Importing Experiments from Other Software¶
Importing from BD FACSDiva™¶
You can import gates, populations, worksheets, compensations and scales from FACSDiva experiments.
- In FACSDiva, right-click on your experiment in the navigator and select Export Experiments. Select an export location, select the directory export option and your experiment name, then click OK. This should create a directory with your FCS files an XML file.
- In CellEngine, create a new experiment from your inbox.
- Drag and drop your FCS files and Diva workspace XML file into your browser. Click "Yes" at the prompt asking if you want to import the Diva workspace.
The Export Experiments function in some versions of Diva exports incorrect FCS files (channels out of order). If this happens, use the Export FCS Files function to export FCS files, and only use the XML file from Export Experiments.
FACSDiva uses a proprietary implementation of biexponential scaling. When your experiment is imported, your gate coordinates will be converted from the biexponential scales used in your Diva experiment to an ArcSinh scale approximating that biexponential scale, taking into consideration the "below zero" settings for each channel. The imported gates typically have event counts within 1% of the FACSDiva experiment.
FACSDiva calculates biexponential scale values for each individual tube in your experiment. When your experiment is imported, the median of those scale values will be used. If any of your tubes have manual scale values set, then the median of the manual values will be used and the values calculated by Diva will be ignored.
Importing from Cytobank¶
Experiment information, FCS files, attachments, gates, populations, panels and scales can be imported from Cytobank experiments. Entire accounts can also be imported. Please contact firstname.lastname@example.org for more information.
Statistics between applications are typically identical or almost identical. Events at gate edges are more likely to be different between applications due to differences in numerical precision and rounding.
Cytobank does not use a population hierarchy for gating. Instead, a hierarchy must be inferred based on the gates that comprise your populations. If you have not used the population manager in Cytobank, your gating hierarchy will be step-wise and import seamlessly. However, if you have used the population manager to create, modify or delete populations, there are several conversions to be aware of:
If the same gate is used below multiple parent populations, that gate will be duplicated for each branch in which it is used. For example, an experiment with one population comprised of
Gate1, Gate3and another population comprised of
Gate2, Gate3has two possible parents for
Gate3. During import,
Gate3will be copied so that
Gate2each have unique child gates.
If more than one gate is different between a population and its next logical descendant, then populations will be automatically created to make the hierarchy step-wise. For example, an experiment with one population comprised of
Gate1and another comprised of
Gate1, Gate2, Gate3has two gates differing between the parent and child populations and is thus not a step-wise hierarchy. During import, a new population comprised of
Gate1, Gate2will be automatically created.
If a gate is not used in any populations, then a population comprised solely of that gate will be created below
Ungated. This is common when you have created readout or signaling gates (e.g. "STAT3+") that you wish to see on multiple populations. After import into CellEngine, you can copy and paste the gate below the populations where you wish to see it displayed.
Each Cytobank gate can have its own scales. During import, if a gate has scales that are different from the current Cytobank experiment scales, the gate will be transformed to the experiment scales. For more information on this transformation, see changing scales after gating.
Cytobank gates that are tailored per population will be imported as separate populations.
- Each Cytobank gate can have its own compensation. CellEngine does not support using different compensations for each gate. Cytobank experiments using multiple compensations are more likely to have differing statistics between applications.