Lectures and seminars AutoSpectral: An optimized unmixing workflow

Hosts: Petter S. Woll and Maria Johansson

If you don't have access to Biomedicum, meet us at the reception in BioMedicum at 14.55.  

About AutoSpectral

AutoSpectral is an R tool that aims to provide an optimized, complete pipeline to make the best use of the information in the single-stained controls and reproducibly generate better unmixing. 

The tool provides methods for extracting the cleanest spectral signatures from controls, reducing contamination by autofluorescent cell “noise”. This process automates several key steps that are time-consuming and difficult if performed manually, with the aim of following best practices for unmixing. 

When it comes to the unmixing, AutoSpectral accounts for cell-to-cell variation in the data to reduce spread and skewing errors. The cellular background, or autofluorescence, varies by cell type in both magnitude and quality. The key insight in AutoSpectral is that this variability is quantifiable and can be used to improve the unmixing if we provide the distribution of possibilities during the unmixing. 

In the talk, I will explain how autofluorescence “contaminates” our fluorophore data in unmixing, and how we can go about identifying the best-fitting autofluorescence spectrum on a per-cell basis.

Bio

I got my PhD at the University of Cambridge, studying the ability of microbial products to modulate autoimmune diabetes under the supervision of Prof. Anne Cooke. I fell in love with flow cytometry and spent quite a bit of time with Nigel Miller, the head of the flow core facility. After that, I began my post-doctoral studies with Dr. Hans Oettgen at Boston Children’s Hospital in Boston, Massachusetts, USA. 

In my post-doctoral work, I focused on the aetiology and treatment of food allergies, working primarily with mouse models. In 2017 I joined Adrian Liston’s lab, at the time in Leuven, Belgium, where I was employed to develop high-throughput, high-parameter flow panels for the analysis of tissue-resident Tregs and brain-resident immune cells. 

In 2019, the lab moved to the Babraham Institute in Cambridge, England, where I was employed as a staff scientist, and in 2023, the lab moved again to the Department of Pathology at the University of Cambridge. In my day-to-day work, I focus on training people in best practices in flow cytometry, developing new panels and techniques for our experiments, and working closely with bioinformaticians to develop new ways of getting meaning out of cytometry data.