FAQ
How do I arrange to run samples at the Flow Cytometry Core Facility?

In order to be able to run samples on the instruments of the Flow Cytometry Facility you will need to attend to a training course specific for the flow cytometer you are going to use. These courses are obligatory for everybody who plans to use the facility resources.
Those courses are free of charge but the instrument used for the training itself will be charged. To register for a course please complete the registration form available on the download section of this website and submit it to
After attending the course you will be assigned a login and a password for the online booking system permitting you to book the machines that your training was for. In order to schedule a cytometer training session or to obtain access to the online booking system please complete and submit the Request Form available from the download section of the website. If you only intend to use the Flow Cytometry Facility resources very occasionally, please ask someone from your lab who has access to the machines to run your samples for you or ask the Flow Cytometry Facility staff to assist you.

How much will I be charged for using the facility resorces?

Invoices are sent every 3 months with a minimum charge of 100 CHF per trimester to cover administration costs. This money contributes to the running costs of all the machines.

Invoices ar send by the accounting department of the University of Lausanne. For any inquiries concerning the invoices please contact

pricelist
Will I be charged for the entire time I sign up for?

No. You are not charged if you finish early, as long as you modify your booking accordingly within 24hours of when the booked session ended.

However we ask that you do not overbook your time deliberately as this prevents others from scheduling during that time.
Please ensure that the time you book is sufficient for running your samples, cleaning the machine and exporting the data.

 

If you are unsure of how much time you will need contact the facility staff for advice.

Can I analyze infected/transfected cells?

All samples classified as BSL-1 or below (e.g. many primary, non-genetically modified cells of veterinary origin) can be analyzed on any of the instruments without fixation.

All primary human cells and some cell lines are considered to be at least BSL-2. Samples above BSL-1 maybe analyzed after they have been fixed with a proven pathogen deactivating substance (e.g 1-2% Paraformaldehyde).

 

For more details, please refer to the Biosafety section of this website.

How many cells should I have in my sample and how many events shall I record?

The cell concentration within your sample should be between 1-5 million per ml with a minimum volume of about 200ul. For your first experiments you should plan a larger volume for your setup tubes, to have enough cells to set up the flow cytometer correctly.

The number of events that you need to record depends on the proportion of the population you are interested in analyzing.
Consider the general case of enumerating a total of 'N' events, of which 'R' meet a certain criterion (positives). The proportion of positives, P=R/N, will also be the probability of any particular event being observed as positive 0=P=1, which is clearly related to the random manner in which cells are selected for analysis. As with all statistical distributions, the variance, Var, is a fundamental parameter and, for the binomial, is defined as follows:


Var(R) = NP(1-P)


The more familiar standard deviation, SD, is the square root of the variance, and the coefficient of variation (CV) is the SD expressed as a percentage of the population:


CV = (v(Var(R))*100)/R


These simple equations can now be used to examine some practical situations.

 

Consider a preparation of peripheral blood mononuclear cells labeled with an antibody to detect B cells. Flow cytometry then indicates that 10% of the cells present are positive for this marker (i.e. P=0.1 and P(1-P)=0.09).

If three data sets were collected for 1,000, 5,000 and 10,000 events, we would expect to observe 100, 500 and 1,000 positive cells with variances of 90, 450 and 900, respectively. Expressed as SDs, these would be 9.5, 21.2 and 30 and as CVs 9.5, 4.2 and 3. Good experimental practice within the biological field usually results in CVs on of the order of 5%. Hence, the example above indicates that a file of 5,000 events will provide that level of confidence as expected from the argument relating to dogma: For a sub-population of 10%, a total of 5,000 events would be adequate. Now consider a subset of T cells representing only 1% of the total, with P=0.01, P(1-P)=0.0099. Collecting the files as above, we would expect to observe 10, 50 and 100 positive events with variances of 10, 50 and 100, and CV's of 32, 14 and 10. Clearly, 5,000 and 10,000 total events are no longer adequate. Collecting 100,000 events in this situation would result in a CV of 3.2, and an important feature becomes evident: To improve the CV by a factor of 10 (32 to 3.2), 100 times more data is required. As a direct consequence of the square root, to improve the CV by x, x2 events are required. The approximation mentioned above can be used for the rare events defined as populations of less than 5% (1 in 20), when P=0.05 and P(1-P)=0.0475. This simplifies the equation for calculating the variance to Var(R)=NP=NR/N=R (i.e. the number of rare events observed). Hence, the SD=vR and CV=vR*100/ R=100/vR. It is obvious that the reliability of the data set is now only dependent on the number of rare events observed. Consequently, for a desired level of reliability (CV), the number of rare events to be counted can be calculated from R=(100/CV)2 . To match typical experimental variations of 5%, approximately 400 rare events need to be observed, which at the defined limit of rare events (1 in 20), a list file of 8,000 would be required. A useful table can now be constructed from values of desired CV and expected size of the subpopulation, showing the total number of events that need to be collected.

