File size: 4,893 Bytes
4a6179c
 
02970af
c4bf66f
 
 
 
 
 
 
4a6179c
c4bf66f
4a6179c
 
1786f57
73fc0d7
02970af
73fc0d7
 
c4bf66f
 
1786f57
73fc0d7
1786f57
4a6179c
 
 
73fc0d7
c4bf66f
4a6179c
73fc0d7
 
 
 
4a6179c
 
 
 
21ff015
4a6179c
21ff015
73fc0d7
 
4a6179c
1786f57
73fc0d7
 
c4bf66f
 
73fc0d7
 
1786f57
73fc0d7
 
c4bf66f
73fc0d7
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
21ff015
73fc0d7
 
 
 
 
 
 
 
 
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
61
62
63
64
65
66
67
68
69
70
71
72
73
74
75
76
77
78
79
80
81
82
83
84
85
86
87
88
89
90
91
92
93
94
95
96
97
98
99
100
101
102
103
104
105
106
107
108
109
110
111
112
113
114
115
116
117
118
119
120
121
122
123
124
125
126
127
128
129
130
131
132
133
134
135
136
137
138
139
140
import asyncio
import re
from pathlib import Path

import streamlit as st
import pandas as pd
import plotly.express as px
from fpdf import FPDF
from streamlit_agraph import agraph

from mcp.orchestrator import orchestrate_search, answer_ai_question
from mcp.knowledge_graph import build_agraph
from mcp.graph_metrics import build_nx, get_top_hubs, get_density
from mcp.protocols import draft_protocol

# Streamlit configuration
st.set_page_config(page_title="MedGenesis AI", layout="wide")

# Initialize session state
if "res" not in st.session_state:
    st.session_state.res = None

# Header UI
st.title("🧬 MedGenesis AI")
llm = st.radio("LLM engine", ["openai", "gemini"], horizontal=True)
query = st.text_input("Enter biomedical question")

# PDF generation helper
def _make_pdf(papers):
    pdf = FPDF()
    pdf.add_page()
    pdf.set_font("Helvetica", size=12)
    pdf.cell(0, 10, "MedGenesis AI – Results", ln=True, align="C")
    pdf.ln(5)
    for i, p in enumerate(papers, 1):
        pdf.set_font("Helvetica", "B", 11)
        pdf.multi_cell(0, 7, f"{i}. {p.get('title','')}")
        pdf.set_font("Helvetica", size=9)
        body = f"""{p.get('authors','')}
{p.get('summary','')}
{p.get('link','')}"""
        pdf.multi_cell(0, 6, body)
        pdf.ln(3)
    return pdf.output(dest="S").encode("latin-1", errors="replace")

# Trigger search
if st.button("Run Search 🚀") and query.strip():
    with st.spinner("Gathering data…"):
        st.session_state.res = asyncio.run(orchestrate_search(query, llm))

# Retrieve results
res = st.session_state.res

# If no results yet, prompt user
if not res:
    st.info("Enter a question and press **Run Search 🚀** to begin.")
else:
    # Create tabs
    tabs = st.tabs(["Results", "Graph", "Clusters", "Variants", "Trials", "Metrics", "Visuals", "Protocols"])
    title_tab, graph_tab, clust_tab, var_tab, trial_tab, met_tab, vis_tab, proto_tab = tabs

    # Results Tab
    with title_tab:
        for i, p in enumerate(res["papers"], 1):
            st.markdown(f"**{i}. [{p['title']}]({p['link']})**")
            st.write(p["summary"])
        c1, c2 = st.columns(2)
        c1.download_button("CSV", pd.DataFrame(res["papers"]).to_csv(index=False),
                           "papers.csv", "text/csv")
        c2.download_button("PDF", _make_pdf(res["papers"]),
                           "papers.pdf", "application/pdf")
        st.subheader("AI summary")
        st.info(res["ai_summary"])

    # Graph Tab
    with graph_tab:
        nodes, edges, cfg = build_agraph(res["papers"], res["umls"], res.get("drug_safety", []), res.get("umls_relations", []))
        hl = st.text_input("Highlight node:", key="hl")
        if hl:
            pat = re.compile(re.escape(hl), re.I)
            for n in nodes:
                n.color = "#f1c40f" if pat.search(n.label) else n.color
        agraph(nodes, edges, cfg)

    # Clusters Tab
    with clust_tab:
        clusters = res.get("clusters", [])
        if clusters:
            df = pd.DataFrame({
                "title": [p['title'] for p in res['papers']],
                "cluster": clusters
            })
            st.write("### Paper Clusters")
            for c in sorted(set(clusters)):
                st.write(f"**Cluster {c}**")
                for t in df[df['cluster'] == c]['title']:
                    st.write(f"- {t}")
        else:
            st.info("No clusters to show.")

    # Variants Tab
    with var_tab:
        variants = res.get("variants", [])
        if variants:
            st.json(variants)
        else:
            st.warning("No variants found. Try a well-known gene like 'TP53'.")

    # Trials Tab
    with trial_tab:
        trials = res.get("clinical_trials", [])
        if trials:
            st.json(trials)
        else:
            st.warning("No trials found. Try a disease name or specific drug.")

    # Metrics Tab
    with met_tab:
        G = build_nx([n.__dict__ for n in nodes], [e.__dict__ for e in edges])
        st.metric("Density", f"{get_density(G):.3f}")
        st.markdown("**Top hubs**")
        for nid, sc in get_top_hubs(G):
            label = next((n.label for n in nodes if n.id == nid), nid)
            st.write(f"- {label}: {sc:.3f}")

    # Visuals Tab
    with vis_tab:
        years = [p.get("published", "")[:4] for p in res["papers"] if p.get("published")]
        if years:
            st.plotly_chart(px.histogram(pd.DataFrame({'year': years}), x='year', nbins=10, title="Publication Year"))

    # Protocols Tab
    with proto_tab:
        hyp = st.text_input("Enter hypothesis for protocol:", key="proto_q")
        if st.button("Draft Protocol") and hyp.strip():
            with st.spinner("Generating protocol…"):
                doc = asyncio.run(draft_protocol(hyp, context=res["ai_summary"], llm=llm))
            st.subheader("Experimental Protocol")
            st.write(doc)