Sync from GitHub (preserve manual model files)

StreamlitApp/StreamlitApp.py

@@ -86,14 +86,10 @@ if "predict_input_widget" not in st.session_state:
     st.session_state.predict_input_widget = ""
 if "analyze_input" not in st.session_state:
     st.session_state.analyze_input = ""             # last analyze input
-if "analyze_input_widget" not in st.session_state:
-    st.session_state.analyze_input_widget = st.session_state.get("analyze_input", "")
 if "analyze_output" not in st.session_state:
     st.session_state.analyze_output = None         # (label, conf_display, comp, props, analysis)
 if "optimize_input" not in st.session_state:
     st.session_state.optimize_input = ""           # last optimize input
-if "optimize_input_widget" not in st.session_state:
-    st.session_state.optimize_input_widget = st.session_state.get("optimize_input", "")
 if "optimize_output" not in st.session_state:
     st.session_state.optimize_output = None       # (orig_seq, orig_conf, improved_seq, improved_conf, history)
 if "visualize_sequences" not in st.session_state:
@@ -110,8 +106,8 @@ if st.sidebar.button("Clear All Fields"):
     # clear only our known keys
     keys = ["predictions", "predict_ran",
             "predict_input_widget",
-            "analyze_input", "analyze_input_widget", "analyze_output",
-            "optimize_input", "optimize_input_widget", "optimize_output",
+            "analyze_input", "analyze_output",
+            "optimize_input", "optimize_output",
             "visualize_sequences", "visualize_df"]
     for k in keys:
         if k in st.session_state:
@@ -135,7 +131,7 @@ if page == "Predict":
     preset_cols = st.columns(2)
     with preset_cols[0]:
         if st.button("Use strong AMP example"):
-            st.session_state.predict_input_widget = "KWKLFKKIGAVLKVL"
+            st.session_state.predict_input_widget = "ILPWKWPWWPWRR"
             st.rerun()
     with preset_cols[1]:
         if st.button("Use weak sequence example"):
@@ -197,7 +193,7 @@ if page == "Predict":
 
     # If user hasn't just run predictions, show the last saved results (if any)
     if st.session_state.predictions and not (run and st.session_state.predict_ran is False):
-        st.subheader("Predictions (last run)")
+        st.write("**Predictions (last run)**")
 
         top_candidate = choose_top_candidate(st.session_state.predictions)
         if top_candidate:
@@ -205,34 +201,23 @@ if page == "Predict":
                 st.markdown(
                     "<div style='border:1px solid #e6e6e6; border-radius:0.6rem; padding:0.8rem; background:#fafafa;'>"
                 , unsafe_allow_html=True)
-                st.markdown("### Top Candidate")
-                st.code(top_candidate["Sequence"], language="text")
+                st.write("**Top Candidate**")
+                seq = top_candidate.get("Sequence", "")
+                cc = st.columns([9, 1])
+                with cc[0]:
+                    st.code(seq, language="text")
+                with cc[1]:
+                    if st.button("Copy", key="copy_top_candidate"):
+                        _try_copy_to_clipboard(seq)
+                        toast_fn = getattr(st, "toast", None)
+                        if toast_fn is not None:
+                            toast_fn("Copied to clipboard")
+                        else:
+                            st.success("Copied to clipboard")
                 st.write(f"Confidence: **{format_conf_percent(top_candidate['predicted_confidence'], digits=1)}**")
                 st.write(f"Reason: {top_candidate['Reason']}")
                 st.markdown("</div>", unsafe_allow_html=True)
 
