- $M$ is the mapping matrix using tangram, with dimension $n_{cell} \times n_{spot}$
- $S$ is the gerated single cell reference from scGPT with dimension $n_{cell} \times n_{genes}$. The selection of genes have two options, the first case chosing its overlap with the marker genes, defined in [tangram](https://www.nature.com/articles/s41592-021-01264-7#Sec12). The second case is using the entire overlap genes between spatial data and generated single cell. Generally, case one would result in a higher correlation.
- $S$ is the gerated single cell reference from scGPT with dimension $n_{cell} \times n_{genes}$. The selection of genes have two options, the first case chosing its overlap with the marker genes, defined in [tangram](https://www.nature.com/articles/s41592-021-01264-7#Sec12). The second case is using the entire overlap genes between spatial data and generated single cell. Generally, case one would result in a higher correlation. Therefore, in the evaluation, we calculated the correlation using the overlapped marker genes.
- $G$ is the actual spatial expression matrix with dimension $n_{spot} \times n_{genes}$. The $n_{genes}$ is determined by the dimension of $S$.
The generation results are all stored within the folder **./scGPT/Generation_Result**. The generated single cell reference is stored in files with name **adata_result_XXX.h5ad**, the tangram mapping results is stored in files with name **scgpt_cell_reference_XXX.h5ad**. To findout the performance of our new method under different circumstances, we created three tables.
**Note**: After scGPT generation, there are **10727** intersected genes between the spatial and single cell expression data with **188** marker genes.
### Ratio Comparsion
**Note**: We used MOp 10Xv3 dataset. The default hyperparameters is $k=500$, $m=52$, $n_{genes}$ is the overlap with marker genes.
...
...
@@ -159,16 +161,18 @@ Seurat* -- Needs to be implemented
### Method Comparsion with Redeconve & Seurat (Human Breast)
**Note**: We used Human Breast dataset. The default hyperparameters is $ratio=0.8$, $k=1000$, $m=100$, $n_{genes}$ is the overlap with marker genes.
**Note**: We used Human Breast dataset. The default hyperparameters is $ratio=0.8$, $k=1000$, $m=100$, $n_{genes}$ is the overlap with marker genes. After scGPT generation, there are **14981** intersected genes between the spatial and single cell expression data with **166** marker genes.
| Methods | Median | Mean |
|----------------|-----------|-------|
| Ours (1000) | 0.712518 | 0.673199 |
| Tangram (1000) | 0.438570 | 0.427745 |
| Redeconve | 0.412310 | 0.423374 |
| Ours (1000) | 0.724916 | 0.680323 |
| Tangram (1000) | 0.423233 | 0.4183469 |
| Redeconve | 0.458208 | 0.453385 |
| Seurat | \ | \ |
| Ours (Full) | 0.928724 | 0.869383 |
| Tangram (Full) | 0.537428 | 0.526031 |
| Ours (Full) | 0.927419 | 0.870528 |
| Tangram (Full) | 0.538813 | 0.528211 |
### Method Comparsion with Redeconve & Seurat (Human Lymph Node)
