[Cytometry] FW: Teaching Flow Cytometry - Yes, Guilty! but...

Leonore A. Herzenberg leeherz at darwin.stanford.edu
Sat May 7 05:03:10 EDT 2011

Having had much to do with building flow the initial cytometers, creating monoclonal antibody reagents, creating new flow applications and broadly disseminating the whole technology, Len (Leonard A. Herzenberg) and I plead guilty -- guilty to having visited upon biomedical scientists and related colleagues a need to learn about technical details that truly have nothing at all to do with their primary research and clinical interests.   Let's face it.  Most people who use flow really do not need to know about overlapping spectra, fluorescence compensation, etc.  They just need the data from their experiments.   Nevertheless, it is certainly true that they cannot adequately plan their flow experiments or interpret their flow data without understanding how to take these intricacies of flow studies into account.  

Recognizing this problem, some years ago Len, Dave Parks, Wayne Moore, Stephen Meehan and I set out to try to build software that would support optimal flow use without forcing people to know how to juggle colors to get the best stain sets for the target cells being studied, how to determine which markers in the stain set require FMO controls, or how to be sure the instrument used to collect the data can support data collection for the chosen fluorochromes.  Our object, quite simply, was to free scientists to focus on their science rather than tie them down with details that are highly relevant to getting good data but basically irrelevant to the work they are trying to do.  

This is still a work in progress.  However, at the last ISAC meeting, Meehan reported that he and his engineering group at Woodside Logic, working closely with Parks, Moore and the biologists in our laborfatory, had completed an initial "CytoGenie AutoColor" version that could compute and optimize potential fluorochrome combinations for stain sets.  This year, Stephen and Woodside Logic will return with a much broader and more complete version.  Among much else, this version takes marker expression levels on target subsets and key staining protocol dictates (methanol fixation, second step reagents, etc.) into account, and generates an ordered list of reagent combinations optimized to work with the user's selected instrument. Meehan's group has also embedded this stain set design technology in an overall protocol design system that does several other of the knowledge-requiring protocol design steps, further removing the need for so much specialized flow knowledge.

FlowJo, Coulter and several other flow cytometry companies have also recognized that flow users need help of this type.  They have responded by putting stain set design tools up on their respective websites.  Meehan's software, which is available through Woodside Logic, offers substantially more robust stain set design capability, plus help with other protocol design tasks.  However, all of the stain set design tools that have been posted provide basic help that is clearly a step in the right direction.  Users, not just novices, should be encouraged to use these tools since they speed up protocol design and provide help for building more optimal protocols.

Does the upsurge of these protocol design aids mean that Howard is wrong in urging that scientists think more and use good reasoning wherever they can?  Of course not.  But the more these tools improve the ability of scientists to their work, and frees their time for thinking about their science, rather than their technology, the better.  It is up to us, the leaders and teachers in flow cytometry, to understand these needs and put out efforts into building and/or demanding  tools that use of modern information technology to make flow work easier and better.  

Howard's Practical Flow Cytometry is an invaluable tool for our education and our reference, and he does well to remind us that thinking and understanding is at the heart of our endeavor.  But we should perhaps always be willing to recognize that not everyone who uses flow needs to be an expert, and we should offer these others the benefit of our expertise, either directly or through the writing and tool building we do.  As Howard says at the end of his letter, the goal is to improve the delivery of information about flow technology so that fewer people have to make mistakes before information is made available to enable others to bypass those mistakes.  Building knowledge-based software to support various flow tasks can help to do this.

Thanks for listening.  In the interests of good science and communication, I think you should know that Len and I founded Woodside Logic together with Stephen Meehan to attack the problems set forth above and to facilitate other aspects of flow cytometry usage.  To date, Len and I are Woodside Logic's principal source of funding.   

Lee Herzenberg

-----Original Message-----
From: cytometry-bounces at lists.purdue.edu [mailto:cytometry-bounces at lists.purdue.edu] On Behalf Of Howard Shapiro
Sent: Friday, May 06, 2011 3:42 PM
To: flow list Purdue flow list Purdue (cytometry at lists.purdue.edu)
Subject: Re: [Cytometry] Teaching Flow Cytometry - Not Guilty!

 Ann Atzberger and Dale Hirschkorn seemed to think I was putting Till Dettmering down in my reply to his original post; nothing of the sort. Till established himself as one of the good guys; he was asking about others' experience and, in mine, one of the big problems in the field overall is now that there seem to be fewer and fewer people as well motivated as Till and the new students he is mentoring.
I suspect cytometry is not the only field in which this is happening, but it is the field in which I am most interested in reversing the trend.

As I have pointed out in previous postings, there are around 4,000 people reading this Mailing List, and recent editions of my flow book have sold about 5,000 copies each, although there are presumably a fair number of downloads of the 4th Edition. There are probably at least 35,000 flow cytometers out there; nothing like that number of people are likely to have taken intensive training courses in their use. Many of the investigators who have brought in the big money that paid for their fancy apparatus take it for granted that the people they hired to run it are experts, but they can't really tell. 

I have no idea how to solve that big problem. Right now, though, we also need to help people who are motivated get the information they need to make optimal use of and take optimal care of the instruments they have. The manufacturers could certainly be helpful with this, even if they don't want to invest more time and money preparing educational materials. They know which institutions have which machines and, in many if not most cases, who is using the apparatus and what types of analyses are being done. As I have sung, "there's no people like flow people." By and large, users are nice, and like to pass on helpful information they have acquired. There's a lot of it in the Mailing List archives; it could be made more accessible and more useful if small groups of users collected and indexed what was relevant to different instrument types.

Methodology, i.e., how to prepare, stain, and analyze cells, usually finds its way into the literature; how to do things on specific machines does not, and that's the information that needs to be made more accessible to get new users up to speed faster. It would also be helpful to get people working on less common applications of the technology together so only a few have to make each of the mistakes that can and will be made along the way.


Cytometry mailing list
Cytometry at lists.purdue.edu
Search the list archive at  http://tinyurl.com/cytometry

More information about the Cytometry mailing list