Systematic identification of intron retention associated variants from massive publicly available transcriptome...

Area of Science:

• Genomics and Bioinformatics
• Molecular Biology
• Computational Biology

Background:

• Genomic variants often cause disease by altering gene splicing.
• Identifying splicing-associated variants is crucial for genomic medicine.
• Existing methods typically require both genomic and transcriptomic data, which are not always available.

Purpose of the Study:

• To develop a novel bioinformatics methodology for detecting splicing-associated variants using only transcriptome sequencing data.
• To identify intron retention associated variants (IRAVs) from large-scale public transcriptomic datasets.
• To pinpoint putative disease-associated IRAVs by integrating with existing variant databases.

Main Methods:

• Developed a computational framework to detect intron retention from transcriptome sequencing data alone.
• Evaluated the sensitivity and precision of the developed methodology.
• Applied the method to 230,988 public transcriptome datasets and cross-referenced identified variants with disease databases.

Main Results:

• Identified 27,049 intron retention associated variants (IRAVs) from 230,988 transcriptome datasets.
• Discovered 3,000 putative disease-associated IRAVs, including variants linked to cancer and autosomal recessive disorders.
• Demonstrated the feasibility of an in-silico framework for automated medical knowledge discovery from public sequencing data.

Conclusions:

• A novel method enables the detection of splicing-associated variants using only transcriptome data, overcoming data availability limitations.
• The identified IRAVs provide a valuable resource for understanding genetic variant impact on gene function and disease.
• The IRAVDB resource (https://iravdb.io/) offers a collection of identified IRAVs for further research.