 


Frequency of Rare Events (1/X)

Desired Coefficient of Variation % (Rare Events Required)

30 (11)

10 (100)

5 (400)

3 (1,111)

20

222

2,000

8,000

22,222

50

556

5,000

20,000

55,556

100

1,111

10,000

40,000

111,111

1,000

11,111

100,000

400,000

1,111,111

10,000

111,111

1,000,000

4,000,000

11,111,111

100,000

1,111,111

10,000,000

40,000,000

111,111,111

1,000,000

11,111,111

100,000,000

400,000,000

1,111,111,111

 

If background counts are appreciable, special care needs to be taken when corrections are performed. If we have an observed count R with an associated background (control) count of B, the variance will be vR and vB, respectively. However, the variance of R-B is the sum of the individual variances, not the difference.

 

var(R-B) = var(R) + var(B)

 

In situations where high background counts begin to affect the reliability of data sets, collecting more data will improve the quality. For example, consider test and background counts of 200 and 100 in list mode files of 10,000 events. The corrected count will be 100 with a variance of 300, SD=17.3, CV=17.3%. If the data sets are increased to 100,000 events, we would expect 2,000 and 1,000 events, respectively, corrected to 1,000 with a variance of 3,000, SD=54.8 and CV reduced to 5.48%. Once again the importance of collecting sufficient data is obvious.

 

The Poisson distribution becomes skewed as very low numbers are considered and, if these situations are avoided, confidence limits can be applied in the same manner as with Gaussian (continuous) distributions. Providing more than about 50 events are observed (after correcting for background counts, if necessary), the standard deviation can be calculated as above and multiplied by 1.64, 1.96 or 2.58 to yield the 90, 95 and 99% confidence limits, respectively, on either side of the determined value. This approach is valuable in clinical situations where, for example, a decision may be made on a flow cytometry-based count of stem cells as to whether sufficient cells are present to provide a successful transplant.

 

For a more detailed discussion about how many events you need to acquire, please see [1].

 

How do I prepare my samples and what do I bring them in?

There are numerous methods for preparing e.g. leukocytes from blood or whole tissues to yield single cell suspensions. A few points are considered here.
In general, cells should be maintained in a nutrient-supplemented, protein-containing medium for staining and analysis. Often simple PBS containing 3% FCS (fetal calf serum) buffered at pH 7.2 is sufficient, and doesn't result in excessive cell death. If using Ca2+ or Mg2+ dependent antibodies or fusion proteins a Ca2+/Mg2+ containing salt solution (such as modified Hanks' Balanced Salt Solution (HBSS) Invitrogen Catalog Number 14025076) might be used. For very sensitive cell types that require more complex media a modified RPMI-1640 (with L-glutamine) (Sigma, Product number R8755-10X1L) that is commercially prepared without biotin, riboflavin, phenol red, and bicarbonate might be an option.

 

Generally for all staining buffers biotin should be avoided since it will interfere with avidin second-step reagents. Riboflavin and phenol red should be eliminated because they can increase apparent cell autofluorescence. The pH is buffered at 7.2 with 10 mM HEPES, and the protein source should be heat inactivated 3% newborn or fetal calf serum or bovine serum albumin. 0.02% sodium azide in staining buffers reduces cellular metabolism and thus prevents cap formation or sloughing of antigens, however in most cases this is not an issue with mammalian cells when kept at 4°C or on ice for short time frames. The presence of sodium azide generally has no effect on viability as long as cells are maintained in the cold, but as it has toxic and environmental polluting potential it should be avoided if not necessary. Cell aggregates and clumps can be reduced by filtering cell suspensions through nylon mesh filters before staining. In addition to filtering, 1-2 mM EDTA may be added to the staining medium to reduce clumping of cells. EDTA is a polyamino carboxylic acid with chelating properties for binding metal ions such as Ca2+ and Fe3+(be aware that binding of some antibodies and fusion proteins is Ca2+ dependent). Sticky extracellular DNA (released by dead cells) can be digested with DNAse prior to staining. DNAse is particularly effective in reducing clumping when tissue preparations exhibit considerable cell death. It may also be useful to further process the cells by density gradient centrifugation with a medium such as Ficoll, which results in a preparation with higher viability and less debris by removal of dead cells [2]. For analysis your samples must be in 12x75mm round bottom Falcon tubes (BD Falcon catalog no 352008).

