Detecting Somatic Insertions/Deletions (Indels) Using Tumor RNA-Seq Data.

Area of Science:

• Genomics
• Bioinformatics
• Computational Biology

Background:

• Somatic indel identification is crucial for cancer genomic analysis but difficult using tumor-only RNA sequencing (RNA-Seq).
• Challenges include the absence of matched normal samples and complex read alignment involving splice junctions and indels.
• Existing methods are limited for accurate indel detection in RNA-Seq data.

Purpose of the Study:

• To introduce RNAIndel, a novel software tool for identifying somatic coding indels from tumor-only RNA-Seq data.
• To address the limitations of current methods in detecting indels in cancer transcriptomes.

Main Methods:

• RNAIndel utilizes indel realignment techniques specific to RNA sequences.
• A machine learning model is employed to classify coding indels as somatic, germline, or artifactual.
• The tool is designed to handle the complexities of RNA-Seq read alignment.

Main Results:

• RNAIndel demonstrates high accuracy in identifying somatic coding indels.
• Validation was performed using RNA-Seq data from diverse tumor types.
• The software effectively distinguishes true somatic indels from sequencing artifacts and germline variants.

Conclusions:

• RNAIndel provides a robust solution for somatic indel detection in tumor-only RNA-Seq.
• This tool enhances the capability of cancer genomic analysis without requiring matched normal samples.
• RNAIndel's validated accuracy makes it a valuable resource for cancer research.