FAQ  •  Register  •  Login

CyTOF normalisation using reference PBMCs

Forum rules
Please be as geeky as possible. Reference, reference, reference.
Also, please note that this is a mixed bag of math-gurus and mathematically challenged, so choose your words wisely :-)
<<

Ondrej

Participant

Posts: 5

Joined: Fri Feb 16, 2018 12:27 pm

Post Fri Feb 16, 2018 3:22 pm

CyTOF normalisation using reference PBMCs

Hello,

I stained thawed "reference" barcoded PMBCs from the same blood donor together with every tested patient sample as a method for future data normalization across different patients and days of measurement. I was wondering if anyone ever used such approach and whether there is already a script how to normalize fcs files across all channels based on reference PBMCs?

I found this paper by Kleinsteuber et al. (https://www.ncbi.nlm.nih.gov/pubmed/?te ... er+corleis) where they used similar approach but only for setting up population gates all across samples, not for normalizing the signal in each channel allowing more complex analysis (such as t-SNE).

I am aware of the beads based normalization by Rachel Fink et al. (https://www.ncbi.nlm.nih.gov/pubmed/23512433). Has anyone tried to modify their method for PBMCs as a standard?

Many thanks for your comments and help!

Ondrej
<<

mleipold

Guru

Posts: 5796

Joined: Fri Nov 01, 2013 5:30 pm

Location: Stanford HIMC, CA, USA

Post Fri Feb 16, 2018 4:52 pm

Re: CyTOF normalisation using reference PBMCs

Hi Ondrej,

The closest I've seen to sample-based Batch Correction was Jeff Hokanson's CLEAN algorithm that he presented at CYTO 2016 in Seattle (see p. 153, here: http://cytoconference.org/cyto/pdfs-doc ... 6-FINAL-(1).aspx)

Unfortunately, I haven't seen that it's been published, and according to his GitHub page (https://github.com/jeffrey-hokanson), he's no longer at MD Anderson.

Though the page does apparently have his CYTO 2016 talk: https://github.com/jeffrey-hokanson/old ... /downloads

One downside to CLEAN is that the BatchCorrectionControl had to be *included in* every sample (like Katja and and you have done), not just a control sample on every plate. This was done in the context of Barcoding, and as I recall was a strict requirement.....the Plate Control sample would not suffice.


Mike
<<

Ondrej

Participant

Posts: 5

Joined: Fri Feb 16, 2018 12:27 pm

Post Mon Feb 19, 2018 10:58 am

Re: CyTOF normalisation using reference PBMCs

Hi Mike,

thanks a lot for this reference! It is indeed exactly what I was looking for! I might try to get in touch with Jeff somehow and get more details. I have barcoded and included the reference control in every patient's sample I have measured so should work ok.

Best wishes,

Ondrej
<<

vtosevski

Contributor

Posts: 44

Joined: Wed Nov 20, 2013 12:50 pm

Location: Zurich, Switzerland

Post Mon Feb 19, 2018 1:20 pm

Re: CyTOF normalisation using reference PBMCs

Hi Ondrej,

Last year at Cyto Sofie Van Gassen also presented a nice work on an algorithm aiming to do exactly what you need. I believe the algorithm has a name by now (CytofNorm) but hasn't been made public yet. However, if you reach out to Sofie or Yvan, they will probably provide the code directly to you to give it a try ahead of time.

Best,
Vinko
<<

Ondrej

Participant

Posts: 5

Joined: Fri Feb 16, 2018 12:27 pm

Post Mon Feb 19, 2018 3:15 pm

Re: CyTOF normalisation using reference PBMCs

Hi Vinko,

many thanks for this suggestion - I will contact Sofie too and see where they are up to with CytofNorm. I realize from all the comments that Cyto seems to be a pretty useful event so will try to go this year as well.

Best wishes,

Ondrej
<<

jamesaries

Participant

Posts: 14

Joined: Thu Sep 22, 2016 2:49 pm

Post Mon Feb 19, 2018 3:23 pm

Re: CyTOF normalisation using reference PBMCs

This looks interesting.

Ondrej - I am / will be staining up to 250 samples so I would be interested to know your experience using the PBMC approach. I'll PM you.

James

Return to CyTOF data analysis

Who is online

Users browsing this forum: No registered users and 11 guests