We are sorry, but NCBI web applications do not support your browser and may not function properly. More information

Results: 5

1.
Figure 5

Figure 5. Multicolor immunophenotyping and the spillover spreading matrix. From: Quantifying Spillover Spreading for Comparing Instrument Performance and Aiding in Multicolor Panel Design.

(Left) Bivariate distributions of CD4+ T lymphocytes is shown for three different combinations of reagents, showing the expression pattern for CD25 (conjugated to Cy5PE) vs CD45RA (on three different fluorochromes) for the same PBMC sample. The degree of spillover spreading (indicated by the dashed lines) limits sensitivity for detection of CD25. Data taken from a published OMIP (6). (Right) The heatmap representation for spillover spreading into the G660 detector (for Cy5PE) on the instrument. The three values corresponding to the reagent combinations shown on the left are indicated.

Richard Nguyen, et al. Cytometry A. ;83(3):306-315.
2.
Figure 4

Figure 4. The spillover spreading matrix (SSM) for instrument comparison. From: Quantifying Spillover Spreading for Comparing Instrument Performance and Aiding in Multicolor Panel Design.

The SSM was computed for the same stained bead samples collected on each of five nearly-identically-configured instruments. (A) A heatmap representation of the SSM is shown for two instruments (LSR-A, LSR-D). The colored bars on the left and top indicate which laser was used for detection (cyan: 488 nm; green: 532 nm; red: 632 nm; violet: 405 nm). (B) The top row show the heatmaps for all five instruments. The bottom row shows the ratio of each instruments’ SS values from the average across all five (for SS values less than 0.5, the ratio was set to zero for the heatmap). Green indicates SS values smaller than average; red indicates SS values larger than average.

Richard Nguyen, et al. Cytometry A. ;83(3):306-315.
3.
Figure 3

Figure 3. Spillover spreading values reflect noise contributions independently from compensation values. From: Quantifying Spillover Spreading for Comparing Instrument Performance and Aiding in Multicolor Panel Design.

(A) Spectral overlap matrix and SSM for a 4-color experiment with standard filters or with an added ND2.0 (i.e., reducing signal by 100-fold) in the V800 parameter. (B) Compensated distributions of CD8 Cy7APC-stained beads in the V800 (spillover) parameter for data collected with different ND filters in the V800 detector. (C) Relationship of the spillover spreading (SS) value to the signal level for CD8 Cy7APC-stained beads in the V800 detector (top) or for FITC stained beads into the G560 (PE) detector (bottom). Note that the SS values are plotted on a log scale; ND values scale logarithmically with signal attenuation. The line is a linear least-squares fit of the log-scaled data; the blue dashed lines indicate a theoretical relationship for which SS scales by the square-root of signal-to-noise. The deviation from this relationship is likely due to an additional linear dependence of SS on signal-to-noise (Appendix; eq. 9).

Richard Nguyen, et al. Cytometry A. ;83(3):306-315.
4.
Figure 1

Figure 1. Computation of spillover spreading and relationship to fluorescence intensity. From: Quantifying Spillover Spreading for Comparing Instrument Performance and Aiding in Multicolor Panel Design.

(A) Compensation beads were labeled to saturation with Cy7APC CD8 (red, “sample”) or left unstained (green, “reference”); data shown are compensated for Cy7APC spillover into the QD800 detector. Lines indicate the approximate locations of various fluorescence measures used in computing spillover spreading. (B) ΔFC7 is the difference in median fluorescence of the reference and sample in the primary parameter (Cy7APC, on the R780 detector). The spillover spreading (indicated by the grey line) theoretically increases with the square root of fluorescence once it exceeds background fluorescence; this relationship is a straight line with a slope of ½ on a log-log plot (see also Supplemental Figure 1). (C) The increased standard deviation due to spillover spreading (ΔσQD) is computed based on the measured variances in the spillover parameter of the reference and sample. (D) Compensation beads were incubated with 2-fold dilutions of Cy7APC CD8. The insert shows the gating for singlet beads. Compensated data are shown. (E, F) The relationship of Δσ to ΔF is shown for Cy7APC into the QD800 detector V800 (E) and FITC into the PE detector G560 (F). Note that the x-axis is non-linearly scaled (by the square-root).

Richard Nguyen, et al. Cytometry A. ;83(3):306-315.
5.
Figure 2

Figure 2. Spillover spreading can be computed from beads or cells. From: Quantifying Spillover Spreading for Comparing Instrument Performance and Aiding in Multicolor Panel Design.

(A) Beads or PBMC were stained with 16 fluorescent conjugates and used to create respective compensation matrices. The compensated data were then used to determine the SSM for each. Each cell of the heatmap is colored according to the magnitude of the SS value. The colored bars on the left and top indicate which laser was used for detection (cyan: 488 nm; green: 532 nm; red: 632 nm; violet: 405 nm). (B) The 240 off-diagonal SS elements of the two SSMs shown in panel A are graphed against each other to demonstrate quantitative similarity. A least-squares regression on the log-transformed values was performed. Most outliers occur with SS values <1, which are less reliable and perhaps trivial (see Supplemental Figure 2). Indeed, the correlation coefficient (r2) for SS values >1 is 0.80. (C) Cells (top) or beads (bottom) labeled with Cy5PE CD8 are shown following compensation for the primary parameter (G660) vs. a spillover parameter (G710). The black (cells) or purple (beads) lines are a depiction of the approximate spreading for each control type; it rises only when the variance from spillover spreading exceeds that of the unstained substrate.

Richard Nguyen, et al. Cytometry A. ;83(3):306-315.

Supplemental Content

Recent activity

Your browsing activity is empty.

Activity recording is turned off.

Turn recording back on

See more...
Write to the Help Desk