 

NOTE:
all samples MUST be filtered through a 40um nylon mesh or cell strainer cap (BD Falcon catalog no 352340) immediately before loading onto the cytometer. Thus bring your mesh or filter top tubes with you and filter your samples as you are about to pass them on the machine. Do not filter them in your lab and then bring them to the machine.

Do I have to fix my cells? If so, what do I fix them in?

Any samples which may contain potentially infectious material and all human derived cells which have not officially been classified to be BSL-1 or less, MUST be fixed prior to analysis. Cells should be fixed in a final concentration of 1-2% Paraformaldehyde in PBS or another proven pathogen deactivating fixation method.

 

Please also refer to the Biosafety section of this website!

What kind of controls do I need?

 

Basically the term "controls" can be subdivided into classes. In terms of flow cytometry you have two groups, instrument setup controls and experiment controls (these can then be further subdivided into isotype controls and fluorescence minus one (FMO) controls etc).

 

Instrument setup controls are defined (as the name suggests) as controls necessary for correctly setting up instrument parameters such as signal amplification voltages and compensation (correction factors to apply to overlapping emission spectra between one dye and another). Instrument Setup controls usually include an unstained sample and single fluorochrome stained samples for each fluorochrome used.
For correct instrument set up and to ensure reliable results, these controls should be used every time an experiment is performed and should undergo the same treatment as the actual sample to be analyzed (fixation, centrifugation steps, special buffers etc).

Single fluorochrome stains must have an adequate signal (at least as bright as the stained sample but not too bright to be off scale) and a high enough percentage of positives to adjust settings.
They do not have to be specific for the same marker used in the experiment, except when tandem dyes such as PE-Cy5, APC-Cy7 etc., are used for staining. There can be significant lot-to-lot differences in the donor-acceptor ratio of the tandem dye that influences the values applied for correct compensation.

 

In some cases, beads such as antibody capture beads are a good alternative to cells for the preparation of single stained fluorochrome stained controls. These are especially useful in cases where the actual number of available cells is low, the positive population within the single stained control is too small or too dim, or shows a big coefficient of variation within the fluorescence parameter, to compute a correct compensation matrix.

 

Experiment controls include isotype matched, fluorescence minus one (FMO), wild type and other biological or non biological controls, which might be necessary to eliminate alternate explanations of experimental results. Not all of these are obligatory for every kind of experiment but some results might be uninterpretable without the use of these.
The use of isotype controls has recently been discussed by Keeney et al., who showed an example of where isotype controls do not work, and can in fact lead to erroneous estimations of the target subpopulation [3][4]. However, there are still applications where the use of isotype controls is appropriate.

Isotype controls must be of the same subclass, species and fluorochrome, as the antibodies used in the experiment. Isotype controls should also be used at the same concentration and fluorochrome:protein ratio as the specific antibody, and be purchased from the same vendor. Isotype controls are only useful as a gross estimate of non-specific binding of your specific antibody and should not be used as an indicator of where to set positive markers. There are many factors which affect the amount of nonspecific binding of an antibody, some of which an isotype control may or may not elucidate.

FMO controls include all positive staining fluorochromes minus one [5][6]. Thus an assay employing 5 fluorescent dyes needs five different FMO controls. Aside from undesirable antibody binding due to nonspecific binding, the intensity of the negative population in FMO controls is determined by compensation (or lack thereof) for spectral overlap. Background, as a result of compensation for spectral overlap is proportional to the standard deviation and mean fluorescence intensity of a positive population and the relative amount of bleed-through into the detector in which the fluorochrome of interest is measured [5][7]. As a result, the coefficient of variation (CV) of a negative population may appear larger in compensated data than in uncompensated data. Since FMO controls are labeled with all fluorochromes involved except one, they show (unlike the single stained positive controls) the same apparent increase in CV of the negative population as the experimental sample. Furthermore, FMO controls help to determine positivity and to set regions in samples that contain multi-labeled subpopulations [8][9][10][11][12].

