Hi Mario and flow-ers
I just want to mention some additional points (learned in the school of the
life) to stress Mario's comments.
It is a well know secret that many labs buy very expensive cytometers (No
name brand) to end doing single color or at best two colors and if we check
how many of them do complex sorting that require three and four colors we
almost end with a few places that have really mastered the (Dark side of
the flow) art of multicolor compensation and learned to have a wise choice
of fluorochroms couple to high affinity antibodies.
It is a well know experience, the imagen of the newly arrive postdoc and
PhDs who want sort the 0.001% population aiming to get its results
published in journals with more than 20 of index impact. At the end you and
your associates get the blame for no being able to get the same results
that were just published.
The learning curve for multicolor flow procedures is quite long. Require
strong commitment from the flow unit and from the users.
I still believe that multicolor have a future but the results for most of
the labs are so descouraging that only a few flow facilities make more than
three colors in a routine basis. The expensive machines sleep most of the
time with only one laser being used. A few guys take the time to do
appropriated compensations and put appropriated controls so far.
In the other side of the river are the companies that produce the
antibodies. I guess if those companies walk the extra mile and produce a
broader variety of fluorochroms but cheaper, the users will be willing to
take the pains of try the multicolors. I have doubts that in home coupling
can compete with the quality controls of the companies. The companies can
argue that living in the sunny state is too expensive!
In regard with the sofware stuff (No name brand), any software will do the
job for acquisition and analysis if you have proper settings and reagents.
Even the internet free sofwares can work in this case.
Multicolor is a serious bussiness that could generate very important data,
not a show off of skills and reagents. If any lab is getting there, they
should side apart people, time, equipment and budget and get into the
multicolor mud, otherwise is not going to work. Regards Rafael
Ps. I am still in the middle of the multicolor mud
>To those considering experiments utilizing more than 4-5 colors...
>
>Let us remember that instrumentation is only one of the three major
>hurdles that must be overcome to perform these experiments
>"routinely." The other two are software and reagents. The latter is
>perhaps the most difficult hurdle--since you can now buy the
>instrument as well as the software.
>
>I won't say much on the software (everyone knows my view by now),
>other than to say that your choice is extremely limited... there's
>really only one package that can handle the complex needs of
>multicolor experiments, from off-line compensation of all the
>different colors to keeping track of the dozens (if not hundreds) of
>distinct gates applied to a single sample.
>
>Reagents, however, represents a much bigger hurdle for the average
>laboratory. Currently, only about 6 colors are commercially
>available (FITC, PE, Cy5PE/PerCP/Cy5.5PerCP, Cy7PE, APC, and Cy7APC).
>For several of those colors, the availability of reagents is
>extremely low. And as you quickly discover, you simply don't want to
>do any second-step stainings if possible (there's your answer,
>Maciej), so having primary conjugates is extremely valuable.
>
>This means that your laboratory will need to learn how to conjugate
>reagents. Of course, there's nothing complex about this--it's pretty
>routine, and the steps are all layed out on my web site. However, it
>means investing in buying (or making) bulk quantities of purified
>antibodies, plus the fluorophores. And if you want to make tandems,
>then it's another level of complexity and reagent purchases.
>
>The major investment in reagents is not dollars (because, in the end,
>if you make them yourself you'll save a lot of money), but time.
>Someone has to learn how to do the conjugations, how to optimize them
>for your conditions, how to properly validate them, and how to
>database & track them (we have over 600 different conjugates in our
>refrigerator).
>
>So don't let an instrument manufacturer sales rep convince you that
>you can start doing 8-color experiments by buying their instrument.
>Before you decide to tackle such a project, realize that you will
>have to commit serious time to reagent manufacture & development.
>Even if you stick to commercially-available reagents, it will take
>some time to develop an understanding of the additional problems
>encountered by these experiments. Nicole Baumgarth and I recently
>published a paper in J. Immunological Methods (243:77-97, 2000),
>which covers some of these problems and presents quite a bit of
>useful information on setting up multicolor experiments. (I can
>EMail the PDF file by request).
>
>Finally, of course, you'll need to learn how to use FlowJo.
>
>mr
\|/
(o o)
________________________________oOo__(_)__oOo_________________________________
___/\_ | Rafael Nunez mailto:rafaeln@vetvir.unizh.ch
/ o \/| | University Inst.for Virology http://www.vetvir.unizh.ch/
/ _| | Winterthurerstr. 266a Telephone: (+41) 1 6358710
/_/\__/-\/ | 8057 Zurich SWITZERLAND Faximile : (+41) 1 6358911
______________________________________________________________________________
This archive was generated by hypermail 2b29 : Sat Mar 10 2001 - 19:31:35 EST