Update pfp

.gitattributes

@@ -33,3 +33,4 @@ saved_model/**/* filter=lfs diff=lfs merge=lfs -text
 *.zip filter=lfs diff=lfs merge=lfs -text
 *.zst filter=lfs diff=lfs merge=lfs -text
 *tfevents* filter=lfs diff=lfs merge=lfs -text
+media/pfp/*.png filter=lfs diff=lfs merge=lfs -text

app.py

@@ -121,7 +121,8 @@ def check_password():
                 # Retrieve and store user name and team
                 st.session_state["name"] = user_db.loc[user_db.username == st.session_state["username"], "name"].values[0]
                 st.session_state["team"] = user_db.loc[user_db.username == st.session_state["username"], "team"].values[0]
-                st.session_state["profile_pic"] = user_db.loc[user_db.username == st.session_state["username"], "profile_pic"].values[0]
+                # st.session_state["profile_pic"] = user_db.loc[user_db.username == st.session_state["username"], "profile_pic"].values[0]
+                st.session_state["profile_pic"] = st.session_state["username"]
 
                 # Don't store the username or password
                 del st.session_state["password"]

menu.py

@@ -1,52 +1,21 @@
 # From https://docs.streamlit.io/develop/tutorials/multipage/st.page_link-nav
 import streamlit as st
-
+import project_config
 
 def authenticated_menu():
 
-    st.markdown("""
-        <style>
-        .circle-image {
-            width: 100px;
-            height: 100px;
-            border-radius: 50%;
-            overflow: hidden;
-            display: flex;
-            justify-content: center;
-            align-items: center;
-            border: 2px solid black;
-            margin: 0 auto 10px auto;
-        }
-        
-        .circle-image img {
-            width: 100%;
-            height: 100%;
-            object-fit: cover;
-        }
-                
-        .username {
-          font-size: 20px;
-          font-weight: bold;
-          text-align: center;
-          margin-top: 0px;
-        }
-        </style>
-        """, unsafe_allow_html=True)
-    
-    # Show the user's profile picture
-    st.sidebar.markdown(f'<div class="circle-image"><img src="{st.session_state.profile_pic}" /></div>', unsafe_allow_html=True)
-
-    # Show the user's name
-    # st.sidebar.markdown(f"Logged in as {st.session_state.name}.")
-    st.sidebar.markdown(f'<div class="username">{st.session_state.name}</div>', unsafe_allow_html=True)
+    # Insert profile picture
+    st.sidebar.image(str(project_config.MEDIA_DIR / 'pfp' / f"{st.session_state.profile_pic}.png"), use_column_width=True)
     st.sidebar.markdown("---")
 
     # Show a navigation menu for authenticated users
     # st.sidebar.page_link("app.py", label="Switch Accounts", icon="🔒")
     st.sidebar.page_link("pages/about.py", label="About", icon="📖")
     st.sidebar.page_link("pages/input.py", label="Input", icon="💡")
-    st.sidebar.page_link("pages/predict.py", label="Predict", icon="🔍")
-    st.sidebar.page_link("pages/validate.py", label="Validate", icon="✅")
+    st.sidebar.page_link("pages/predict.py", label="Predict", icon="🔍",
+                         disabled=("query" not in st.session_state))
+    st.sidebar.page_link("pages/validate.py", label="Validate", icon="✅",
+                         disabled=("query" not in st.session_state))
     # st.sidebar.page_link("pages/explore.py", label="Explore", icon="🔍")
     if st.session_state.role in ["admin"]:
         st.sidebar.page_link("pages/admin.py", label="Manage Users", icon="🔧")
@@ -54,7 +23,6 @@ def authenticated_menu():
     # Show the logout button
     st.sidebar.markdown("---")
     st.sidebar.button("Log Out", on_click=lambda: st.session_state.clear())
-    
 
 
 def unauthenticated_menu():

pages/about.py

@@ -18,3 +18,7 @@ st.markdown(f"Hello, {st.session_state.name}! Welcome to GRAVITY, a **GR**aph **
 
 # Subheader
 st.subheader("About GRAVITY", divider = "grey")
+
+st.markdown("""
+Knowledge graphs (KGs) are data structures that use network topology to represent relational information, including and especially in biology and medicine. Graph artificial intelligence (AI) models trained on these biomedical KGs can enable many important link prediction tasks, such as predicting disease progression, diagnosing genetic disorders, identifying therapeutic targets, and discovering new drugs. However, especially in biomedical settings, it is important for clinicians and scientists to evaluate whether KG-grounded AI models are safe and trustworthy, and whether the predictions of these models are biologically explainable. To address this challenge, we developed GRAVITY, an interactive user interface for graph-based explainable AI. GRAVITY enables human users to query and interpret KG-grounded AI models for biomedical link prediction tasks.
+""")

pages/input.py

@@ -10,6 +10,7 @@ from pathlib import Path
 
 # Custom and other imports
 import project_config
+from utils import load_kg
 
 # Redirect to app.py if not logged in, otherwise show the navigation menu
 menu_with_redirect()
@@ -25,8 +26,11 @@ st.subheader("Construct Query", divider = "red")
 # Checkbox to allow reverse edges
 allow_reverse_edges = st.checkbox("Reverse Edges", value = False)
 