How do I reduce non specific binding?

Non-specific binding can be due to several factors.

    1. Too much antibody can increase the amount of non-specific binding of your negative population, thus reducing your signal:noise ratio. If the antibody you are using has not been titrated, than a titration should be done to determine the optimal concentration [2].
      Titrating reagents is relatively simple. Usually, a 2-fold serial dilution series is generated, starting at what is presumed to be a saturating concentration of reagent. Stains are adjusted to have equivalent final volumes. A second "color" (spectrally not interfering with the reagent you are titrating for) which will positively identify the cells which also stain for the titrated reagent maybe be included (for instance, when titrating CD3, the counter-stain could be anti-TCR). This is so that the median fluorescence of the positive and negative populations can be accurately determined even if the antibody stain is so dilute as to generate overlapping distributions for the positive and negative populations. The second color should be added after generating the dilution series. Finally, add the number of cells to the stain that will typically be used under experimental staining conditions (such as 1 million in 50ul volume). All incubations and washes should be performed under standard staining conditions [2]. To choose the right antibody concentration out of a titration series the stain index might be chosen as a arithmetical proof. It represents the effective brightness of a reagent, which is expressed as a linear function of the difference (D) between the positive (red) and the negative (blue, green) populations and the spread of the negative population (W) which is equal to 2s.d.

       

      Note: For most antibodies, if properly titrated and used at or near saturation, cell number is essentially irrelevant (at least, up to 10's of millions of cells). This is because the saturating concentration of an antibody is such that there is typically a vast excess of antibody over antigen, making the binding kinetics insensitive to cell number [2].
    SI


    1. Non-specific binding can also be due to FcR-mediated binding. Using monoclonal antibodies, such as the supernatant of the hybridoma cell line 2.4G2 (ATCC Number: HB-197) specific for the FcγR, a receptor expressed by macrophages and Fc?R bearing lymphoid cells can help reduce background binding. Alternatively you can use IgG (e.g. www.Lampire.com) of the same species as your antibody to block "nonspecific" binding. Incubate samples with a final concentration of 1-2ug/ml of IgG for 5-10 minutes at room temperature before adding specific antibodies.

       

      NOTE: If you use IgG or FcR antibody to block FcR it is not possible to use indirect staining with a polyclonal anti-mouse (or Rat) IgG-coupled antibody as this will bind directly to the blocking agent as well as the antibody it was intended for.

    1. The use of directly conjugated antibodies can also reduce the amount of non-specific binding.
      If your antibody is not commercially available as a directly conjugated antibody, there are a number of simple procedures available with which you can easily conjugate your antibody. Invitrogen (http://www.invitrogen.com) offers conjugation kits for their Alexa dyes and Pacific Blue which are very simple to perform. In addition, they also offer the Zenon labeling kits that utilize fluorochrome conjugated Fab antibody fragments directed against the specific isotype of your antibody. A simple, 10 minute incubation is all that is required. Other options include conjugation kits for PE and APC are available from Prozyme, Inc. (www.prozyme.com).

    1. The use of a biotinylated antibody with a streptavidin fluorochrome conjugate can be a source of nonspecific binding in some cells. Biotin is a component of normal cellular metabolism, and as such, there will be truckloads of it within some mammalian cells. However, indirect staining e.g. with biotin:streptavidin conjugates can also amplify weak signals and thus enhance staining of some weakly expressed antigens.

    1. Dead cells are notorious for non-specifically binding antibodies. Inclusion of a viability dye (i.e. PI, 7-AAD, DAPI) in your assay will allow you to exclude the dead cell population from your analysis. However, you should ensure that your viability dye is compatible with the other fluorochromes in your sample.

 

How often are data files removed from the computers?

In order to maintain satisfactory computer performance, data must be deleted from the database of the cytometer workstations. This is done infrequently, and without notification when the database exceeds 15GB. Employees of the Ludwig Institute for Cancer Research are requested to export their data to the local backup folder after they have finished their acquisition (and not directly to the public server), as this folder is backed up daily to a remote server. Other users should export and copy their data to external storage media after having finished their acquisition to prevent data loss.

How do I analyze data acquired on the cytometers?

There are several tools available for analyzing data. Some are available in-house, some are provided as free software solutions and some are commercially distributed.

