UNDER CONSTRUCTION
Testis 4 Individual Clustering
Normalize Data
sobj <- FindVariableFeatures(
sobj,
selection.method = "vst",
nfeatures = 2000,
verbose = FALSE
)
top20 <- head(VariableFeatures(sobj), 20)
Scale Data
all_genes <- rownames(sobj)
sobj <- ScaleData(
sobj,
features = all_genes,
vars.to.regress = "nCount_RNA",
verbose = FALSE
)Dimensionality Reduction
Dimensional Gene Loadings

Top Genes Heatmap

JackStraw

Elbow Plot

Dimension Selection
Clustering
| Resolution | Num of Clusters |
|---|---|
| 0.4 | 10 |
| 0.5 | 12 |
| 0.6 | 12 |
| 0.7 | 14 |
| 0.8 | 17 |
| 0.9 | 19 |
| 1 | 19 |
| 1.1 | 22 |
| 1.2 | 22 |

Look at marker genes
Spermatogonia


Spermatocytes


CySC


TE


PC


hub


Cluster Markers
| cluster | Num Biomarkers per Cluster |
|---|---|
| 0 | 105 |
| 1 | 927 |
| 2 | 536 |
| 3 | 120 |
| 4 | 560 |
| 5 | 1,162 |
| 6 | 786 |
| 7 | 1,167 |
| 8 | 2,188 |
| 9 | 1,652 |
| 10 | 719 |
| 11 | 1,090 |