Network-based approach to identify principal isoforms among four cancer types

Area of Science:

• Genomics
• Molecular Biology
• Cancer Research

Background:

• Protein isoforms, arising from alternative splicing, can have diverse biological functions.
• Isoform expression is often altered in tumors, impacting tumorigenesis and prognosis.
• Previous studies identified principal isoforms based on normal tissue interactions.

Purpose of the Study:

• To identify higher degree isoforms (HDIs) of multi-isoform genes (MIGs) in various cancer types.
• To compare cancer-derived HDIs with those identified by proteomic and prognostic data.
• To analyze associations between HDIs, non-HDIs (nHDIs), gene expression, and drug responses in cancer cell lines.

Main Methods:

• Applied a meta-analytical framework to RNA-seq data from breast, lung, leukemia, and colon cancer cell lines.
• Calculated isoform-isoform co-expression to identify HDIs within multi-isoform genes (MIGs).
• Compared identified HDIs with proteomic data, prognostic/predictive evidence, and analyzed transcript expression and drug responses.

Main Results:

• Identified complex properties of HDIs within genes across different cancer types using co-expression networks.
• Found distinct associations between HDIs and nHDIs concerning transcript expression and drug sensitivity.
• Highlighted the potential of HDIs as novel therapeutic targets in cancer treatment.

Conclusions:

• Isoform co-expression networks in cancer reveal complex patterns of higher degree isoforms (HDIs).
• HDIs exhibit distinct relationships with gene expression and drug responses compared to non-HDIs.
• HDIs represent promising novel therapeutic targets for developing targeted cancer therapies.