FAQ  •  Register  •  Login

Acquisition Templates - how do you handle these?

<<

jimbomahoney

Master

Posts: 73

Joined: Wed Feb 27, 2019 11:21 am

Post Wed Nov 20, 2019 3:43 pm

Acquisition Templates - how do you handle these?

Hi all,

How do you currently program the acquisition template into the CyTOF software?

e.g. I'm thinking of the following scenarios / options:

1) User sits with operator @ CyTOF and verifies that acquisition panel is correct.
2) User creates template using "offline" copy of CyTOF software and operator loads the .tem file from this.
3) User provides operator with Excel template.
4) User simply e-mails a list to the operator.

In cases 1 & 2, the user is thus responsible for ensuring their panel is correct.
In cases 3 & 4, the operator could be blamed for any mistakes.

Case #1 = Probably easiest and I would guess most common?
Case #2 = Easiest for operator. Potentially annoying for user.
Case #3 = Potentially annoying for Operator, easy for user.
Case #4 = Potentially very annoying for operator. Easiest for User?

I've devised a method that I hope achieves the following:

1) Easy for user.
2) Provides some guidance / basic checks to both the user and operator (e.g. if they've forgotten to add Ir / Pt / bead channels / all barcodes, presence of illegal characters).
3) Easy for operator (load a .tem file, rather than program the template in the CyTOF software).
4) Ensures that the user is primarily responsible for ensuring panel is correct.

UPDATE - web app built and tested - see post below.

So far, I've built a spreadsheet that the user fills out by marking which isotopes are in their panel and what they want them named / labelled.
It includes all the existing Fluidigm isotopes (including the new Cd).
It's "pre-loaded" with the compulsory bead channels, Ir and Pt, but the user could change these if they wish.
It provides some basic instructions to the user, as well as some "sanity checks" (e.g. the illegal symbols in CyTOF software are found and the user alerted)
Once sent to the operator, an R-script reads the Excel file, re-runs the sanity checks and spits out a .tem file that the operator simply opens on the CyTOF.

Image

Intrigued at other's input to what method they currently use and whether my method would be useful for anyone else? (I don't know how common R would be for operators, but I've found it both useful and fun!). :geek:
Last edited by jimbomahoney on Tue Nov 26, 2019 5:12 pm, edited 1 time in total.
<<

dahern

Participant

Posts: 12

Joined: Tue Mar 31, 2015 3:45 pm

Post Wed Nov 20, 2019 4:11 pm

Re: Acquisition Templates - how do you handle these?

A combination of option 3 and 4 for us. The user will email us either a list or excel file (with the masses in order from low to high) of their panel.
We then make the antigen name additions to a standard default template that has all the usable channels already selected and named as Blank1....Blank xx for the antigen channels and already filled in with Barcode1-6, live/dead, iridium1 etc. We would always then collect all the channels in the default template even if they are not being used by that users panel. One potential downside of your method is when users are starting out with their experiments and might not be using every channel from 141-176, these channels still have valuable information for them while they work up their panel where they can see any potential spillover effects etc. Particularly for users doing any titration/validation experiments, some of the most valuable information is in the channels they are not using and you might be missing out on this. Perhaps a tweak so your script names all the channels in the 141-176 range correctly but then also names those in that range not in use as Blank1...Blank x. So these channels are still always collected.
<<

jimbomahoney

Master

Posts: 73

Joined: Wed Feb 27, 2019 11:21 am

Post Thu Nov 21, 2019 8:01 am

Re: Acquisition Templates - how do you handle these?

dahern wrote: Perhaps a tweak so your script names all the channels in the 141-176 range correctly but then also names those in that range not in use as Blank1...Blank x. So these channels are still always collected.


Thanks!

That should be easy to do.
<<

jimbomahoney

Master

Posts: 73

Joined: Wed Feb 27, 2019 11:21 am

Post Thu Nov 21, 2019 9:27 am

Re: Acquisition Templates - how do you handle these?

OK, that's done, but thinking about this - wouldn't it be sufficient / better to simply add any n+1 channels, rather than everything in the range 141-176?

My understanding is that the n+1 is the most likely spillover due to a very strong signal in the n channel.

