Normalization across ROIs in different sample batches

Looking for ways to analyze your dataset? Need help with a software package?
Computational image analysis of multiplexed Imaging Mass Cytometry acquisitions
Post Reply
Posts: 3
Joined: Fri Dec 20, 2019 1:29 pm

Normalization across ROIs in different sample batches

Post by MalteLehmann » Thu Sep 24, 2020 10:57 am

Hey everybody,

following problem: we measured a batch of samples derived from our pathology bank. Further on in the analysis, we found that the older samples were probably fixed longer or less time in PFA resulting in a lower overall staining (Except the Iridium channels of course). An Epithelial marker, for example, that should be evenly distributed, shows a mean intensity of 3,6 in a newer sample, compared to 0.2 in an older one. However, from the biological point of view they should be about the same.

This results in a bunch of problems along the way, since, for example, those epithelial cells will be differently clustered in a tSNE, or Phenograph, respectively, due to their different expression of the epithelial marker.

Can you recommend any script or processing step before loading the data into HistoCAT, to cope with this problem?

A similiar question has been raised here, but since the post was already a year old I created a new one. They can be joined if you feel that is better.

Thank you very much for your help!

Post Reply