Re: Purity of sorted populations

Gerald E. Marti (gemarti@helix.nih.gov)
Wed, 29 Mar 95 14:08:52 -0500

>With regards to high speed sorting, I recently saw one or two articles by
>Carleton Steward. Perhaps he will have them available in electronic form?

I only have references to texts.
>
>Alice Longobardi Givan, Flow Cytometry: First Principles, pp. 34-40.
>This is an introductory overview of the concepts used in sorting.
>
>Howard M. Shapiro, Practical Flow Cytometry, 2nd ed., pp. 106-114.
>More detailed approach. A newer edition is available, but I don't have it.
>
>Tore Lindmo et. al., Flow Sorters for Biological Cells, in: Flow Cytometry
>and Sorting, 2nd ed., Melamed et. al. pub., pp. 145-169.
>Probably the most complete reference for sorting.
>
>All of the above references include references. (Sorry about the 'incomplete'
>references; if you have problems finding these, I can give you complete,
>Cytometry-format references).
>
>We also have a FACS Vantage. The papers that were published that included
>data based on sorted cells from our lab usually address the purity obtained.
>If I remember correctly, papers based on cells that were sorted with our
>previous cytometer, an EPICS 753, referred to "95% or better" purity. Some
>papers based on cells sorted with the Vantage refer to "98% or better" purity.
>(If the Vantage is set up correctly, I usually get close to 99%, but if one
>sort out of 10 comes back with 98%, then we have to say 98% or better for the
>complete set of 10 sorts).
>
>Now, that is sorting under optimal conditions. I also sort cells for a group
>from the University of Toledo under non-optimal conditions. First of all, the
>distinction between negatively-stained (right sort) and positively-stained
>(left sort) cells is terrible! The positively-stained cells are defined by
>the "upper shoulder" of a fluorescence histogram, having fluorescence intensity
>greater than the control histogram; i.e., there is no separation between a
>negative and a positive peak, as staining produces only one peak. The
>negatively-stained cells are defined as cells having fluorescence intensity
>less than the mode (peak channel) fluorescence. Since positively-stained cells
>rarely have fluorescence this low, the negatively-stained cells are collected
>with >98% purity. However, negative cells are unavoidably present in my
>left sort window, so I get maybe >50% or >75% purity (I can't remember the
>actual figure offhand). I also sort the cells at relatively high speeds,
>somewhere between 6000 to 8000 cells/second threshold. The higher sort rates
>produce lower purities, but really effect recovery (yield) more than purity.
>
>A big factor effecting purity on the Vantage is the drop delay offset (ddo).
>This is set on one of the electronic boards that "live" under the lasers. I
>think the manual explains how to perform a drop delay matrix, but doesn't go
>into actually setting the ddo. (You don't have to set the ddo, but it makes
>life easier than trying to remember that you must set your drop delay 1.25
>drops less than whatever number AutoSORT comes up with.) Also, there are two
>ddo settings: one for standard, rectangular sort windows; and one for SEM,
>non-rectangular sort windows.
>
>Another purity factor is sidestream formation (phase). If your sidestreams are
>not stable, i.e. the phase is incorrect, then you will not collect what you
>programmed the cytometer to collect. This is especially a problem on long,
>several hour sorts, when there is more chance that your breakoff distance
>might change (as a result of sheath pressure changes that result from the
>consumption of sheath fluid).
>
>Another observation is this: sorting non-fluorescent populations from fluor-
>escent populations (a "negative" sort) usually produces better purity. This
>is probably due to the fact that postive-negative doublets are detected as
>positively-fluorescent (and other reasons?).
>
>/\/\/\_ Eric Van Buren, vanburen%flovax.dnet@rocdec.roc.wayne.edu
>\ \ \ Immunology & Microbiology
> \_^_/ Wayne State University, Detroit, Michigan

Gerald E. Marti
Flow and Image Cytometry Section
Laboratory of Medical and Molecular Genetics
Division of Cell and Gene Therapies
CBER FDA NIH Bdg 29 Rm 502
8800 Raockville Pike
Bethesda,MD 20892


Home Page Table of Contents Sponsors Web Sites
CD ROM Vol 2 was produced by staff at the Purdue University Cytometry Laboratories and distributed free of charge as an educational service to the cytometry community. If you have any comments please direct them to Dr. J. Paul Robinson, Professor & Director, PUCL, Purdue University, West Lafayette, IN 47907. Phone:(317) 494-0757; FAX (317) 494-0517; Web http://www.cyto.purdue.edu EMAIL robinson@flowcyt.cyto.purdue.edu