Poster #87 - Eloise Gaudet

Characterizing Transcriptional Landscape in Gliomas with Oxford Nanopore Long-Read Sequencing/Translational Bioinformatics

Eloise Gaudet(BSE)1,2 Karleena Rybacki(MS)1,3 Joe Chan(BS)3 Yuanquan Song(Ph.D)3,4,5 Marilyn Li(MD, MS)3,5,6 Kai Wang(Ph.D)1,3,5 1. Department of Bioengineering, University of Pennsylvania, Philadelphia, PA, USA 2. Department of Computer and Information Science, University of Pennsylvania, Philadelphia, PA, USA 3. Raymond G. Perelman Center for Cellular and Molecular Therapeutics, Children's Hospital of Philadelphia, Philadelphia, PA, USA 4. Neuroscience Graduate Group, University of Pennsylvania, Philadelphia, PA, USA 5. Department of Pathology and Laboratory Medicine, University of Pennsylvania, Philadelphia, PA, USA 6. Division of Genomic Diagnostics, Children's Hospital of Philadelphia, Philadelphia, PA, USA

Transcriptional complexities, such as gene fusions, novel isoforms and alternative splicing events, play crucial roles in cancer development. While previous short-read RNA sequencing of 49 high- and low-grade glioma samples using the CHOP Fusion Panel was negative for gene fusions, this traditional short-read RNA sequencing approach faces limitations in detecting complete transcript structures, complex splicing patterns, novel isoforms, and potentially novel gene fusions. To address these limitations and comprehensively characterize transcriptome complexity associated with glioma, we performed Oxford Nanopore Technologies long-read RNA sequencing and isoform detection (IsoQuant and FLAIR) on the same 49 glioma samples. Using 22 brain tissues from GTEx (v9) as controls, differential expression analysis (DESeq2, edgeR, and DRIMSeq) was used to identify significant isoform-level alterations in glioma. This analysis revealed 1,159 significantly altered transcript isoforms from 911 genes (IsoQuant) and 895 isoforms from 639 genes (FLAIR), relative to controls. Many of these isoforms exhibited structural changes, including exon skipping, premature stop codons, and truncated transcripts, with potential implications for protein function. Notably, a highly overexpressed truncated protein-coding transcript isoform of CTSB, a gene implicated in glioma development and poor prognosis, was found largely in high-grade glioma. Analysis of cancer-relevant gene sets revealed significant isoform alterations in CHOP Fusion Panel genes, COSMIC Cancer Census genes, and essential genes identified in CRISPR knockout screens in glioma-related cancer cell lines from the Sanger Dependency Map, such as TPM3, RACK1, and MEAF6. We also identified novel, previously unannotated, splice variants in genes such as GFAP and SQSTM1, with pathway analysis implicating their association to glioma progression and neurodegenerative disorders. These findings demonstrate the advantages of using long-read sequencing, which enabled the detection of differentially expressed and novel transcripts, including biologically relevant splicing events that may contribute to glioma pathogenesis.