The n+16 may also be useful, due to oxidation?

I realise there are also n+isotope impurity, but that's going to be very difficult to program due to the variation in isotopic impurities!
<<

mleipold

Guru

Posts: 2086

Joined: Fri Nov 01, 2013 5:30 pm

Location: Stanford HIMC, CA, USA

Post Thu Nov 21, 2019 3:29 pm

Re: Acquisition Templates - how do you handle these?

I feel the need to clarify a common misunderstanding.

A lot of people think about "M+1" as the major spill, but that's not precisely correct. The usual major spill is the *next mass isotope*.

Yes, in most cases, this is M->M+1. However, not every element is that way. For example, Europium is 151 and 153: there is no 152 isotope of Eu (well, at least stable isotope, maybe an accelerator could make it).

There's also an issue of what the starting abundance is in nature. For example, natural Indium is~3% 113, 97% 115. So, you can easily get 99.9% 115 from Trace, but the 113 you buy is only 93% or so. Anything higher would be stupid expensive.

This is also why Pd102 is 10x the price of the other Pd isotopes, and is often less pure (more total non-102 levels).
<<

jimbomahoney

Master

Posts: 73

Joined: Wed Feb 27, 2019 11:21 am

Post Thu Nov 21, 2019 3:45 pm

Re: Acquisition Templates - how do you handle these?

Thanks Mike - agreed.

I believe my previous comment is accurate in that regard?

(M+1 = abundance sensitivity; M+16 = oxidation; M+whatever = isotope impurities)

I suppose programming the isotope channels around the user-selected would be possible...

I already created an Excel lookup spreadsheet for that...

Image
<<

mleipold

Guru

Posts: 2086

Joined: Fri Nov 01, 2013 5:30 pm

Location: Stanford HIMC, CA, USA

Post Thu Nov 21, 2019 7:23 pm

Re: Acquisition Templates - how do you handle these?

In my experience, at least with the Helios instruments, Abundance sensitivity is seldom a problem since the arrival times are more uniform for a given mass.

However, Abundance sensitivity technically hits not only M+1 but also M-1: it's basically left-leg (M-1) and right-leg (M+1) spillover of the desired channel (M), which is an ion peak of a given width. You are more likely to get it at high mass: Cs133 is a tighter, more uniformly shaped peak than Ir193.


The only times I've ever seen abundance sensitivity left/right leg spillover has been when my signal at M is >1e4 Dual. Once it happened because of a contaminant in the sample (La139, spilling into both 138 and 140) (yes, there's a super minor La138 isotope in nature, but the 138 and 140 spills were of similar magnitude, implying it was an abundance sensitivity issue). And one time I over-stained my Ir signal such that the right leg of Ir193 wiped out my barcode signal in the Pt194 channel.


Mike
<<

jimbomahoney

Master

Posts: 73

Joined: Wed Feb 27, 2019 11:21 am

Post Fri Nov 22, 2019 2:05 pm

Re: Acquisition Templates - how do you handle these?

Ok, with all of this in mind, I've designed a web-based Template creator that should perform all these functions.

Please feel free to try it and provide feedback?

https://jimbomahoney.shinyapps.io/shiny/

UPDATE 24th Nov - OK, now has upload from Excel functionality. I've also tested to ensure that the add/remove buttons don't accidentally add/remove anything that the user may have added from the Excel file.

I think it's working pretty well! I'll also ask some users in my centre to test it. I used one of the panels they had already designed in Excel. It worked nicely and using the "Add Blanks" feature adds more blanks than they had added manually, as well as adding the 140 channel, which is necessary for normalisation and had been missed.

I'm keeping track of issues on GitHub, which is also linked to in the web app.

UPDATE 25th Nov - OK, adding blank channels now adds M +/- 1 and isotopic impurities!

UPDATE 26th Nov - OK, added +16 to blank channels, plus loads of other features. Tested today on a simple Excel-uploaded 5-panel Helios run with loads of contaminant, blank channels and quick-add Cell ID channels. :D

Return to CyTOF general discussion

Who is online

Users browsing this forum: No registered users and 1 guest

cron