1. Semi-automated background removal limits loss of data and normalises the images for downstream analysis of imaging mass cytometry data
- Author
-
de Miranda Nf, Boudewijn P. F. Lelieveldt, Antonios Somarakis, Thomas Höllt, and Marieke E. Ijsselsteijn
- Subjects
Text mining ,business.industry ,Computer science ,Tissue Processing ,Mass cytometry ,Computational biology ,Signal intensity ,Tissue morphology ,business - Abstract
Imaging mass cytometry (IMC) allows the detection of multiple antigens (approximately 40 markers) combined with spatial information, making it a unique tool for the evaluation of complex biological systems. Due to its widespread availability and retained tissue morphology, formalin-fixed, paraffin-embedded (FFPE) tissues are often a material of choice for IMC studies. However, antibody performance and signal-to-noise ratio can differ considerably between FFPE tissues as a consequence of variations in tissue processing, including fixation. We investigated the effect of immunodetection-related signal intensity fluctuations on IMC analysis and phenotype identification in a cohort of twelve colorectal cancer tissues. Furthermore, we explored different normalisation strategies and propose a workflow to normalise IMC data by semi-automated background removal, using publicly available tools. This workflow can be directly applied to previously obtained datasets and considerably improves the quality of IMC data, thereby supporting the analysis and comparison of multiple samples.
- Published
- 2020
- Full Text
- View/download PDF