+# Load knowledge graph
+kg_nodes = load_kg()
+
 with st.spinner('Loading knowledge graph...'):
-    kg_nodes = nodes = pd.read_csv(project_config.DATA_DIR / 'kg_nodes.csv', dtype = {'node_index': int}, low_memory = False)
+    # kg_nodes = nodes = pd.read_csv(project_config.DATA_DIR / 'kg_nodes.csv', dtype = {'node_index': int}, low_memory = False)
     node_types = pd.read_csv(project_config.DATA_DIR / 'kg_node_types.csv')
     edge_types = pd.read_csv(project_config.DATA_DIR / 'kg_edge_types.csv')
 
@@ -64,6 +68,10 @@ if st.button("Submit Query"):
     # st.write(st.session_state.query)
     st.write("Query submitted.")
 
+    # Switch to the Predict page
+    st.switch_page("pages/predict.py")
+
+
 st.subheader("Knowledge Graph", divider = "red")    
 display_data = kg_nodes[['node_id', 'node_type', 'node_name', 'node_source']].copy()
 display_data = display_data.rename(columns = {'node_id': 'ID', 'node_type': 'Type', 'node_name': 'Name', 'node_source': 'Database'})

pages/predict.py

@@ -14,7 +14,7 @@ from huggingface_hub import hf_hub_download
 
 # Custom and other imports
 import project_config
-from utils import capitalize_after_slash
+from utils import capitalize_after_slash, load_kg
 
 # Redirect to app.py if not logged in, otherwise show the navigation menu
 menu_with_redirect()
@@ -28,13 +28,12 @@ st.image(str(project_config.MEDIA_DIR / 'predict_header.svg'), use_column_width=
 st.subheader(f"{capitalize_after_slash(st.session_state.query['target_node_type'])} Search", divider = "blue")
 
 # Print current query
-st.markdown(f"**Query:** {st.session_state.query['source_node']} ➡️ {st.session_state.query['relation']} ➡️ {st.session_state.query['target_node_type']}")
+st.markdown(f"**Query:** {st.session_state.query['source_node'].replace('_', ' ')} ➡️ {st.session_state.query['relation'].replace('_', '-')} ➡️ {st.session_state.query['target_node_type'].replace('_', ' ')}")
 
-with st.spinner('Loading knowledge graph...'):
-    kg_nodes = nodes = pd.read_csv(project_config.DATA_DIR / 'kg_nodes.csv', dtype = {'node_index': int}, low_memory = False)
-
-# Get paths to embeddings, relation weights, and edge types
-with st.spinner('Downloading AI model...'):
+@st.cache_data(show_spinner = 'Downloading AI model...')
+def get_embeddings():
+    # Get paths to embeddings, relation weights, and edge types
+    # with st.spinner('Downloading AI model...'):
     embed_path = hf_hub_download(repo_id="ayushnoori/galaxy",
                                 filename="2024_03_29_04_12_52_epoch=3-step=54291_embeddings.pt",
                                 token=st.secrets["HF_TOKEN"])
@@ -44,13 +43,23 @@ with st.spinner('Downloading AI model...'):
     edge_types_path = hf_hub_download(repo_id="ayushnoori/galaxy",
                                         filename="2024_03_29_04_12_52_epoch=3-step=54291_edge_types.pt",
                                         token=st.secrets["HF_TOKEN"])
+    return embed_path, relation_weights_path, edge_types_path
 
-# Load embeddings, relation weights, and edge types
-with st.spinner('Loading AI model...'):
+@st.cache_data(show_spinner = 'Loading AI model...')
+def load_embeddings(embed_path, relation_weights_path, edge_types_path):
+    # Load embeddings, relation weights, and edge types
+    # with st.spinner('Loading AI model...'):
     embeddings = torch.load(embed_path)
     relation_weights = torch.load(relation_weights_path)
     edge_types = torch.load(edge_types_path)
 
+    return embeddings, relation_weights, edge_types
+
+# Load knowledge graph and embeddings
+kg_nodes = load_kg()
+embed_path, relation_weights_path, edge_types_path = get_embeddings()
+embeddings, relation_weights, edge_types = load_embeddings(embed_path, relation_weights_path, edge_types_path)
+
 # # Print source node type
 # st.write(f"Source Node Type: {st.session_state.query['source_node_type']}")
 
@@ -78,7 +87,7 @@ with st.spinner('Computing predictions...'):
     edge_type_index = [i for i, etype in enumerate(edge_types) if etype == (source_node_type, relation, target_node_type)][0]
 
     # Get target nodes indices
-    target_nodes = kg_nodes[kg_nodes.node_type == target_node_type]
+    target_nodes = kg_nodes[kg_nodes.node_type == target_node_type].copy()
     dst_indices = target_nodes.node_index.values
     src_indices = np.repeat(src_index, len(dst_indices))
 
@@ -126,7 +135,8 @@ with st.spinner('Computing predictions...'):
     display_data['Database'] = display_data.apply(lambda x: map_dbs[target_node_type](x['ID']), axis = 1)
 
     # Use multiselect to search for specific nodes
-    selected_nodes = st.multiselect('Search for specific nodes.', display_data.Name)
+    selected_nodes = st.multiselect(f"Search for specific {st.session_state.query['source_node_type'].replace('_', ' ')} nodes to determine their ranking.",
+                                    display_data.Name, placeholder = "Type to search...")
 
