RCANE: a deep learning algorithm for whole-genome pan-cancer somatic copy number aberration prediction using RNA-seq...

Area of Science:

• Genomics
• Bioinformatics
• Cancer Research

Background:

• Transcriptome sequencing (RNA-seq) is crucial for cancer research, analyzing gene expression.
• Somatic copy-number aberrations (SCNAs) are key drivers of cancer development.
• Inferring SCNAs from RNA-seq offers a cost-effective alternative to DNA assays.

Purpose of the Study:

• To introduce RCANE, a deep-learning framework for predicting genome-wide SCNAs using only RNA-seq data.
• To demonstrate RCANE's effectiveness across diverse cancer types.
• To provide a scalable and robust solution for SCNA profiling.

Main Methods:

• Developed a deep-learning framework named RCANE.
• Trained RCANE on The Cancer Genome Atlas (TCGA) and DepMap cell-line cohorts.
• Evaluated RCANE's performance against existing SCNA inference methods.

Main Results:

• RCANE accurately predicts genome-wide SCNAs from RNA-seq data.
• The framework demonstrates superior performance compared to current approaches.
• RCANE provides a scalable and robust solution for SCNA profiling.

Conclusions:

• RCANE enhances SCNA profiling using only RNA-seq data.
• This method improves cancer diagnostics and therapeutic decision-making.
• RCANE offers a valuable tool for cancer research and clinical applications.