 

FlowJo is a software package for analyzing flow cytometry data developed by Tree Star, Inc. It became a commercially available product in 1996 for Macintosh Operating Systems. The first Windows version was released in 2002. In FlowJo, samples are organized in a "Workspace" window, which presents a hierarchical view of all the samples and their analyses (gates and statistics). Viewing an entire experiment in a Workspace permits organizing and managing complex cytometry experiments and produces detailed graphical reports. FlowJo's ability to automate repetitive operations facilitates the production of statistics tables and graphical reports when the experiment involves many samples, parameters and/or operations.
Within a workspace, samples can be grouped or sorted by various attributes such as the panel of antibodies with which they are stained, tissue type, or patient from whom they came. When an operation on a group is initiated, FlowJo can perform the same operation on every sample belonging to that group. Thus, you can apply a gate to a sample, copy it to the group, and that gate will be automatically placed on all samples in the group.
FlowJo provides tools for the creation of:

  • Histogram and other plot overlays
  • Cell cycle analysis
  • Calcium flux analysis
  • Proliferation analysis
  • Quantitation
  • Cluster identification
  • backgating and bi-exponential transformation

Employees of the Ludwig Institute for Cancer Research can request access to their FlowJo site license by contacting Dr. Anne Wilson.
Many labs in the Department of Biochemistry using flow cytometry will have some common use dongles available.

 

BD FACSDiVa software is a data acquisition and analysis package running on Microsoft Windows XP (Service Pack 2 or higher) based operating systems. The software provides analysis features including one-click Snap-To gating tools, hierarchical gating, and biexponential display. To simplify experiment and data management, BD FACSDiVa software uses a database based browser view that allows users to easily organize experiments, group specimens and tubes and design global or tube-specific analyses layouts. The browser also allows you to manage and process recorded data in the context of a single tube as well as an entire experiment.
DiVa is installed on the computer workstations of the FACSCanto, LSR-II and SORP LSR-II but may only be used for data analysis when the computers are not required to run the cytometers for sample acquisition purposes.

 

BD CellQuest Pro software allows you to analyze data stored in FCS 2.0 file format with a maximum of 8 parameters, FCS2.0 files from some cytometers (e.g. file exported as FCS 2.0 from BDFACS DiVa software) need to be converted to the right byte-order using the BDFACS Convert tool.
Within the BD CellQuest Pro experiment document window, you can create several types of plots including multicolor contour plots and overlaid histograms; and you can generate statistics for dot plots, histograms, density plots, 3D plots, and contour plots. You can also save everything in an Experiment document, including plots, regions, gates, markers, statistics, calculated expressions, Browser contents, annotations, and color preferences, and then restore it all later.
Workstations with this software are available on the FACScans, the FACSCalibur but may only be used for data analysis when these computers are not required to run the cytometers. An additional CellQuest installation for data analysis can be found in Room C316.

 

FCS Express is a fully featured analysis package provided by De Novo Software. The graphic user interface of FCS Express resembles Office applications such as Microsoft Power Point and therefore offers an intuitive and easy to use design. A useful and rare feature is the ability to import BD FACS DiVa experiment files, which include the gating strategy used for acquisition.
FCS Express runs on all common Microsoft Windows operating systems (NT/2000/2003/XP/Vista/7).

 

Kaluza is a relatively new and promising analysis tool for Flow Cytometry data that was launched by Beckman Coulter in late 2009. This software includes remarkably advanced features, while maintaining an intuitive, user-friendly interface. Because of the simple nature of Kaluza, you'll spend less time searching for options and more time analyzing data. It offers several new data visualization possibilities such as the tree plot, which provides a useful data comparison tool, as one tree plot can condense data from up to 28 bivariate plots or the Radar Plot, which maps multi-dimensional data onto a two-dimensional surface. Kaluza runs on Microsoft® Windows XP 32 bit Operating System with Service Pack 3, or Windows Vista® 32 bit Operating System with Service Pack 2. A free Kaluza Trial version can be requested at http://www.coulterflow.com/bciflow/kaluza.php (and is definitely worth a try).

 

Cyflogic as a useful freeware tool (http://www.cyflogic.com/) can also be considered for analyzing Flow Cytometry data. It is designed to run in a Microsoft Windows XP environment and is able to read FCS1.0, FCS2.0 and FCS3.0 files (see below). It offers all "regular" analysis capabilities, such as dot plots, histograms and statistics. In addition, Cyflogic has some new and innovative tools for data analysis. Cyflogic is free for non-commercial academic use. All normal analysis capabilities exist in the free version.

