systemic lupus erythematosus, shows differences in the 3¢ UTR (Fig. 4), where the A allele of rs10954213 creates a functional polyadenylation site, shortening its 3¢ UTR8,9. This result for IRF5 contrasts the original predicted change at the gene expression level10,13 and occurs because the Illumina array interrogates IRF5 with a probe in the 3¢ UTR specific to the long isoform. Other examples previously classified as expression changes include PTER, which we show to have a variation in the 3¢ UTR, and C17orf81 (also known as DERP6), which shows alternative splicing of a cassette exon. Another interesting example is ERAP2, which has been reported as having an expression change10. Our results confirm this variation in expression; however, we additionally detect alternative splice-site use in one of the exons (Fig. 2a). Many platforms have been used so far in these population-wide expression analyses, and although there is substantial overlap between the studies, signifi- cant discordance also exists. A recent paper identified 374 gene￾expression phenotypes associated with SNP markers from a study of 3,554 genes10. Differences in statistical strin￾gency and false discovery rate most likely explain the higher proportion of SNP associa￾tions in their study. However, their set of 3,554 genes was preselected for the most variable expression phenotypes among an original set of 48,000 genes. This restricted set of genes may exclude examples of isoform changes without an accompanying change in whole-gene expression, which we observed in our study. In future expression association studies, comparative meta-analyses across dif￾ferent microarray designs may help eliminate platform-specific technical artifacts and allow the elucidation of true isoform and gene-level variations. We show that tools such as the exon array, targeting probes to many regions of the gene, give a more complete picture of the true complexity of variation in gene expression than previously believed. This variation exists at all levels of transcript processing, beginning with initiation of transcription, through pre￾mRNA splicing16,20,21, to alternative polyadenylation, and it has the potential to exert diverse cellular responses and phenotypic effects. 20 18 16 14 12 –log10 (P-value) 2821369 2821370 2821371 2821372 2821373 2821374 2821378 2821379 2821382 2821383 2821385 2821387 2821389 2821390 2821392 2821394 2821399 2821400 2821402 2821403 2821404 2821406 2821408 2821409 Probe set ERAP2 log2 (fold change) –2 –1 0 1 a b d log2 (fold change) 15 10 –log 5 10 (P-value) 3421631 3421634 3421635 3421636 3421637 3421638 3421640 3421643 3421644 3421645 3421650 3421652 3421653 3421654 3421655 3421656 3421657 3421658 3421661 3421666 3421667 3421668 Probe set 0.5 1.0 1.5 0 2.0 0.0 25 20 15 10 5 0 –log10 (P-value) 2868182 2868181 2868180 2868179 2868178 2868177 2868176 2868173 2868172 2868170 2868169 2868168 2868167 2868161 2868160 2868157 2868156 2868155 2868154 2868152 2868151 2868146 2868145 2868144 2868143 2868142 2868141 2868134 2868133 Probe set ERAP1 CCT2 log2 (fold change) –1.0 –0.5 0.5 0.0 1.5 1.0 c 10 8 6 4 2 –log10 (P-value) 3550142 3550143 3550144 3550145 3550152 3550153 3550154 3550155 3550156 3550157 3550159 3550160 3550161 3550162 3550163 3550164 3550165 3550166 3550167 3550169 3550170 3550174 3550176 3550177 3550178 3550181 3550182 3550183 3550184 3550185 Probe set TCL6 log2 (fold change) –2 –1 0 1 0 2 –3 Expression 39% Initiation 11% Splicing 26% Termination 18% Complex 6% Figure 3 Classification of genes showing expression changes at the exon and/or transcript level. The 324 genes were classified into separate categories depending on the nature of the isoform change occurring: expression changes at the whole transcript level (green), transcription initiation changes (yellow), alternative splicing of a cassette exon (blue), transcription termination changes (purple), and complex changes of multiple event types (red). The percentages shown assume a uniform false-positive rate for all results. To obtain a lower bound for the relative frequency of isoform variants, we have also recalculated the frequencies of the isoform changes (but not whole-gene expression and complex changes) based on our current false positive rate estimate of B20% (from validation experiments). Thus, we obtained the following ranges for each of the changes: whole gene expression, 39–44%; initiation, 10–11%; splicing, 24–26%; termination, 16–18%; and complex events, 6–7%. Figure 2 Examples of different types of transcript isoform events observed. Data is graphed as in Figure 1b. (a) Gene expression level changes of ERAP2, including alternative splicing of a cassette exon. (b) Differential 3¢ UTR change of ERAP1 resulting in long and short isoforms with alternative stop codon use. (c) Expression of two TCL6 transcript isoforms that contain different 5¢ and 3¢ ends. (d) Increasing significance and fold change in expression levels toward the 3¢ end of the CCT2 gene, suggesting genetic variation associated with mRNA stability. NATURE GENETICS VOLUME 40 [ NUMBER 2 [ FEBRUARY 2008 227 LETTERS © 2008 Nature Publishing Group http://www.nature.com/naturegenetics