01-TCMSP.Rmd

```{r}
my_packages <- c("tidyverse", "data.table",
                 "ggplot2", "ggpubr", "patchwork",
                 "ggvenn")
pacman::p_load(char = my_packages)

tmp <- list()
```

# Preparation
Shen Ling Bai Zhu San(SLBZS)

## Uniprot
```{r}
# gancao_drug <- fread("Input/TCMSP/gancao/chemical_attr.csv")
# gancao_target <- fread("Input/TCMSP/gancao/chemical_tar.csv")
```

Uniprot:
```{r}
uniprot_df <- fread("Input/uniprotkb_AND_model_organism_9606_AND_r_2024_01_22.tsv.gz")

# table(gancao_target$target_name %in% uniprot_df$`Protein names`)
```

```{r}
uniprot_df$ProteinNames <- gsub(" \\(.*", "", 
                                   uniprot_df$`Protein names`)

# table(gancao_target$target_name %in% uniprot_df$ProteinNames)
# gancao_target$target_name[!gancao_target$target_name %in% uniprot_df$ProteinNames] %>% head()
```

Discard unmapped targets.

```{r}
uniprot_df <- separate(uniprot_df, col = `Gene Names`, into = c("GeneNames", "others"),sep = " ")
```


```{r}
# gancao_target_sel <- gancao_target[!gancao_target$target_name %in% uniprot_df$ProteinNames,] 
# gancao_target_sel_vec <- unique(gancao_target_sel$target_name)
# 
# test <- lapply(gancao_target_sel_vec, function(target){
#   grep(target, uniprot_df$ProteinNames)
# })
```

## TCMSP
```{r}
herbs <- dir("Input/TCMSP")
herbs <- herbs[!herbs %in% c("gancao")]
target_df <- lapply(herbs, function(x){
  df <- fread(paste0("Input/TCMSP/", x, "/chemical_tar.csv"))
  df$herb <- x
  df
})
target_df <- do.call(rbind, target_df)

attr_df <- lapply(herbs, function(x){
  df <- fread(paste0("Input/TCMSP/", x, "/chemical_attr.csv"))
  df$herb <- x
  df
})
attr_df <- do.call(rbind, attr_df)
```


# Filter chemicals
Ref: Gancao Nurish-Yin Decoction medicated serum inhibits growth and migration of ovarian cancer cells: Network pharmacology-based analysis and biological validation
with cut off values for oral bioavailability (OB) ≥ 30% and drug-like properties (DL) ≥ 0.18.

```{r}
attr_df2 <- attr_df[attr_df$ob >= 30 & attr_df$dl >= 0.18,]
```

```{r}
target_df2 <- target_df[target_df$MOL_ID %in% attr_df2$MOL_ID,]

table(unique(target_df2[,c("herb", "molecule_name")])$herb)
```

# Merge with uniprot
```{r}
uniprot_df3 <- uniprot_df[,c(3,5,7)]

target_df3 <- merge(target_df2, uniprot_df3, 
                    by.x = "target_name", by.y = "ProteinNames")
```

```{r}
table(unique(target_df3[,c("herb", "molecule_name")])$herb)
```

# Merge with disease genes
genecard:
```{r}
genecard_df <- fread("Input/Target/GeneCards-SearchResults.csv")
genecard_df2 <- genecard_df[genecard_df$`Relevance score` > 35,]
```

DisGeNET:
```{r}
disgenet_df <- fread("Input/Target/C0919267__C0029925__C0677886_disease_gda_evidences.tsv")
disgenet_df2 <- disgenet_df[disgenet_df$Score_gda > .4,]
```


```{r}
ovarian_genes <- unique(c(genecard_df2$`Gene Symbol`, disgenet_df2$Gene))

target_df4 <- target_df3[target_df3$GeneNames %in% ovarian_genes,]

length(unique(target_df4$GeneNames))
table(unique(target_df4[,c("herb", "molecule_name")])$herb)
```

# ggVenn
```{r}
ovarian_genes_list <- list(
  `SLBZS genes` = unique(target_df3$GeneNames),
  `Ovarian cancer genes` = ovarian_genes
)

p1 <- ggvenn(ovarian_genes_list, set_name_size = 6)
p1
```

```{r}
ggsave(plot = p1,
  filename = "Plot/p1.png",
  width = 7,             
  height = 7,           
  units = "in",         
  dpi = 300)
```

# Save
```{r}
saveRDS(target_df4, file = "Output/01/target_df4.Rds")
```