FIX: openai api call

alphastats/gui/pages/02_Import Data.py

@@ -279,9 +279,9 @@ def load_sample_data():
 
     loader = MaxQuantLoader(file=filepath)
     ds = DataSet(loader=loader, metadata_path=metadatapath, sample_column="sample")
-    metadatapath = os.path.join(_this_directory, "sample_data/metadata.xlsx").replace(
+    metadatapath = os.path.join(_this_directory, "sample_data", "metadata.xlsx").replace(
         "pages/", ""
-    )
+    ).replace("pages\\", "")
 
     loader = MaxQuantLoader(file=filepath)
     ds = DataSet(loader=loader, metadata_path=metadatapath, sample_column="sample")

alphastats/gui/pages/05_GPT.py

@@ -1,3 +1,4 @@
+import os
 import streamlit as st
 from openai import OpenAI
 
@@ -256,20 +257,20 @@ def select_analysis():
         display_proteins([], st.session_state["downregulated"])
 
 st.session_state["instructions"] = (
-    "You are an expert biologist and have extensive experience in molecular biology, medicine and biochemistry.\n"
+    f"You are an expert biologist and have extensive experience in molecular biology, medicine and biochemistry.{os.linesep}"
     "A user will present you with data regarding proteins upregulated in certain cells "
     "sourced from UniProt and abstracts from scientific publications. They seek your "
     "expertise in understanding the connections between these proteins and their potential role "
-    "in disease genesis. \nProvide a detailed and insightful, yet concise response based on the given information. "
+    f"in disease genesis. {os.linesep}Provide a detailed and insightful, yet concise response based on the given information. "
     f"The data you have has following groups and respective subgroups: {str(get_subgroups_for_each_group(st.session_state.dataset.metadata))}."
     "Plots are visualized using a graphical environment capable of rendering images, you don't need to worry about that."
 )
 if "column" in chosen_parameter_dict and "upregulated" in st.session_state:
     st.session_state["user_prompt"] = (
         f"We've recently identified several proteins that appear to be differently regulated in cells "
         f"when comparing {chosen_parameter_dict['group1']} and {chosen_parameter_dict['group2']} in the {chosen_parameter_dict['column']} group. "
-        f"From our proteomics experiments, we know that the following ones are upregulated: {', '.join(st.session_state['upregulated'])}.\n\n"
-        f"Here is the list of proteins that are downregulated: {', '.join(st.session_state['downregulated'])}.\n\n"
+        f"From our proteomics experiments, we know that the following ones are upregulated: {', '.join(st.session_state['upregulated'])}.{os.linesep}{os.linesep}"
+        f"Here is the list of proteins that are downregulated: {', '.join(st.session_state['downregulated'])}.{os.linesep}{os.linesep}"
         f"Help us understand the potential connections between these proteins and how they might be contributing "
         f"to the differences. After that provide a high level summary"
     )

alphastats/gui/utils/gpt_helper.py

@@ -1,3 +1,4 @@
+import copy
 from typing import Optional, Union
 from pathlib import Path
 import requests
@@ -315,8 +316,8 @@ def get_assistant_functions(
                             "description": "False Discovery Rate cutoff for SAM",
                         },
                     },
+                "required": ["column", "group1", "group2"],
                 },
-                "required": ["group", "subgroups"],
             },
         },
         {"type": "code_interpreter"},
@@ -574,10 +575,10 @@ def get_info(genes_list: list[str], organism_id: str) -> list[str]:
 
 def get_gene_function(gene_name: Union[str, dict], organism_id: str = None) -> str:
     """
-    Get the gene function and description by UniProt lookup of gene identifier/name.
+    Get the gene function and description by UniProt lookup of gene identifier / name.
 
     Args:
-        gene_name (Union[str, dict]): Gene identifier/name for UniProt lookup.
+        gene_name (Union[str, dict]): Gene identifier / name for UniProt lookup.
         organism_id (str): The UniProt organism ID to search in.
 
     Returns:
@@ -615,7 +616,7 @@ def turn_args_to_float(json_string: Union[str, bytes, bytearray]) -> dict:
 
 
 def get_gene_to_prot_id_mapping(gene_id: str) -> str:
-    """Get protein id from gene id. If gene id is not present, return gene id, as we might already have a gene id (LLM moment).
+    """Get protein id from gene id. If gene id is not present, return gene id, as we might already have a gene id.
     'VCL;HEL114' -> 'P18206;A0A024QZN4;V9HWK2;B3KXA2;Q5JQ13;B4DKC9;B4DTM7;A0A096LPE1'
     Args:
         gene_id (str): Gene id
@@ -624,11 +625,12 @@ def get_gene_to_prot_id_mapping(gene_id: str) -> str:
         str: Protein id or gene id if not present in the mapping.
     """
     import streamlit as st
-
-    print(st.session_state["gene_to_prot_id"])
-    if gene_id in st.session_state["gene_to_prot_id"]:
-        return st.session_state["gene_to_prot_id"][gene_id]
-    for gene, prot_id in st.session_state["gene_to_prot_id"].items():
+    session_state_copy = dict(copy.deepcopy(st.session_state))
+    if "gene_to_prot_id" not in session_state_copy:
+        session_state_copy["gene_to_prot_id"] = {}
+    if gene_id in session_state_copy["gene_to_prot_id"]:
+        return session_state_copy["gene_to_prot_id"][gene_id]
+    for gene, prot_id in session_state_copy["gene_to_prot_id"].items():
         if gene_id in gene.split(";"):
             return prot_id
     return gene_id