SCYM Data Analysis 3 — Questions and Answers
Question 1: In dimensionality reduction analysis of high-dimensional cytometry data, what does t-SNE primarily preserve?
- Global structure and overall distances between all clusters
- Local neighborhood relationships between similar cells (Correct answer)
- Linear relationships between all measured parameters
- Absolute fluorescence intensity values for each marker
Correct answer: Local neighborhood relationships between similar cells
t-SNE (t-distributed Stochastic Neighbor Embedding) preserves local neighborhood structure, placing similar cells near each other while global distances between clusters are not preserved.
Question 2: During cytometry data analysis, 'doublet discrimination' is performed primarily using which parameter combination?
- FL1-A vs FL2-A scatter plot
- FSC-A vs FSC-H or SSC-A vs SSC-H pulse geometry plots (Correct answer)
- CD45 vs CD3 bivariate dot plot
- Viability dye vs autofluorescence histogram
Correct answer: FSC-A vs FSC-H or SSC-A vs SSC-H pulse geometry plots
Doublet discrimination uses pulse area (A) vs. pulse height (H) plots because doublets have proportionally higher area than height compared to singlets of the same size.
Question 3: Which UMAP parameter most controls the balance between preserving local versus global data structure?
- n_neighbors (Correct answer)
- min_dist
- metric
- n_components
Correct answer: n_neighbors
The n_neighbors parameter in UMAP determines how many neighboring points are considered during embedding, with higher values preserving more global structure and lower values emphasizing local clusters.
Question 4: When performing cluster analysis of cytometry data using FlowSOM, what does the meta-clustering step accomplish?
- Assigns fluorescence intensities to each node in the self-organizing map
- Groups similar SOM nodes into biologically meaningful metaclusters (Correct answer)
- Normalizes data across all samples before clustering begins
- Applies compensation matrices to reduce spillover between markers
Correct answer: Groups similar SOM nodes into biologically meaningful metaclusters
FlowSOM meta-clustering aggregates individual SOM nodes into larger metaclusters, reducing the hundreds of nodes into a manageable number of biologically interpretable populations.
Question 5: In the context of cytometry data, what is 'arcsinh transformation' most commonly used for?
- Normalizing data to a standard Gaussian distribution
- Compressing high-intensity values while expanding low-intensity values for better visualization (Correct answer)
- Converting raw channel values to absolute molecule counts
- Removing background autofluorescence before analysis
Correct answer: Compressing high-intensity values while expanding low-intensity values for better visualization
Arcsinh transformation compresses the high end of the dynamic range while expanding detail in the low-signal region, making it ideal for cytometry data with a wide range of intensities.
Question 6: What is the primary advantage of using Cytobank's SPADE algorithm over simple manual gating for complex samples?
- SPADE eliminates the need for any antibody panel optimization
- SPADE automatically identifies hierarchical cell lineage relationships across all markers simultaneously (Correct answer)
- SPADE corrects spectral spillover without requiring single-color controls
- SPADE generates absolute cell counts without reference to sample volume
Correct answer: SPADE automatically identifies hierarchical cell lineage relationships across all markers simultaneously
SPADE creates a spanning-tree visualization that reveals hierarchical relationships between cell populations across the entire high-dimensional dataset simultaneously.
Question 7: Which approach is recommended when comparing cytometry data collected on different instruments or across multiple experimental days?
- Always re-gate all samples using a single universal gate set
- Use bead-based or cross-platform normalization to harmonize data before comparison (Correct answer)
- Analyze only samples collected on the same day without normalization
- Apply identical PMT voltages across all instruments regardless of their individual specifications
Correct answer: Use bead-based or cross-platform normalization to harmonize data before comparison
Bead-based or cross-platform normalization aligns fluorescence intensity scales across instruments or time points, enabling valid inter-experiment or inter-instrument comparisons.
SCYM Data Analysis 3
In dimensionality reduction analysis of high-dimensional cytometry data, what does t-SNE primarily preserve?