Computational approach to discriminate human and mouse sequences in patient-derived tumour xenografts.

摘要: Patient-Derived Tumour Xenografts (PDTXs) have emerged as the pre-clinical models that best represent clinical tumour diversity and intra-tumour heterogeneity. The molecular characterization of PDTXs using High-Throughput Sequencing (HTS) is essential; however, presence mouse stroma challenging for HTS data analysis. Indeed, high homology between two genomes results in a proportion reads being mapped human. In this study we generated Whole Exome (WES), Reduced Representation Bisulfite (RRBS) RNA sequencing (RNA-seq) from samples with known mixtures human DNA or cohort breast cancers their derived PDTXs. We show an silico Combined human-mouse Reference Genome (ICRG) alignment discriminates up to 99.9% accuracy decreases number false positive somatic mutations caused by misalignment >99.9%. also model estimate content independent PDTX samples. For RNA-seq RRBS analysis, use ICRG allows dissecting computationally transcriptome methylome cells stroma. direct comparison previously reported approaches, our method showed similar higher while requiring significantly less computing time. computational pipeline describe here valuable tool analysis well any other mixture species.