如何确定细胞聚类的PC数

library(Seurat)

load(file = 'Cluster_seurat.Rdata') # data.filt
seurat_data <- data.filt

# Explore heatmap of PCs
DimHeatmap(seurat_data, dims = 1:6, cells = 500, balanced = TRUE)

DimHeatmap(seurat_data , dims = 7:12, cells = 500, balanced = TRUE)

# Plot the elbow plot
ElbowPlot(object = seurat_data ,  ndims = 30)

# Slow slow slow
seurat_data  <- JackStraw(object = seurat_data, dims = 50)
seurat_data  <- ScoreJackStraw(seurat_data, dims = 1:50)
JackStrawPlot(object = seurat_data, dims = 1:50)

# Determine percent of variation associated with each PC
pct <- seurat_data [["pca"]]@stdev / sum( seurat_data [["pca"]]@stdev) * 100


# Calculate cumulative percents for each PC
cumu <- cumsum(pct)


# Determine which PC exhibits cumulative percent greater than 90% and % variation associated with the PC as less than 5
co1 <- which(cumu > 90 & pct < 5)[1]
co1

# Determine the difference between variation of PC and subsequent PC
co2 <- sort(which((pct[1:length(pct) - 1] - pct[2:length(pct)]) > 0.1), decreasing = T)[1] + 1


# last point where change of % of variation is more than 0.1%.
co2

# Minimum of the two calculation
pcs <- min(co1, co2)
pcs

# Create a dataframe with values
plot_df <- data.frame(pct = pct,   cumu = cumu,   rank = 1:length(pct))


# Elbow plot to visualize 
ggplot(plot_df, aes(cumu, pct, label = rank, color = rank > pcs)) + 
  geom_text() + 
  geom_vline(xintercept = 90, color = "grey") + 
  geom_hline(yintercept = min(pct[pct > 5]), color = "grey") +
  theme_bw()

# Printing out the most variable genes driving PCs
print(x = seurat_data [["pca"]],  dims = 1:25,  nfeatures = 5)

PC_ 1 
Positive:  NEIL1, LTB, KLF2, TP53INP1, CD27 
Negative:  TYMS, MKI67, PCLAF, RRM2, NUSAP1 
PC_ 2 
Positive:  GZMA, ARL4C, PRF1, CST7, GZMM 
Negative:  SLC35E3, ID3, PRDX1, TOP2B, RPLP0 
PC_ 3 
Positive:  HBA2, HBB, HBA1, AHSP, HBD 
Negative:  RPS18, RPL18A, RPS2, RPSA, RPL37A 
PC_ 4 
Positive:  IGLL1, SLC35E3, PCDH9, CD38, F13A1 
Negative:  CCL17, HMBS, BLVRB, AQP1, CD36 
PC_ 5 
Positive:  GYPC, RPS18, RPS2, C1QTNF4, RPL18A 
Negative:  MNDA, LYZ, S100A9, S100A8, FCN1 
PC_ 6 
Positive:  PLK1, CDC20, CENPA, HMMR, CENPE 
Negative:  GINS2, MCM6, HELLS, MCM4, MCM3 
PC_ 7 
Positive:  GYPC, C1QTNF4, LIMS1, NRIP1, S100A9 
Negative:  SPIB, TAGLN2, MS4A1, IGLC6, PTPRC 
PC_ 8 
Positive:  FCGR3A, GZMB, SPON2, KLRF1, MYOM2 
Negative:  CCR7, CD3G, CD3D, IL7R, GPR183 
PC_ 9 
Positive:  CCL17, LTB, TMEM154, CCND2, HSPA12B 
Negative:  ACTG1, LGALS1, IGLL1, CCDC81, TOP2B 
PC_ 10 
Positive:  AHNAK, VIM, EMP1, LMNA, CD27 
Negative:  MT1X, CCL17, FTL, HSP90B1, NSMCE1 
PC_ 11 
Positive:  NEIL1, LTB, FTH1, CFD, CST3 
Negative:  LCN2, RETN, S100A8, LTF, CAMP 
PC_ 12 
Positive:  RPS12, RPLP1, RPL18A, EEF1B2, RPS5 
Negative:  HNRNPU, NCL, AHNAK, AC245060.5, EMP1 
PC_ 13 
Positive:  CD3D, TRAC, CD3G, IGLC6, CD27 
Negative:  MARCH1, MS4A1, BANK1, ADAM28, LINC02397 
PC_ 14 
Positive:  SCIMP, SRGN, GUSB, SHISA2, MARCH1 
Negative:  MS4A1, ZNF608, ENAM, CCND2, CCL17 
PC_ 15 
Positive:  ATF5, HSPA5, PSAT1, PHGDH, MARCH1 
Negative:  NT5E, GIMAP4, TP53INP1, SHISA2, DBI 
PC_ 16 
Positive:  ACSM3, IGLC6, SHISA2, REXO2, MT1X 
Negative:  CD82, GCHFR, PRDX1, UBASH3B, PTGDR 
PC_ 17 
Positive:  MARCKSL1, FTH1, S100A1, CRIP2, EMP2 
Negative:  HSP90B1, HSPA5, UBASH3B, PPIB, FKBP5 
PC_ 18 
Positive:  MARCH1, H3F3A, CALM2, ACTB, PRDX1 
Negative:  HSP90B1, ATF5, HSPA5, MT-ND6, CANX 
PC_ 19 
Positive:  TRGC2, LGALS1, KLRG1, CCL5, PTMS 
Negative:  CCR7, TXK, FCER1G, CD7, TCF7 
PC_ 20 
Positive:  PIM1, SOCS3, ADGRE5, RGCC, EPHA4 
Negative:  LRMP, BANK1, MS4A1, CLEC4E, NME1 
PC_ 21 
Positive:  CCR7, CMTM2, S100A11, LRMP, TXK 
Negative:  TRGC2, RPS12, KLRG1, LCN6, RPS18 
PC_ 22 
Positive:  CTGF, PMAIP1, FOS, KLF6, FOSB 
Negative:  FUT7, SLC9A3R2, LCN6, PPP1R14A, EMP3 
PC_ 23 
Positive:  ATF5, PSAT1, HSP90B1, PHGDH, HSPA5 
Negative:  CTHRC1, NSMCE1, MAP1A, IGLL1, BTNL9 
PC_ 24 
Positive:  SERINC2, LST1, NAMPT, MT1X, SLC25A37 
Negative:  SHISA2, DEPP1, GADD45A, PSTPIP2, CD33 
PC_ 25 
Positive:  CDKN1C, RHOB, BATF3, CX3CR1, SERPINA1 
Negative:  FOS, ALDH2, MGST1, MPO, FOSB