What are the differences between FCS1.0, FCS2.0 and FCS3.0 files?

Briefly, a Flow Cytometry Standard (FCS) file is a particular way that information obtained from a flow cytometer is encoded for storage in a computer file. The first data file standard (FCS 1.0), was introduced in 1984 by Murphy and Chused [13]. This provided a standard "data container" for transporting information obtained from a flow cytometer in a dynamic format. As information technology and flow cytometry developed synchronously to collect and process an increasing number of parameters and data with increased resolution, the FCS1.0 file standard was revised and replaced by the FCS2.0 standard in 1989. This was then replaced by FCS3.0 in 1997[14].
Basically all three standards provide a detailed description of a data file structure designed such that the file can include all of the information necessary to describe the following parameters fully:

  1. the instrument used to obtain the data;
  2. the sample measured;
  3. the data obtained; and
  4. the results of data analysis [14].

Recently, Spindlen et al published a paper about a FCS 3.1 file standard [15], which represents a minor revision of FCS 3.0. Only a few cytometers (and none on site) still employ the FCS 1.0 file standard.

 

A FCS file is by definition composed of four segments, termed HEADER, TEXT, DATA, and ANALYSIS (optional). Information in HEADER, TEXT and ANALYSIS is coded in ASCII (American Standard Code for Information Interchange) for FCS1.0 and FCS2.0 and can be ASCII or Unicode in FCS3.0. Unicode allows display of international characters that are not supported by ASCII. Information in the DATA area is (in the majority of cases) simply a binary coded list (list mode) of the data values, the length of which in bits is defined by the keyword $PnB in the TEXT area where 'n' is the parameter number. These values are mostly stored binary coded, usually with 16 bit resolution (FCS1.0 & FCS2.0), whereas only 10 bits are actually used by most cytometers from that time (e.g. BD FACScan or BD FACSCalibur). While a single bit can be either a value of 1 or 0, this gives 1024 code possibilities by coding data in 10 bits, or an effectively usable scale for "sorting" data into 1024 bins (Channel 0 – 1023) for each stored parameter.
In FCS3.0 files, the DATA segment is usually coded by 32 bits. This provides a possible resolution, and therefore theoretically usable scale of max. ~ 4.2 x 10E9 channel values. However until now (05/2010), no analytical cytometer currently on the market makes use of the entire scale. The analytical cytometer currently offering the highest resolution is the Accuri C6 which supports a dynamic range of 24 bits (16,777,216 Channels) followed by the Beckman Coulter Gallios, with 20 bits (1,048,576 Channels). Becton Dickinson's state of the art cytometers (LSR-II, FACSCanto, Fortessa) offer a dynamic range of 16 bits (65,536 Channels) for signal height measurements and 18 bits (262,144 Channels) for signal area measurements.

Most state of the art analysis tools are compatible with all current flow cytometry standard file formats; however some earlier programs, such as CellQuest or WinMDI, which are still used by some users do not offer full FCS3.0 compatibility and, in the case of CellQuest, only a limited number of 8 parameters in FSC2.0 file format. BD FACSDiVa software allows data export in both FCS2.0 and FCS3.0 formats, and thus either can be used for analysis. However, some features such as bi-exponential transformation [16] [17] or re-compensation of acquired data are not supported at all, or only supported to a limited extent in FCS2.0 files. Compensation may be increased (e.g. if undercompensated) in a FCS2.0 file that has been recorded with an "earlier" flow cytometer (e.g. FACSCalibur) or exported from DiVa as FCS2.0. A reduction in compensation (e.g. if data is overcompensated) in FCS2.0 files exported from DiVa or acquired on "earlier" cytometers, is usually not possible, as once a data point hits the edge of the plot, it registers as zero in DiVa (FCS2.0) and its true position can't be re-calculated precisely. Many programs do not support writing negative values, as earlier cytometers cannot process negative events (events inducing a voltage to the detector below the baseline voltage was processed as 0) or uncompensated data in addition to a compensation matrix for offline compensation to FCS2.0, even if that's actually not a limit of the file format itself.

These, and the fact that FCS2.0 files cannot exceed 100Mb (99,999,999 bytes = 95,37 Mb), represent the most limiting point of this format. While this 100Mb border was not of an issue in the past, it has become more so with further instrument and computer developments.

Ch. des Boveresses 155 - CH-1066 Epalinges  - Switzerland  -  Phone +41 21 692 59 67  -  Fax  +41 21 692 59 95