Figure 3. PRMT5 inhibition leads to increased intron retention and exon skipping events.

(A) Distribution of isoforms across different structural categories and their regulation status. Isoforms significantly up-regulated in PRMT5-inhibited samples (P < 0.05) are shown in red; significantly down-regulated isoforms are shown in blue. The x-axis represents the number of isoforms, and the y-axis indicates structural subcategories as defined by the SQANTI classification system. Reference_match, alternative_5end, alternative_3end, and alternative_3end5end correspond to isoforms with known exon structures that match the reference transcript model. 5prime_fragment, 3primefragment, mono-exon, and intron_retention represent isoforms that lack terminal exons but still align to known splice junctions. Combination_of_known_junctions and combination_of_known_splicesites refer to isoforms that use only annotated splice sites in new combinations. At_least_one_novel_splicesite and multi-exon represent isoforms that include at least one previously unannotated splice site. (B) Comparison of up-regulated (red) and down-regulated (blue) isoforms across various structural categories. Statistical significance is annotated on the bars using asterisks, based on P-values from a chi-squared test. Asterisks indicate statistical significance: P < 0.05 (*), P < 0.01 (**), and P < 0.001 (***). (C) Dot plot visualizing Gene Ontology (GO) enrichment analysis for isoforms with increased intron retention. The x-axis represents the gene ratio (the proportion of input genes associated with each GO term). The y-axis lists the enriched GO-term descriptions. Dot size corresponds to the number of genes associated with each term, whereas the dot color represents the adjusted P-value, with the red-blue gradient indicating significance levels. Terms are sorted by gene ratio for clarity. (D) Cytoscape Enrichment Map of significantly enriched gene sets in all up-regulated intron retention containing isoforms. Gene sets are colored based on −log₁₀ (P-value) with darker red indicating lower P-values. Clusters were completed using AutoAnnotate and are labeled based on the pathway in each cluster with the lowest adjusted P-value.