diff --git a/R/general_help.R b/R/general_help.R
index 67e6e7c13..853bfc91d 100644
--- a/R/general_help.R
+++ b/R/general_help.R
@@ -272,6 +272,50 @@ rank_binarize_wrapper <- function(
 }
 
 
+## chatgpt queries ####
+
+#' @title writeChatGPTqueryDEG
+#' @name writeChatGPTqueryDEG
+#' @description This function writes a query as a .txt file that can be used with 
+#' ChatGPT or a similar LLM service to find the most likely cell types based on the 
+#' top differential expressed genes (DEGs) between identified clusters.
+#' @param DEG_output the output format from the differenetial expression functions
+#' @param top_n_genes number of genes for each cluster
+#' @param tissue_type tissue type
+#' @param folder_name path to the folder where you want to save the .txt file
+#' @param file_name name of .txt file
+#' @returns writes a .txt file to the desired location
+#' @details This function does not run any LLM service. It simply creates the .txt 
+#' file that can then be used any LLM service (e.g. OpenAI, Gemini, ...)
+#' @export
+writeChatGPTquery = function(DEG_output, 
+                             top_n_genes = 10, 
+                             tissue_type = 'human breast cancer', 
+                             folder_name = getwd(), 
+                             file_name = 'chatgpt_query.txt') {
+  
+  chatgpt_query = paste0("Identify cell types of ", tissue_type, " tissue using the following markers. Identify one cell type for each row. Only provide the cell type name and the marker genes used for cell type identification.")
+  
+  selected_DEG_output = DEG_output[, head(.SD, top_n_genes), by="cluster"]
+  
+  finallist = list()
+  finallist[[1]] = chatgpt_query
+  
+  for(clus in unique(selected_DEG_output$cluster)) {
+    x = selected_DEG_output[cluster == clus][['feats']]
+    x = c(clus, x)
+    finallist[[as.numeric(clus)+1]] = x
+  }
+  
+  outputdt = data.table::data.table(finallist)
+  
+  cat('\n start writing \n')
+  data.table::fwrite(x = outputdt, 
+                     file = paste0(folder_name,'/', file_name),
+                     sep2 = c(""," ",""), col.names = F)
+  
+} 
+
 
 
 # IDs ####