Quality Control Report

demo chkv

Author

Affiliation

Eugénie Lohmann

CRCM (CiBi Group)

Report Parameters

Input parameters

	Sélection
fc	1
p_value	0.05
transform	FALSE
adcode	FALSE
analyse_type	QC
annotation_column	CellSubset
batch	acqdate
cluster_id	cluster_id
conditions	cond
log_bar	FALSE
min_cells	10
path	/root/analycyte-projects/demo_chkv/0823ECE7
project	demo chkv
rds	dataFE.rds
show_parameters	TRUE
show_summary	TRUE
show_heatmap_on_off	TRUE
sorted	TRUE
plot_legends	Gender

This document has been formated using knitr¹ and quarto².

1 Features summary

Samples informations

Sample table with information on conditions (cond), batch (acqdate), ncells (Number of Cells)

• percent.cluster : Percentage of clusters with cells (> 10 cells) in this fcs. see 1

1.0.1 Number of fcs per conditions

Condition-based distribution

Acute	Conv
43	43

Metacluster informations

Information on metaclusters is grouped in this table with the number of cells and the corresponding description.

• n_cells : Number of cells in the cluster.
• percent.n_cells : Percentage of total cells in the cluster.
• percent.fcs : Percentage of samples with cells (> 10 cells) in this cluster. see 1

Markers informations

A total of 37 markers are included in this analysis report.

Type markers (7): CD19, CD38, CD4, CD45RA, CD54, CD8, CXCR5.

State markers (30): CCR4, CCR5, CCR6, CCR7, CD11b, CD11c, CD123, CD127, CD14, CD141, CD16, CD161, CD1c, CD20, CD209, CD25, CD27, CD3, CD40, CD45, CD56, CD57, CD66b, CD80, CD86, CHIKV, CX3CR1, CXCR3, HLADR, PD1.

2 Presence / Absence data

With a minimum of 10 per cluster.

Figure 1: Heatmap Presence Absence

3 Number of cells Barplot

• Check the number of cells in each fcs.

• Controls the possible association with a batch effect.

Bars are sorted by conditions. Annotations columns are added on the barplot’s top to control possible batch effects.

Figure 2: Barplot number of cells per fcs

Less cells fcs : 55 1885 Acute (1446 cells), 43 1863 Acute (3493 cells), 59 1889 Acute (3549 cells), 21 1823 Acute (3807 cells), 19 1822 Acute (4087 cells), 41 1862 Acute (4171 cells)

Number of Cells per quantiles

0%	25%	50%	75%	100%
1446	8309.25	12556.5	16110.25	24631

Figure 3: Number of cells per fcs Histogram

(a) Cummulative annotation

Figure 4: Barplot number of cells per clusters

Less cells clusters : T Cell_unassigned (0.01%), Debris (0.11%), Dendritic Cell (Type 1, CD141+) (0.16%), Basophil (0.18%), CD4+ T Cell (EMRA) (0.23%), CD8+ T Cell (Effector Memory) (0.24%)

4 Density Heatmap

A special distance method, denoted as ‘ks,’ measures the similarity between distributions by computing the Kolmogorov-Smirnov statistic between two distributions.

Figure 5: Density heatmap of perCellCountsNorm clustered by ks

5 MFI’s Heatmap

• Check or help annotate clusters.

The input is the MFI cluster x marker array. The MFI is already transformed (typically using asinh(intensity/cofactor)).

Heatmap of the MFI with clusters in rows and markers in columns.

A barplot shows the number of cells in each metacluster.

The first heatmap covers all fcs.

Figure 6: Heatmap of the MFI

Figure 7: Heatmap of the MFI for Acute

Figure 8: Heatmap of the MFI for Conv

6 Mfi x Abundance Heatmap

• Heatmap grouping the MFI of each marker per cluster and cluster abundance per fcs.

• MFIs are normalized per line between 0 and 1 to give a comparable distribution of marker expression per cluster.

• Markers are grouped hierarchically to facilitate cluster reading/interpretation.

Relative abundances shown on the second heat map are transformed abundances relative to the cluster mean. The green-purple color scale applies to the entire matrix.

Figure 9: Abundance heatmap with clusters associated with mfi

7 Abundance Heatmap

• Visualize samples organized according to hierarchical grouping or by condition.

• Check that sample states are homogeneous; identify outliers or gender effects.

• Check whether conditions have already been separated.

Percentage heatmap with samples in columns and clusters in rows.

By default, clusters and samples are grouped using hierarchical clustering. It is also possible to classify samples by condition, gender or patient, and to give clusters as an ordered vector.

Group abundance is shown in the right-hand barplot with log values.

Figure 10: Abundance Heatmap

8 PCA of samples using percentages per cluster

• Visualize samples in a reduced space.

