Microbiome Signatures in Host - Spatial and Single Cell Transcriptomics

Zurich Seminars in Bioinformatics - Stine Anzböck

  • 12:15 UZH Irchel Y55-l-06/08 and ZOOM Call

Abstract The tumor-associated microbiome is an integral part of the tumor microenvironment and holds great promise for a better understanding of carcinogenesis, response to therapy and early diagnosis of cancer. However, current methods limit the understanding of the spatial distribution of the microbiome in the tumor and their interaction with host cells. To this end, we have developed SpaceMicrobe, a novel computational framework to detect and profile the microbiome in 10X Visium Spatial Gene Expression data, a widely used commercially-available spatial transcriptomics technology. We demonstrate the application of our framework on a colorectal cancer spatial transcriptomics dataset and showed the spatial distribution of putative microbiome biomarkers for CRC, such as Fusobacterium nucleatum and Parvimonas micra, as well as the detection of tumor regions with Epstein-Barr virus (EBV) and its associated impact on the host transcriptome. Furthermore, we show that SpaceMicorbe can be applied to additional 10X technologies such as 10X Chromium single-cell RNA sequencing data by profiling colon biopsies from an ulcerative colitis dataset.

Applying SpaceMicrobe on more spatial and single-cell transcriptomics tumor biopsy datasets may help us better understand the tumor-associated microbiome in different cancers and help contribute to developing new strategies for cancer biomarker discovery and patient stratification based on the tumor-associated microbiome.