FAQ  •  Register  •  Login

Standardising exported data with different channels

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 :-)
<<

levelup

Participant

Posts: 3

Joined: Wed Jan 30, 2019 4:53 pm

Post Thu Jan 31, 2019 10:17 am

Standardising exported data with different channels

Hi,
I’m been having a real headache trying to analyse samples which have had a few new channels added to the sample data upon acquisition. This has resulted in basically two panels which need to be analysed side by side.
I have very basic understanding of R, the analysis I’ve been trying to perform has been on CyTOFkit.
I’ve attempted several different ways of standardising the fcs files, transforming into excel, deleting extra channels, and standardising channel names, then saving as FCS with no luck. I’ve tried using Premessa R program which was recommended here, this was useful as I was able to remove extra channels but for some reason channel names changed between different sample acquisition, and even using Premessa and changing channel names I’m still not getting a unified panel to allow me to analyse all of my samples.
I’ve attached an image to show the problem, even after removing discrepant channels and changing the names of the channels to match the samples are still not standardised!
Any help would be much appreciated!
Thank you.

Before Premessa
https://tinyurl.com/ycffreug

After Premessa
https://tinyurl.com/ycd9arqx
<<

jamesaries

Participant

Posts: 14

Joined: Thu Sep 22, 2016 2:49 pm

Post Thu Jan 31, 2019 1:18 pm

Re: Standardising exported data with different channels

Hello,

I use CyTOFkit and have found CyTOF CORE (in R) helpful for standardising files.

I have a template file and then standardised all files to this one and it works well (i.e. CyTOFkit functions well).

James
<<

levelup

Participant

Posts: 3

Joined: Wed Jan 30, 2019 4:53 pm

Post Thu Jan 31, 2019 3:36 pm

Re: Standardising exported data with different channels

Ok, thank you for the reply. I'll have a look at the CyTOF Core package!
<<

sgranjeaud

Master

Posts: 123

Joined: Wed Dec 21, 2016 9:22 pm

Location: Marseille, France

Post Thu Jan 31, 2019 4:05 pm

Re: Standardising exported data with different channels

Hi,

We have identified a few packages to manage harmonization of FCS.
Here is our doc (french) http://impact-cyto.inserm.fr/harmoniser-des-panels/.
Untranslated it yet. Try Google, deepl...
Nevertheless, the final example (end of the doc) is in english.

Best.

To summarize

ChangeFCSParameters allows to answer a simple and punctual need. cytutils presents a synthetic report of the channel-marker association and allows to rename the channel and the marker. premessa proposes an interface for the renaming of the marker only and the possibility to delete channels. cytofCore allows to rename the marker associated to a channel (but not the channel), but especially to delete channels and to create channels (with a zero intensity) by guaranteeing that all the FCS files have the same channels with the same names and in the same order.

premessa is probably the simplest way (because it is graphical) to reconcile 2 panels by keeping only the common channels and renaming the markers. The combination of cytuils + cytofCore is more powerful because of its ability to rename channels and markers, and its ability to add channels.

Tool selection by strategy

Use cytutils for an overview of channels and markers

Change the name of the marker for each channel ==> cytutils/premessa/cytofcore

Decide which channels should be renamed ==> cytutils

Decide if channels are considered identical: for example a change of metal but no marker ==> cytutils

Decide if the reconciled panel contains the intersection of channels (or less) ==> premessa/cytofcore, or if missing channels should be added ==> cytofcore
<<

levelup

Participant

Posts: 3

Joined: Wed Jan 30, 2019 4:53 pm

Post Fri Feb 08, 2019 3:49 pm

Re: Standardising exported data with different channels

sgranjeaud wrote:Hi,

We have identified a few packages to manage harmonization of FCS.
Here is our doc (french) http://impact-cyto.inserm.fr/harmoniser-des-panels/.
Untranslated it yet. Try Google, deepl...
Nevertheless, the final example (end of the doc) is in english.

Best.

To summarize

ChangeFCSParameters allows to answer a simple and punctual need. cytutils presents a synthetic report of the channel-marker association and allows to rename the channel and the marker. premessa proposes an interface for the renaming of the marker only and the possibility to delete channels. cytofCore allows to rename the marker associated to a channel (but not the channel), but especially to delete channels and to create channels (with a zero intensity) by guaranteeing that all the FCS files have the same channels with the same names and in the same order.

premessa is probably the simplest way (because it is graphical) to reconcile 2 panels by keeping only the common channels and renaming the markers. The combination of cytuils + cytofCore is more powerful because of its ability to rename channels and markers, and its ability to add channels.

Tool selection by strategy

Use cytutils for an overview of channels and markers

Change the name of the marker for each channel ==> cytutils/premessa/cytofcore

Decide which channels should be renamed ==> cytutils

Decide if channels are considered identical: for example a change of metal but no marker ==> cytutils

Decide if the reconciled panel contains the intersection of channels (or less) ==> premessa/cytofcore, or if missing channels should be added ==> cytofcore


Thank you very much for this, I will try to implement your workflow this weekend!

Cheers!
<<

dtelad11

Master

Posts: 129

Joined: Mon Oct 31, 2016 6:26 pm

Post Fri Feb 08, 2019 8:17 pm

Re: Standardising exported data with different channels

I'm one of the developers behind cytutils -- feel free to reach out at el.ad.david.amir@gmail.com if you need help with the utility.

Return to CyTOF data analysis

Who is online

Users browsing this forum: No registered users and 9 guests

cron