• Check that sample states are homogeneous; identify outliers or gender effects.

• Check whether conditions can already be separated.

• Based on the factormineR[tbd] PCA function.

Input is the percentages table sample x cluster. Percentages are previously transformed with asinh(%+cte) with cte = 0.03 and centered by cluster.

• check that the first 2 or 3 components retains the main information

• Graphical figures generated with factoextra³ functions.

8.1 Barplot of variance of each PC

Figure 11: Barplot of variance of each PC

8.2 Dim.1 vs Dim.2

Figure 12: Dim.1 vs Dim.2 by cond

Figure 13: Dim.1 vs Dim.2 by acqdate

Figure 14: Dim.1 vs Dim.2 by Gender

Are annotated on biplot the correlations with cos2 on axes 1, 2 > to the cos2 mean

Figure 15: Dim.1 vs Dim.2

8.3 Dim.1 vs Dim.3

Figure 16: Dim.1 vs Dim.3 by cond

Figure 17: Dim.1 vs Dim.3 by acqdate

Figure 18: Dim.1 vs Dim.3 by Gender

Are annotated on biplot the correlations with cos2 on axes 1, 3 > to the cos2 mean

Figure 19: Dim.1 vs Dim.3

8.4 Dim.2 vs Dim.3

Figure 20: Dim.2 vs Dim.3 by cond

Figure 21: Dim.2 vs Dim.3 by acqdate

Figure 22: Dim.2 vs Dim.3 by Gender

Are annotated on biplot the correlations with cos2 on axes 2, 3 > to the cos2 mean

Figure 23: Dim.2 vs Dim.3

8.5 Contribution and Correlation of variables to each axis

The table presents contributions, ensuring that at least one parameter within the first five primary dimensions contributes a minimum of 1%.

9 PCA of samples using MFI

Plot variables in order to identify their contribution to each PC

PCA of samples using MFI per cluster Similar to Nowicka et al., but at the clustering level (not the cell level).

• View the samples in a reduced space.

• Verify that States of samples are homogeneous; identify outliers or Gender effects.

• Check if the conditions could already be separated.

Input is the MFI table sample x cluster. MFI are previously transformed.

• Based on the factormineR[tbd] PCA function.

Figure 24: Barplot of variance of each PC

9.1 Dim.1 vs Dim.2

Figure 25: Dim.1 vs Dim.2 by cond

Figure 26: Dim.1 vs Dim.2 by acqdate

Figure 27: Dim.1 vs Dim.2 by Gender

Are annotated on biplot the correlations with cos2 on axes 1, 2 > to the cos2 mean

Figure 28: Dim.1 vs Dim.2

Figure 29: Heatmap Top 10 cos2 for :Dim.1 vs Dim.2

9.2 Dim.1 vs Dim.3

Figure 30: Dim.1 vs Dim.3 by cond

Figure 31: Dim.1 vs Dim.3 by acqdate

Figure 32: Dim.1 vs Dim.3 by Gender

Are annotated on biplot the correlations with cos2 on axes 1, 3 > to the cos2 mean

Figure 33: Dim.1 vs Dim.3

Figure 34: Heatmap Top 10 cos2 for :Dim.1 vs Dim.3

9.3 Dim.2 vs Dim.3

Figure 35: Dim.2 vs Dim.3 by cond

Figure 36: Dim.2 vs Dim.3 by acqdate

Figure 37: Dim.2 vs Dim.3 by Gender

Are annotated on biplot the correlations with cos2 on axes 2, 3 > to the cos2 mean

Figure 38: Dim.2 vs Dim.3

Figure 39: Heatmap Top 10 cos2 for :Dim.2 vs Dim.3

9.4 Contribution and Correlation of variables to each axis

The table presents contributions, ensuring that at least one parameter within the first five primary dimensions contributes a minimum of 1%.

Glossary

• counts: Raw count abundances.
• perCellCounts: Frequencies abundances.
• perCellCountsNorm: Argsinh frequencies / 0.03 then centered by the mean per clusters.

R session info

R version 4.3.3 (2024-02-29)

R version 4.3.3 (2024-02-29)

