Benchmarking mouse contamination removing protocols in patient-derived xenografts genomic profiling

Area of Science:

• Genomics
• Cancer Research
• Bioinformatics

Background:

• Patient-derived xenograft (PDX) models are crucial for cancer research.
• Genomic and transcriptomic profiling of PDX models are often contaminated by mouse sequencing reads.

Purpose of the Study:

• To investigate the impact of mouse read contamination on RNA sequencing (RNAseq), Whole Exome Sequencing (WES), and Whole Genome Sequencing (WGS) data from PDX models.
• To benchmark computational protocols for removing mouse reads from PDX data.

Main Methods:

• Analysis of RNAseq, WES, and WGS data from 21 PDX models.
• Systematic benchmarking of 12 computational protocols for mouse read removal.
• Evaluation using gold standard datasets.

Main Results:

• Mouse read contamination increases expression of immune and stromal genes and inflates somatic mutation counts.
• Detection of gene fusions and copy number alterations is minimally affected.
• Pseudo-alignment protocols showed superior performance and efficiency.
• Xengsort was identified as the best-performing tool.

Conclusions:

• Removing mouse reads is essential for accurate genomic and transcriptomic analysis of PDX models.
• Adopting advanced tools like Xengsort is recommended for PDX genomic studies.