     # Filter nodes
     if len(selected_nodes) > 0:
@@ -152,3 +162,6 @@ with st.spinner('Computing predictions...'):
                                                                            display_text = display_database)})
     else:
         st.dataframe(display_data.iloc[:top_k], use_container_width = True)
+
+    # Save to session state
+    st.session_state.predictions = display_data

pages/validate.py

@@ -18,5 +18,9 @@ st.image(str(project_config.MEDIA_DIR / 'validate_header.svg'), use_column_width
 
 st.subheader("Validate Predictions", divider = "green")
 
+# Print current query
+st.markdown(f"**Query:** {st.session_state.query['source_node'].replace('_', ' ')} ➡️ {st.session_state.query['relation'].replace('_', '-')} ➡️ {st.session_state.query['target_node_type'].replace('_', ' ')}")
+
+
 # Coming soon
 st.write("Coming soon...")

utils.py

@@ -1,5 +1,13 @@
-import base64
 import streamlit as st
+import pandas as pd
+import project_config
+import base64
+
+@st.cache_data(show_spinner = 'Loading knowledge graph...')
+def load_kg():
+    # with st.spinner('Loading knowledge graph...'):
+    kg_nodes = pd.read_csv(project_config.DATA_DIR / 'kg_nodes.csv', dtype = {'node_index': int}, low_memory = False)
+    return kg_nodes
 
 def capitalize_after_slash(s):
     # Split the string by slashes first
@@ -52,4 +60,98 @@ def add_logo(png_file):
     st.markdown(
         logo_markup,
         unsafe_allow_html=True,
-    )
+    )
+
+
+# @st.cache_resource()
+# def generate_profile_pic():
+    
+#     st.markdown("""
+#     <style>
+#     .circle-image {
+#         width: 100px;
+#         height: 100px;
+#         border-radius: 50%;
+#         overflow: hidden;
+#         display: flex;
+#         justify-content: center;
+#         align-items: center;
+#         border: 2px solid black;
+#         margin: 0 auto 10px auto;
+#     }
+    
+#     .circle-image img {
+#         width: 100%;
+#         max-width: 100px;
+#         height: 100%;
+#         object-fit: cover;
+#     }
+            
+#     .username {
+#         font-size: 20px;
+#         font-weight: bold;
+#         text-align: center;
+#         margin-top: 0px;
+#     }
+#     </style>
+#     """, unsafe_allow_html=True)
+    
+#     # Show the user's profile picture
+#     st.sidebar.html(f'<div class="circle-image"><img src="{st.session_state.profile_pic}" /></div>')
+
+#     # Show the user's name
+#     st.sidebar.html(f'<div class="username">{st.session_state.name}</div>')
+
+#     return None
+
+# # Load image using PIL
+# from PIL import Image, ImageDraw, ImageOps
+# from io import BytesIO
+# import requests
+
+# def PIL_profile_pic():
+
+#     #  Load user profile picture
+#     profile_pic = st.session_state.profile_pic
+#     response = requests.get(profile_pic)
+#     img = Image.open(BytesIO(response.content))
+
+#     # Create a circular mask
+#     min_dimension = min(img.size)
+#     mask = Image.new('L', (min_dimension, min_dimension), 0)
+#     draw = ImageDraw.Draw(mask)
+#     draw.ellipse((0, 0, min_dimension, min_dimension), fill=255)
+
+#     # Crop the image to a square of the smallest dimension
+#     left = (img.width - min_dimension) // 2
+#     top = (img.height - min_dimension) // 2
+#     right = (img.width + min_dimension) // 2
+#     bottom = (img.height + min_dimension) // 2
+#     img_cropped = img.crop((left, top, right, bottom))
+
+#     # Apply the circular mask to the cropped image
+#     img_circular = ImageOps.fit(img_cropped, (min_dimension, min_dimension))
+#     img_circular.putalpha(mask)
+
+#     st.markdown(
+#     """
+#     <style>
+#         [data-testid=stSidebar] [data-testid=stImage]{
+#             text-align: center;
+#             display: block;
+#             margin-left: auto;
+#             margin-right: auto;
+#             width: 100%;
+#         }
+#     </style>
+#     """, unsafe_allow_html=True
+#     )
+
+#     # Display the image
+#     st.sidebar.image(img_circular, width=200)
+#     st.sidebar.subheader(f"{st.session_state.name}")
+
+
+
+    # Generate the user's profile picture
+    # generate_profile_pic()