Add custom domain information

[sdhutchins]

Hi, is it possible to add custom domain information? Below is my script I'm using, but not sure if there is something set up within the package to do this.

The domain_data_formatted variable is what I'd be inputting.

@phoeguo @zhangrener

# Load necessary libraries
library(dplyr)
library(g3viz)
library(jsonlite)
library(dplyr)


# Define the UniProt REST API URL for BMPR2
uniprot_url <- "https://rest.uniprot.org/uniprotkb/Q13873.json"  # BMPR2 UniProt ID

# Fetch and parse the JSON data directly
parsed_data <- fromJSON(uniprot_url)

# Extract relevant domain information
domains <- parsed_data$features %>%
  filter(type %in% c("Signal", "Domain", "Transmembrane", "Topological domain")) %>%
  transmute(
    Description = description,
    Start = sapply(location$start, as.numeric),  # Safely extract start positions
    End = sapply(location$end, as.numeric),      # Safely extract end positions
    Type = type
  )

# Print the extracted domains
print(domains)

# Format retrieved domain data for g3viz
domain_data_formatted <- list(
  length = parsed_data$sequence$length,  # Total length of BMPR2 protein
  domainType = "Custom Domains",
  details = domains %>%
    transmute(
      start = Start,
      end = End,
      name = Description
    )
)

# Print formatted domain data
print(domain_data_formatted)

[bmpr2_clinvar_pathogenic.csv](https://github.com/user-attachments/files/18461874/bmpr2_clinvar_pathogenic.csv)

# Load your data (assuming it's saved as a CSV file)
data <- read.csv("data/bmpr2_clinvar_pathogenic.csv")

# Prepare data for g3Lollipop
processed_data <- data %>%
  mutate(
    Hugo_Symbol = GeneSymbol,  # Map GeneSymbol to Hugo_Symbol
    Mutation_Class = case_when(
      Type == "SNV" & consequence == "Stop gain" ~ "Nonsense_Mutation",
      Type == "SNV" & consequence == "Missense" ~ "Missense_Mutation",
      TRUE ~ "Other"
    ),
    Protein_Change = paste0("p.", ref_aa, pos_aa, alt_aa),  # Create Protein_Change column
    # Extract AA_Position from pos_aa
    AA_Position = as.numeric(pos_aa)
  ) %>%
  select(Hugo_Symbol, Mutation_Class, Protein_Change, AA_Position)  # Select required columns

# Verify processed data
print(head(processed_data))

plot.options <- g3Lollipop.theme(theme.name = "cbioportal",
                                 title.text = "BMPR2 gene",
                                 y.axis.label = "# of BMPR2 Mutations")

# Use processed data with g3viz
g3Lollipop(
  processed_data,
  gene.symbol = "BMPR2",  # Specify target gene
  gene.symbol.col = "Hugo_Symbol",  # Specify gene symbol column
  aa.pos.col = "AA_Position",  # Specify amino acid position column
  protein.change.col = "Protein_Change",  # Specify protein change column
  factor.col = "Mutation_Class",  # Specify mutation classification column
  plot.options = plot.options,  # Use default theme
  output.filename = "BMPR2_ClinVar_Visualization"  # Save output
)