-        # Copy-to-clipboard + compact row view (buttons placed next to each sequence).
-        st.markdown("### Results with Copy")
-        for idx, row in enumerate(st.session_state.predictions):
-            seq = row.get("Sequence", "")
-            label = row.get("Prediction", "")
-            conf = row.get("Confidence", 0.0)
-            conf_display = round(float(conf) * 100, 1) if label == "AMP" else round((1 - float(conf)) * 100, 1)
-            cols = st.columns([7, 2, 1])
-            with cols[0]:
-                st.code(seq, language="text")
-            with cols[1]:
-                st.write(f"**{label}**")
-                st.caption(f"{conf_display}% confidence")
-            with cols[2]:
-                if st.button("Copy", key=f"copy_pred_{idx}"):
-                    _try_copy_to_clipboard(seq)
-                    toast_fn = getattr(st, "toast", None)
-                    if toast_fn is not None:
-                        toast_fn("Copied to clipboard")
-                    else:
-                        st.success("Copied to clipboard")
-
         # Keep the original dataframe for full overview/download compatibility.
         st.dataframe(pd.DataFrame(st.session_state.predictions), use_container_width=True)
         csv = pd.DataFrame(st.session_state.predictions).to_csv(index=False)
@@ -242,22 +227,11 @@ if page == "Predict":
 elif page == "Analyze":
     st.header("Sequence Analysis")
 
-    preset_cols = st.columns(2)
-    with preset_cols[0]:
-        if st.button("Use strong AMP example"):
-            st.session_state.analyze_input_widget = "KWKLFKKIGAVLKVL"
-            st.rerun()
-    with preset_cols[1]:
-        if st.button("Use weak sequence example"):
-            st.session_state.analyze_input_widget = "GAGAGAGAGAGA"
-            st.rerun()
-
     # show the last saved analyze output if user navigated back
     last_seq = st.session_state.analyze_input
     seq = st.text_input(
         "Enter a peptide sequence to analyze:",
-        value=st.session_state.get("analyze_input_widget", last_seq),
-        key="analyze_input_widget",
+        value=last_seq,
     )
 
     warn = sequence_length_warning(seq)
@@ -349,21 +323,40 @@ elif page == "Analyze":
             "Net Charge": "Favorable" if charge > 0 else "Neutral" if charge == 0 else "Unfavorable",
             "Molecular Weight": "Acceptable" if 500 <= mw <= 5000 else "Extreme"
         }
-        st.table(pd.DataFrame([
-            {"Property": "Length", "Value": length, "Favorability": favorability["Length"]},
-            {"Property": "Hydrophobic Fraction", "Value": hydro, "Favorability": favorability["Hydrophobic Fraction"]},
-            {"Property": "Net Charge", "Value": charge, "Favorability": favorability["Net Charge"]},
-            {"Property": "Molecular Weight", "Value": mw, "Favorability": favorability["Molecular Weight"]}
-        ]))
+        def _info_icon(tooltip_text: str) -> str:
+            safe = _html.escape(tooltip_text, quote=True)
+            return (
+                "<span "
+                f"title='{safe}' "
+                "style=\"display:inline-flex; align-items:center; justify-content:center; "
+                "margin-left:6px; width:16px; height:16px; border-radius:50%; "
+                "background:#f2f2f2; border:1px solid #d9d9d9; color:#333; "
+                "font-size:12px; font-weight:700; cursor:help;\">(i)</span>"
+            )
 
-        _tooltip_label("Hydrophobic Fraction", "Fraction of residues that prefer non-aqueous environments")
-        _tooltip_label("Net Charge", "Positive charge helps peptides bind bacterial membranes")
+        # Render the favorability table with working inline tooltips.
+        hydro_label = f"Hydrophobic Fraction{_info_icon('Fraction of residues that prefer non-aqueous environments')}"
+        charge_label = f"Net Charge{_info_icon('Positive charge helps peptides bind bacterial membranes')}"
+        table_html = (
+            "<table style='width:100%; border-collapse:collapse;'>"
+            "<thead>"
+            "<tr>"
+            "<th style='text-align:left; padding:8px; border-bottom:1px solid #e6e6e6;'>Property</th>"
+            "<th style='text-align:right; padding:8px; border-bottom:1px solid #e6e6e6;'>Value</th>"
+            "<th style='text-align:left; padding:8px; border-bottom:1px solid #e6e6e6;'>Favorability</th>"
+            "</tr>"
+            "</thead>"
+            "<tbody>"
+            f"<tr><td style='padding:8px;'>{_html.escape('Length')}</td><td style='padding:8px; text-align:right;'>{_html.escape(str(length))}</td><td style='padding:8px;'>{_html.escape(favorability['Length'])}</td></tr>"
+            f"<tr><td style='padding:8px;'>{hydro_label}</td><td style='padding:8px; text-align:right;'>{_html.escape(str(hydro))}</td><td style='padding:8px;'>{_html.escape(favorability['Hydrophobic Fraction'])}</td></tr>"
+            f"<tr><td style='padding:8px;'>{charge_label}</td><td style='padding:8px; text-align:right;'>{_html.escape(str(charge))}</td><td style='padding:8px;'>{_html.escape(favorability['Net Charge'])}</td></tr>"
+            f"<tr><td style='padding:8px;'>{_html.escape('Molecular Weight')}</td><td style='padding:8px; text-align:right;'>{_html.escape(str(mw))}</td><td style='padding:8px;'>{_html.escape(favorability['Molecular Weight'])}</td></tr>"
+            "</tbody>"
+            "</table>"
+        )
+        st.markdown(table_html, unsafe_allow_html=True)
 
