SCYM Data Analysis 5 — Questions and Answers
Question 1: In a proliferation assay using CFSE dilution, how should the data be analyzed to calculate division index?
- Measure MFI of the CFSE peak and compare to an unstained control
- Use peak-fitting software to enumerate cells in each division generation and calculate the mean divisions per original cell (Correct answer)
- Set a single positive/negative gate on CFSE and report the percent positive
- Calculate the ratio of FSC-A to CFSE fluorescence intensity
Correct answer: Use peak-fitting software to enumerate cells in each division generation and calculate the mean divisions per original cell
Division index analysis requires fitting individual generation peaks and calculating the mean number of cell divisions undergone by all cells in the original population.
Question 2: Which quality control metric best evaluates the consistency of sample acquisition across multiple tubes in a batch run?
- Percent positive cells in the first tube as a reference standard
- Coefficient of variation of bead MFI values across all acquisition tubes (Correct answer)
- Total event count per tube normalized to acquisition time
- FSC-A pulse height range observed in each sample
Correct answer: Coefficient of variation of bead MFI values across all acquisition tubes
Tracking bead MFI CV across tubes monitors instrument stability, laser fluctuation, and alignment consistency throughout a multi-tube acquisition session.
Question 3: When using viSNE (Cytobank implementation of t-SNE), which parameter controls the size of the neighborhood considered during embedding?
- Iterations
- Perplexity (Correct answer)
- Theta
- Learning rate
Correct answer: Perplexity
Perplexity in t-SNE/viSNE controls the effective number of neighbors considered, balancing attention between local and global data structure in the resulting 2D map.
Question 4: In phospho-flow cytometry data analysis, why is it important to use unstimulated cells as a reference rather than an FMO control?
- FMO controls cannot be prepared for intracellular phospho-epitopes due to fixation requirements
- The biological baseline of phosphorylation (unstimulated state) defines the true negative population for stimulation comparisons (Correct answer)
- Unstimulated cells have lower autofluorescence than FMO controls
- FMO controls overestimate compensation in fixed and permeabilized samples
Correct answer: The biological baseline of phosphorylation (unstimulated state) defines the true negative population for stimulation comparisons
In phospho-flow, the relevant comparison is between stimulated and unstimulated cells; the unstimulated sample defines the basal phosphorylation level that serves as the functional negative reference.
Question 5: What is the purpose of 'back-gating' in hierarchical flow cytometry analysis?
- Applying a gate defined in a later analysis step to an earlier scatter plot to verify gate placement (Correct answer)
- Re-analyzing previously gated events with updated compensation values
- Reversing the order of sequential gates to simplify the gating hierarchy
- Using fluorescence parameters to redefine FSC/SSC scatter gates automatically
Correct answer: Applying a gate defined in a later analysis step to an earlier scatter plot to verify gate placement
Back-gating overlays a population identified in a later gating step onto an earlier scatter plot (e.g., FSC vs. SSC), confirming that the population is appropriately positioned and not contaminated by debris.
Question 6: In high-dimensional cytometry studies, what does the Earth Mover's Distance (EMD) metric quantify?
- The total number of events required to achieve statistical significance between two populations
- The minimum work required to transform one distribution into another, reflecting differences between samples (Correct answer)
- The degree of spectral overlap between two fluorochromes in a panel
- The number of cell clusters that differ between two experimental conditions
Correct answer: The minimum work required to transform one distribution into another, reflecting differences between samples
EMD (also called Wasserstein distance) measures the effort to morph one probability distribution into another, making it useful for comparing cytometry samples in high-dimensional space.
Question 7: Which approach is recommended to validate findings from an unsupervised clustering analysis of cytometry data?
- Re-run the same algorithm multiple times and average the cluster assignments
- Confirm cluster identities by manual back-gating and validate with orthogonal methods or biological markers (Correct answer)
- Increase the number of clusters until all events are assigned to unique singletons
- Apply the same cluster labels to all future experiments without additional validation
Correct answer: Confirm cluster identities by manual back-gating and validate with orthogonal methods or biological markers
Validation requires manual inspection of marker expression within each cluster via back-gating, cross-referencing with established immunophenotyping knowledge, and ideally confirming with an independent assay.
SCYM Data Analysis 5
In a proliferation assay using CFSE dilution, how should the data be analyzed to calculate division index?