FAQs

• Resolve applies an adaptive unmixing algorithm that performs event-specific calculations to adjust for autofluorescence and cellular characteristics, and does not produce negative values in its output.
• To support 1-click analysis, Ozette has migrated to an updated version of the algorithm that produces 0-inflated output. For a given fluor/marker combination, you can interpret this output as:
  • Exactly 0: the event is negative for the marker/fluor pair in question.
  • Above 0: the event is either expressing background signal or true positive signal for the marker/fluor pair in question.
• The Resolve Analysis Platform generates a novel positivity score for each fluor on each event. This is a quantitative metric that can identify signal above background from noise, autofluorescence, and unmixing artifacts.

Only for existing customers. Ozette will continue to support the previous version of the Resolve unmixing algorithm for all customers using it for their current analyses. New customers will only be able to use the version that supports 1-click analysis.

No, FMOs remain the gold standard for interpretation of outputs. You may upload and include FMOs in your Resolve analysis platform. When data requirements are not met, FMOs are very useful for tuning the positivity threshold generated in the Resolve system. However, if you are not able to regularly generate FMO/FMX controls, the positivity score provides an objective measure for identifying cellular phenotypes in your cytometry data.

No, not necessarily. The performance of the 1-click analysis largely depends on the suitability of the provided unstained controls. If the unstained controls are sub-optimal, the spectra measured in single-color controls can still be used to generate high-quality unmixed outputs that require some adjusting of the default positivity thresholds in the 1-click analysis.

No. Compensation is not supported in the Resolve Analysis Platform, because negative unmixed values cannot be generated by the mathematical constraints of the unmixing model implemented in Resolve.

On a given event, the unmixed value assigned to a marker/fluor is a weight telling you how much of that fluor’s spectral signature is needed to recover the observed measurement on the cytometer.

Yes, and the manifestation of that error is different in Resolve from output produced by most cytometers. Spillover and spreading error depends on the panel design, and manifests in Resolve as a positive unmixed value assigned to an event for a marker/fluor pair that is caused by signals from other fluor/marker pairs “spreading” into the affected marker/fluor pair. The Resolve Positivity Score provides a quantitative diagnostic to identify and avoid calling such errors as positive unmixed values for the fluor/marker pair in question.

The mathematical model implemented in the Resolve Analysis Platform prevents these outputs from being generated. This allows you to interpret the unmixed outputs of the Resolve Analysis Platform as the amount of a fluor’s spectra needed to recover a measured raw signal. Any system that produces negative unmixed output prevents this interpretation, and indeed, produces unmixed values for events that do not admit physical interpretation.

Yes. From the point of view of the Resolve Analysis Platform, the only difference between a conventional and spectral cytometer is the number of detectors. The same unmixing algorithm is applied to all files processed by the Resolve Analysis Platform.

Yes. The hierarchy estimated by the Resolve Analysis Platform is merely a good starting point. You are able to edit and create the best strategy for your panel & project, and re-generate 1-click outputs in the system.

No. While the cellular phenotype hierarchy estimated by the Resolve Analysis Platform looks identical to a gating strategy (and indeed can be used this way), the output is a “filtering tree”: each node has logical filters applied to identify a given subset. The platform is able to achieve this filtering tree because the Resolve Positivity Score is event and fluor specific, and so accounts for event-specific variation in brightness and cell size.

No. The Resolve Analysis Platform cannot correct experimental errors. It uses material-specific unstained controls and single-color controls to unmix the data, but the quality of that data is still bounded by your controls and panel design.

Yes, the Resolve Analysis Platform corrects for cellular autofluorescence for all files. This is why you must provide unstained controls for each material and condition you unmix. If your material type has unique autofluorescence characteristics, you can dedicate a detector to help Resolve model and correct for it.

Yes, either works. We require running at least beads at saturating titers, and ideally across any material type you intend to assay, so you can compare spectral shape across material types in Resolve.

Spectral selection and filtering: identifying true positives in your control set. Re-generate controls for any fluor in which you are unable to confidently identify true positive events in the Resolve system.

No. With each new control set, you can inherit gating from a previous set and review/edit from there instead of starting from scratch. Inheriting from a previous control set will be available as an option when you “proceed to gating” with your new control set.

• Email support@ozette.com with any questions.
• Expert review and recommendations for your control sets, available on request.