Platform: x86_64-pc-linux-gnu (64-bit)

locale: LC_CTYPE=en_US.UTF-8, LC_NUMERIC=C, LC_TIME=en_US.UTF-8, LC_COLLATE=en_US.UTF-8, LC_MONETARY=en_US.UTF-8, LC_MESSAGES=en_US.UTF-8, LC_PAPER=en_US.UTF-8, LC_NAME=C, LC_ADDRESS=C, LC_TELEPHONE=C, LC_MEASUREMENT=en_US.UTF-8 and LC_IDENTIFICATION=C

attached base packages: grid, stats4, stats, graphics, grDevices, utils, datasets, methods and base

other attached packages: knitr(v.1.46), checkmate(v.2.3.1), ggiraph(v.0.8.9), reactable(v.0.4.4), plotly(v.4.10.4), tibble(v.3.2.1), kableExtra(v.1.4.0), ggrepel(v.0.9.5), ggplot2(v.3.5.1), tidyr(v.1.3.1), data.table(v.1.15.4), colorRamp2(v.0.1.0), randomcoloR(v.1.1.0.1), RColorBrewer(v.1.1-3), ComplexHeatmap(v.2.18.0), SummarizedExperiment(v.1.32.0), Biobase(v.2.62.0), GenomicRanges(v.1.54.1), GenomeInfoDb(v.1.38.8), IRanges(v.2.36.0), S4Vectors(v.0.40.2), BiocGenerics(v.0.48.1), MatrixGenerics(v.1.14.0), matrixStats(v.1.3.0), dplyr(v.1.1.4), readr(v.2.1.5) and analycyte.utils(v.3.0.11.2)

loaded via a namespace (and not attached): rstudioapi(v.0.16.0), jsonlite(v.1.8.8), shape(v.1.4.6.1), magrittr(v.2.0.3), TH.data(v.1.1-2), estimability(v.1.5), farver(v.2.1.1), rmarkdown(v.2.26), GlobalOptions(v.0.1.2), zlibbioc(v.1.48.2), vctrs(v.0.6.5), Cairo(v.1.6-2), RCurl(v.1.98-1.14), rstatix(v.0.7.2), htmltools(v.0.5.8), S4Arrays(v.1.2.1), curl(v.5.2.1), broom(v.1.0.5), SparseArray(v.1.2.4), htmlwidgets(v.1.6.4), sandwich(v.3.1-0), emmeans(v.1.10.1), zoo(v.1.8-12), lubridate(v.1.9.3), uuid(v.1.2-0), lifecycle(v.1.0.4), iterators(v.1.0.14), pkgconfig(v.2.0.3), Matrix(v.1.6-5), R6(v.2.5.1), fastmap(v.1.1.1), GenomeInfoDbData(v.1.2.11), clue(v.0.3-65), digest(v.0.6.35), colorspace(v.2.1-0), crosstalk(v.1.2.1), ggpubr(v.0.6.0), labeling(v.0.4.3), fansi(v.1.0.6), timechange(v.0.3.0), httr(v.1.4.7), abind(v.1.4-5), compiler(v.4.3.3), withr(v.3.0.0), doParallel(v.1.0.17), pander(v.0.6.5), backports(v.1.4.1), carData(v.3.0-5), highr(v.0.10), ggsignif(v.0.6.4), MASS(v.7.3-60.0.1), DelayedArray(v.0.28.0), rjson(v.0.2.21), scatterplot3d(v.0.3-44), flashClust(v.1.01-2), tools(v.4.3.3), FactoMineR(v.2.11), glue(v.1.7.0), Rtsne(v.0.17), cluster(v.2.1.6), generics(v.0.1.3), gtable(v.0.3.4), tzdb(v.0.4.0), hms(v.1.1.3), car(v.3.1-2), xml2(v.1.3.6), utf8(v.1.2.4), XVector(v.0.42.0), foreach(v.1.5.2), pillar(v.1.9.0), stringr(v.1.5.1), ggprism(v.1.0.5), circlize(v.0.4.16), splines(v.4.3.3), lattice(v.0.22-5), survival(v.3.5-8), tidyselect(v.1.2.1), V8(v.4.4.2), svglite(v.2.1.3), xfun(v.0.43), factoextra(v.1.0.7), DT(v.0.33), stringi(v.1.8.3), lazyeval(v.0.2.2), yaml(v.2.3.8), evaluate(v.0.23), codetools(v.0.2-19), multcompView(v.0.1-10), cli(v.3.6.2), xtable(v.1.8-4), systemfonts(v.1.0.6), munsell(v.0.5.0), jquerylib(v.0.1.4), Rcpp(v.1.0.12), png(v.0.1-8), parallel(v.4.3.3), leaps(v.3.1), bitops(v.1.0-7), viridisLite(v.0.4.2), mvtnorm(v.1.2-4), scales(v.1.3.0), purrr(v.1.0.2), crayon(v.1.5.2), GetoptLong(v.1.0.5), rlang(v.1.1.3), cowplot(v.1.1.3) and multcomp(v.1.4-25)

References

References

1.

Xie, Y. Dynamic documents with R and knitr. (Chapman; Hall/CRC, 2015).

2.

Allaire, J. Quarto: R interface to ’quarto’ markdown publishing system. (2022).

3.

Kassambara, A. & Mundt, F. Factoextra: Extract and visualize the results of multivariate data analyses. (2020).

Reuse