         st.subheader("Property Radar Chart")
-        _tooltip_label(
-            "Radar",
-            "Radar chart compares length, hydrophobicity, net charge, and molecular weight against ideal AMP ranges.",
-        )
         categories = ["Length", "Hydrophobic Fraction", "Net Charge", "Molecular Weight"]
         values = [min(length / 50, 1), min(hydro, 1), 1 if charge > 0 else 0, min(mw / 5000, 1)]
         values += values[:1]
@@ -390,7 +383,6 @@ elif page == "Analyze":
             st.write(f"- {line}")
 
         # Export analysis report
-        st.markdown("---")
         st.subheader("Export Analysis Report")
         export_format = st.radio("Format", ["CSV", "TXT"], horizontal=True)
 
@@ -437,23 +429,10 @@ elif page == "Analyze":
 elif page == "Optimize":
     st.header("AMP Sequence Optimizer")
 
-    preset_cols = st.columns(2)
-    with preset_cols[0]:
-        if st.button("Use strong AMP example"):
-            st.session_state.optimize_input_widget = "KWKLFKKIGAVLKVL"
-            st.session_state.optimize_input = "KWKLFKKIGAVLKVL"
-            st.rerun()
-    with preset_cols[1]:
-        if st.button("Use weak sequence example"):
-            st.session_state.optimize_input_widget = "GAGAGAGAGAGA"
-            st.session_state.optimize_input = "GAGAGAGAGAGA"
-            st.rerun()
-
-    # Single entry point: text input retained across navigation (widget uses a dedicated key)
+    # Single entry point: text input retained across navigation
     seq = st.text_input(
         "Enter a peptide sequence to optimize:",
-        value=st.session_state.get("optimize_input_widget", ""),
-        key="optimize_input_widget",
+        value=st.session_state.get("optimize_input", ""),
     )
 
     # Run optimization when user changes input and clicks button
@@ -626,6 +605,13 @@ elif page == "Visualize":
 elif page == "About":
     st.header("About the Project")
     st.markdown("""
-**Problem:** Antimicrobial resistance is a global health threat. Traditional peptide screening is slow and costly.  
-**Solution:** This tool predicts antimicrobial activity directly from sequence using deep learning, speeding up AMP discovery.
+PeptideAI is a lightweight Streamlit app for exploring antimicrobial peptide (AMP) sequences.
+
+It uses a trained neural network to estimate whether a peptide is likely to be antimicrobial, then helps you interpret and improve candidates:
+- **Predict**: run batch predictions and highlights the best candidate using simple profile heuristics.
+- **Analyze**: show amino-acid composition, physicochemical properties, and an AMP-range radar chart.
+- **Optimize**: iteratively propose residue mutations and summarize how confidence/charge/hydrophobic balance changes.
+- **Visualize**: embed sequences and inspect clusters with t-SNE.
+
+Note: predictions are model-based heuristics and are not a substitute for wet-lab validation.
 """)