The cyclins control progression through the cell cycle and have well-characterized patterns of expression across cell cycle phases.
cyclin A: activates DNA replication in S phase
cyclin B: assembly of mitotic spindle to preparate for mitosis
Cyclin D: move from G0 to G1 and S
Cyclin E: prepares for DNAreplication just before S phase.
The plots are saved in “outs/old/Cyclins”
The key assumption is that the cell cycle effect is orthogonal to other aspects of biological heterogeneity like cell type. This justifies the use of a reference involving cell types that are different from the cells in our dataset, provided that the cell cycle transcriptional program is conserved across datasets (Bertoli, Skotheim, and Bruin 2013; Conboy et al. 2007). Non-orthogonality can introduce biases where, one cell type is consistently misclassified as being in a particular phase because it happens to be more similar to that phase’s profile in the reference. A healthy dose of skepticism is required when interpreting these assignments. Our hope is that any systematic assignment error is consistent across clusters and conditions such that they cancel out in comparisons of phase frequencies.
Filter only the genes related with cell cycle (GO:0007049) from the reference dataset
Map cells from our dataset to reference
Calculate number of cells in each phase for each one of the clusters and condition.
outs/old/cellcycle.csv contains a table with the number of cells found in each phase, for each one of the clusters, divided by WT and KO.
We can compare if there is a significant difference in the distribution of cells across the G1, G2M and S phases between the WT and the KO.
##
## Pearson's Chi-squared test
##
## data: freq_WT.KO_G1.G2M.S
## X-squared = 15.216, df = 2, p-value = 0.0004965We can also perform the same test in each one of our cell types. Due to the presence of small counts (expected frequencies <5) we use the fisher’s exact test, as advised in (Hae-Young Kim et. al.)
## $Astrocyte
##
## Fisher's Exact Test for Count Data
##
## data: freq_WT.KO_G1.G2M.S
## p-value = 0.216
## alternative hypothesis: two.sided
##
##
## $OligoAstro
##
## Fisher's Exact Test for Count Data
##
## data: freq_WT.KO_G1.G2M.S
## p-value = 1
## alternative hypothesis: two.sided
##
##
## $Oligo
##
## Fisher's Exact Test for Count Data
##
## data: freq_WT.KO_G1.G2M.S
## p-value = 0.0596
## alternative hypothesis: two.sided
##
##
## $OPCs
##
## Fisher's Exact Test for Count Data
##
## data: freq_WT.KO_G1.G2M.S
## p-value = 0.2368
## alternative hypothesis: two.sided
##
##
## $Neuron
##
## Fisher's Exact Test for Count Data
##
## data: freq_WT.KO_G1.G2M.S
## p-value = 0.4468
## alternative hypothesis: two.sided
##
##
## $Lymphocytes
##
## Fisher's Exact Test for Count Data
##
## data: freq_WT.KO_G1.G2M.S
## p-value = 0.3811
## alternative hypothesis: two.sided
##
##
## $`Gran & Mono`
##
## Fisher's Exact Test for Count Data
##
## data: freq_WT.KO_G1.G2M.S
## p-value = 0.4639
## alternative hypothesis: two.sided
##
##
## $BAMs
##
## Fisher's Exact Test for Count Data
##
## data: freq_WT.KO_G1.G2M.S
## p-value = 1
## alternative hypothesis: two.sided
##
##
## $Microglia
##
## Fisher's Exact Test for Count Data
##
## data: freq_WT.KO_G1.G2M.S
## p-value = 0.1442
## alternative hypothesis: two.sided
##
##
## $DCs
##
## Fisher's Exact Test for Count Data
##
## data: freq_WT.KO_G1.G2M.S
## p-value = 0.5014
## alternative hypothesis: two.sided
##
##
## $Endothelial
##
## Fisher's Exact Test for Count Data
##
## data: freq_WT.KO_G1.G2M.S
## p-value = 0.4909
## alternative hypothesis: two.sided
##
##
## $Mural_cells
##
## Fisher's Exact Test for Count Data
##
## data: freq_WT.KO_G1.G2M.S
## p-value = 0.569
## alternative hypothesis: two.sided
##
##
## $ChP_epithelial
##
## Fisher's Exact Test for Count Data
##
## data: freq_WT.KO_G1.G2M.S
## p-value = 0.09148
## alternative hypothesis: two.sided
##
##
## $Ependymocytes
##
## Fisher's Exact Test for Count Data
##
## data: freq_WT.KO_G1.G2M.S
## p-value = 0.4979
## alternative hypothesis: two.sided
##
##
## $OEG
##
## Fisher's Exact Test for Count Data
##
## data: freq_WT.KO_G1.G2M.S
## p-value = 0.6115
## alternative hypothesis: two.sided