Scheduling
Considerations
1. The size of the cells or particles in
your sample. The majority of cells Investigators
usually sort are between 10 and 30 micrometers (um or microns u) in diameter
when on ice. Cell sizes range from <4 um to >120 um depending on species,
type of cell, temperature and phase of cell cycle and very few are spherical in
shape. For example:
·
Human granule, Red Blood and sperm
cells are 4 -8 um on ice.
·
Hematopoiesis lineage cells including T
and B lymphocytes as well as fibroblasts and myoblasts are between 10 and 20 um
on ice and 20 to 50 um at 37 C.
·
The Soma (or head) of neurons can range
from 4 to >100 um.
The
size of the cells in the sample along with other sorting preferences will
influence the sorting parameters you chose such as nozzle size, sample flow
rate and precision settings.
2. The number of cells in your sample(s).
If you are planning to sort from a sample with a very large number of cells
(>40 million), time and cost of sorting may be a very important factor and
the choice of nozzle size, sample flow rate, and precision settings will be highly important trade-offs.
3. The percentage of target cells in the
sample. If you are trying to sort a very small sub population (less
than 2%), you may well prefer to sort once with low purity and high yield
precision settings and then re-sort the collected fraction with higher purity
and lower yield settings. You may decide
to use a different technique like column separation or centrifugation rather than
FACS sorting to enrich the sample before FACS sorting.
4. The purity of the cell preparations.
The
amount of debris, especially labeled debris, and the number and amount of
contaminants in the media will directly affect the accuracy of the sort.
5. The nozzle size used for analysis or
sorting. If you are sorting from a sample with a very large number of
small (<25um) cells, and you are sure you have a clean prep with few
cells clumped, you should chose the
smaller 70 um nozzle. The 70 um nozzle produces less volume in the collection
tube which you may prefer. The 70 um
nozzle may allow for 2 to 3 times faster sample velocity if you can provide 2
-3 times higher concentration (10 million cells/ml with the 70um nozzle vs. 4
million cells/ml with the 100 um nozzle) than the 100 um nozzle. For example,
you may be able to sort 40 million cells at 10,000 events/ s in 1 hour with the
70um nozzle vs. sorting 40 million cells at 3,700 events/s in 3 hours with the
100 um nozzle.
6. Viability of the collected cells.
If
your cells are small, unlikely to clump, not fragile and you plan to sort a
large volume, the 70 um nozzle is a good choice. The pressure is higher so the
stream is faster. The sort would probably be much faster than with the 100 um
nozzle. If your cells are large and
fragile or if your cells are likely to form aggregates or clumps, the 100 um
nozzle is a better choice. The stream is
at a lower pressure and the nozzle has a much larger opening which is forgiving
with small clumps and fragile cells.
7. Instrument Settings. There
are two more instrument settings that can affect performance that are usually
set at default values for the Aria. They are Threshold (default is 5000
intensity units triggered off of Forward Scatter) and Window Extension (default
is 2 micro seconds)). The Operator/ Manager can explain these to you in more
detail, or you may choose to read the two BD Technical Bulletins included-
Conflict_Resolution.pdf
Aria_Aquisition
Window_Extension.pdf
Briefly,
the higher you set Threshold, the more noise is rejected and for forward
scatter trigger, the more debris and small particles are ignored, but you may
inadvertently exclude some of your target cells. If triggering off
fluorescence, you may exclude dim particles/cells.
The
lower you set Threshold, the more noise potentially is processed as signals.
This can result in a higher number of electronic aborts. Electronic aborts
occur whenever two signal pulses are too close together to be considered
independent events. If a target cell is
involved in an electronic abort, that cell goes to waste. Operators try to keep
electronic aborts to less than 10 percent of the sort rate and ideally close to
zero.
Window
Extension allows extra time for the electronics to integrate a signal’s area
parameter in addition to the time during which the signal’s intensity is
greater than the Threshold value. Too large a WE can lead to electronic aborts
and poor resolution of dim signals. Too small a value can lead to inconsistency
in area and other signal parameter calculations. So the bottom line here may be
that if you try to sort very fast, these two settings can come into play even
if they are otherwise correctly set.
8. Target Population Recovered.
If
your target sub-population is 2 million cells, you cannot expect 2 million
target cells and zero non-target cells in your collection tube. Due to the
random (probably Poisson) sequence of cells entering the stream from the sample
and passing the interrogation point for the lasers, you cannot know whether a
target cell in a given drop will be near the beginning, middle or end of that
drop nor whether it will be followed by a target or a non-target cell in any of
the next one to twenty or thirty drops. You must choose whether you want high
purity (precision =0, 32, 0) in which case you will get far fewer than 2
million cells including some non-target cells because of conflicts; or high
yield (precision = 32, 0, 0) in which case you will get a lot more than 2
million cells including a fairly large number of targeted as well as
non-targeted cells and far fewer conflicts. Conflicts occur whenever a target
cell has been detected but, due to the precision settings chosen, the system
decides not to sort out the target cell.
There
is a counter in the sort layout window labeled “Efficiency”. This is not an
indication of how well the sorter is doing on getting your 2 million cell sub-
population. It is defined as Efficiency = 100 (sorted events/sorted events +
sort conflicts). If you want high
purity, you might be very happy with a low efficiency number (50%) and if you
want high yield, you would expect Efficiency to be closer to but not
necessarily equal to 100 %.
If
you are collecting just one or two populations, you should always check the box
in the sort control window for “save conflicts” and set up properly to collect
conflicts. The conflicts collection tube will have many target cells that were
not collected in the target cell collection tube.
Users must also expect that sometimes the machine clogs
badly or breaks. This is bad luck for
everyone. We will do everything we can
to minimize delays and keep Users informed about any problems or delays.
