Prediction of transcript structure and concentration using RNA-Seq data.

Area of Science:

• Genomics
• Molecular Biology
• Bioinformatics

Background:

• Alternative splicing (AS) significantly increases proteomic diversity, with ~90% of human genes undergoing this process.
• Aberrant AS transcripts are implicated in diseases including breast cancer, lung cancer, and glioblastoma.
• Identifying novel AS transcripts offers potential for new therapeutic and prognostic targets in drug discovery.

Purpose of the Study:

• To develop an R package, Finding Alternative Splicing Events (FASE), for identifying and quantifying differential alternative splicing events.
• To predict the precise structure and relative expression of transcripts under various conditions.
• To enable the discovery of novel transcripts and AS-based biomarkers.

Main Methods:

• FASE integrates AS events with exon, intron, and junction data using graph theory to predict transcript structures.
• It estimates transcript concentration as relative expression in log2CPM (counts per million).
• The pipeline was applied to TCGA-BRCA data to identify unique transcripts.

Main Results:

• FASE successfully identified unique transcripts for EMILIN1 and SLK genes in TCGA-BRCA data.
• Experimental validation using RT-PCR confirmed the accuracy and precision of the identified transcripts.
• The study demonstrated FASE's capability to detect novel transcripts often missed by conventional methods.

Conclusions:

• The FASE pipeline accurately predicts novel transcripts and their expression levels.
• It serves as a valuable tool for identifying AS-based biomarkers and therapeutic targets across various genomic scales and experimental conditions.
• FASE enhances the discovery of disease-associated transcripts for improved diagnostics and therapeutics.