Thousands of exon skipping events differentiate among splicing patterns in sixteen human tissues

Abstract

Alternative splicing is widely recognized for its roles in regulating genes and creating gene diversity. However, despite many efforts, the repertoire of gene splicing variation is still incompletely characterized, even in humans. Here we describe a new computational system, ASprofile, and its application to RNA-seq data from Illumina's Human Body Map project (>2.5 billion reads). Using the system, we identified putative alternative splicing events in 16 different human tissues, which provide a dynamic picture of splicing variation across the tissues. We detected 26,989 potential exon skipping events representing differences in splicing patterns among the tissues. A large proportion of the events (>60%) were novel, involving new exons (~3000), new introns (~16000), or both. When tracing these events across the sixteen tissues, only a small number (4-7%) appeared to be differentially expressed ('switched') between two tissues, while 30-45% showed little variation, and the remaining 50-65% were not present in one or both tissues compared. Novel exon skipping events appeared to be slightly less variable than known events, but were more tissue-specific. Our study represents the first effort to build a comprehensive catalog of alternative splicing in normal human tissues from RNA-seq data, while providing insights into the role of alternative splicing in shaping tissue transcriptome differences. The catalog of events and the ASprofile software are freely available from the Zenodo repository ( http://zenodo.org/record/7068; doi: 10.5281/zenodo.7068) and from our web site http://ccb.jhu.edu/software/ASprofile.

Figure 1.. A high-level view of alternative splicing in sixteen human tissues: numbers of multi-exon ‘genes’ and transcripts from de novo transcript assemblies produced by Cufflinks (left), and by Cuffcompare (right).

Since Cufflinks may break transcripts and genes into multiple fragments when there is insufficient read coverage, we used Cuffcompare to compare transfrags against the Ensembl reference annotations to produce a better estimate for the number of genes and transcripts in the samples. Results in the right panel show the total number of Ensembl annotated as well as novel genes, and respectively transcripts, found in each sample. The number of novel isoforms identified by Cuffcompare is shown in the bottom panel.

Figure 2.. A novel alternatively spliced exon (chr1: 153,611,844–153,611,927) at the CHTOP gene locus, which does not overlap any known annotation.

This novel 84-bp exon, marked with a red circle in the figure, is the 4 ^th exon in one of the transcripts from heart tissue (heart.79763.3), and it appears exclusively in that tissue, although a partial form is present in a skeletal muscle transcript. The two introns flanking the event and the spanning intron are supported by 5, 59 and 702 reads, respectively, in the 16 tissues.

Figure 3.. Splicing variation at skipped exon events as measured by the exon inclusion ratio R = FPKM _on/(FPKM _on+FPKM _off) in the sixteen tissues.

Most events within a given tissue are single variant (top). When both isoforms are present in a tissue, the exon is typically contained in the major form (R ≥ 0.5) (bottom).

Figure 4.. Splicing patterns for the 26,989 exon skipping events are compared between any two tissues, and events are classified by the difference in the splicing ratios.

The 255 x 255 matrix shows the dynamics of exon skipping events between a tissue and each of the others. The numbers of similar (blue), variable (green), switch (purple) and not present (red) events between any two tissues are shown along one line.

Figure 5.. Distribution of novel and known features by the number of tissues in which they occurred.

( A) The percentage of exons found in 1, 2, …, 16 tissues are shown as horizontal bars, for the 2,914 novel exons (‘novel-X’) and 24,075 known exons (‘known-X’). ( B) Similarly for the 15,958 novel introns (‘novel-I’) and 11,031 known introns (‘known-I’).

Figure S4.. Proportion of predicted novel splice forms from the total number of transcript assemblies, including novel and known (top), and absolute counts of novel splice forms (bottom) detected in the 16 tissues.

Novel and known isoforms are determined by comparison with ENSEMBL v.61 gene annotations using the program Cuffcompare (see caption to Table S4 for more details).

Figure S6.. Number of event variations due to small differences in exon and intron boundaries as a fraction of the total number of events, with varying cutoffs for the allowable difference (V bp).

Values shown are for exon skipping (ES) events derived from the Illumina Human Body Map data (‘Cufflinks’) and from Ensembl v.61 gene annotations, from columns 4 and 8 in Table S6.

Figure S8.. Histograms of reads in the 16 tissue samples supporting the two introns flanking the alternatively spliced exon (blue, red) and the intron spanning the exon (green), respectively, for the 26,989 identified exon skipping events.

( A) Diagram of an exon skipping event, illustrating the three types of introns: left – blue, right – red, and intron spanning – green. ( B) Read histograms for the three categories of introns. The x-axis represents the number of supporting reads for an intron, grouped in bins, and the y-axis shows the percentage of introns by levels of supporting reads (in bins). Most events have deep support (>100 reads) for the flanking introns, and to a lesser extent for the spanning intron.

Figure S10.. Novel exon skipping event at the human ASB15 gene locus.

(Top) Exon chr7:123,257,633–123,257,718, present in heart (heart.653543.1 and heart.653543.2) and with partial support in skeletal muscle tissue (skel_muscle.206929.2), is novel. Red arrows point to the exon appearing in two isoforms reconstructed in the heart sample, and to the (potentially) partial form in a skeletal muscle transcript for the ASB15 gene. The spanning intron in the heart sample is also novel. Additionally, we found a potentially retained intron (chr7:123,269,489–123,270,019; thyroid.843995.1), circled in red, whose 531 bp sequence is conserved across multiple vertebrates. (Bottom) Read support for the putative intron retention event above in the thyroid sample, whereas the flanking introns are devoid of intronic reads. RefSeq exon annotations are shown in blue.

Figure S12.. Expression levels of genes for the 13,946 events not comparable between adrenal and adipose tissues.

2,085 (15.0%) event genes are not expressed (FPKM<0.1), whereas 3,487 (33.7%) have FPKM values less than 10.0 and therefore may be incompletely reconstructed, which can cause a splice form to be missed.

Figure S14.. Classes of alternative splicing events detected by ASprofile by pairwise transcript comparisons.

( A) Exon skipping (SKIP) and cassette exons (MSKIP); ( B) retention of single (IR) and multiple (MIR) introns; ( C) alternative exon ends (AE); ( D) alternative transcription start site (TSS); and ( E) alternative transcription termination site (TTS). Alternatively spliced features are shown in red.