Upload app (1).py

app (1).py

@@ -0,0 +1,444 @@
+import gradio as gr
+import pandas as pd
+import numpy as np
+import json
+import re
+import matplotlib
+matplotlib.use("Agg")
+import matplotlib.pyplot as plt
+from io import BytesIO
+from PIL import Image
+
+BG   = "#0f172a"
+CARD = "#1e293b"
+ACC  = "#f97316"
+ACC2 = "#38bdf8"
+TXT  = "#f1f5f9"
+
+# ── TAB 1 — BRCA2 miRNA ──────────────────────────────────────────────────────
+MIRNA_DB = {
+    "BRCA2": [
+        {"miRNA":"hsa-miR-148a-3p","log2FC":-0.70,"padj":0.013,"targets":"DNMT1, AKT2","pathway":"Epigenetic reprogramming"},
+        {"miRNA":"hsa-miR-30e-5p", "log2FC":-0.49,"padj":0.032,"targets":"MYC, KRAS","pathway":"Oncogene suppression"},
+        {"miRNA":"hsa-miR-551b-3p","log2FC":-0.59,"padj":0.048,"targets":"SMAD4, CDK6","pathway":"TGF-beta / CDK4/6"},
+        {"miRNA":"hsa-miR-22-3p",  "log2FC":-0.43,"padj":0.041,"targets":"HIF1A, PTEN","pathway":"Hypoxia / PI3K"},
+        {"miRNA":"hsa-miR-200c-3p","log2FC":-0.38,"padj":0.044,"targets":"ZEB1, ZEB2","pathway":"EMT suppression"},
+    ],
+    "BRCA1": [
+        {"miRNA":"hsa-miR-155-5p", "log2FC":-0.81,"padj":0.008,"targets":"SHIP1, SOCS1","pathway":"Immune evasion"},
+        {"miRNA":"hsa-miR-146a-5p","log2FC":-0.65,"padj":0.019,"targets":"TRAF6, IRAK1","pathway":"NF-kB signalling"},
+        {"miRNA":"hsa-miR-21-5p",  "log2FC":-0.55,"padj":0.027,"targets":"PTEN, PDCD4","pathway":"Apoptosis"},
+        {"miRNA":"hsa-miR-17-5p",  "log2FC":-0.47,"padj":0.036,"targets":"RB1, E2F1","pathway":"Cell cycle"},
+        {"miRNA":"hsa-miR-34a-5p", "log2FC":-0.41,"padj":0.049,"targets":"BCL2, CDK6","pathway":"p53 axis"},
+    ],
+    "TP53": [
+        {"miRNA":"hsa-miR-34a-5p", "log2FC":-1.10,"padj":0.001,"targets":"BCL2, CDK6","pathway":"p53-miR-34 axis"},
+        {"miRNA":"hsa-miR-192-5p", "log2FC":-0.90,"padj":0.005,"targets":"MDM2, DHFR","pathway":"p53 feedback"},
+        {"miRNA":"hsa-miR-145-5p", "log2FC":-0.75,"padj":0.012,"targets":"MYC, EGFR","pathway":"Growth suppression"},
+        {"miRNA":"hsa-miR-107",    "log2FC":-0.62,"padj":0.023,"targets":"CDK6, HIF1B","pathway":"Hypoxia / cell cycle"},
+        {"miRNA":"hsa-miR-215-5p", "log2FC":-0.51,"padj":0.038,"targets":"DTL, DHFR","pathway":"DNA damage response"},
+    ],
+}
+
+def predict_mirna(gene):
+    return pd.DataFrame(MIRNA_DB.get(gene, []))
+
+# ── TAB 2 — TP53 siRNA ───────────────────────────────────────────────────────
+SIRNA_DB = {
+    "LUAD": [
+        {"Gene":"SPC24","dCERES":-0.175,"log2FC":1.13,"Drug_status":"Novel","siRNA":"GCAGCUGAAGAAACUGAAU"},
+        {"Gene":"BUB1B","dCERES":-0.119,"log2FC":1.12,"Drug_status":"Novel","siRNA":"CCAAAGAGCUGAAGAACAU"},
+        {"Gene":"CDC45","dCERES":-0.144,"log2FC":1.26,"Drug_status":"Novel","siRNA":"GCAUCAAGAUGAAGGAGAU"},
+        {"Gene":"PLK1", "dCERES":-0.239,"log2FC":1.03,"Drug_status":"Clinical","siRNA":"GACGCUCAAGAUGCAGAUU"},
+        {"Gene":"CDK1", "dCERES":-0.201,"log2FC":1.00,"Drug_status":"Clinical","siRNA":"GCAGAAGCACUGAAGAUUU"},
+    ],
+    "BRCA": [
+        {"Gene":"AURKA","dCERES":-0.165,"log2FC":1.20,"Drug_status":"Clinical","siRNA":"GCACUGAAGAUGCAGAAUU"},
+        {"Gene":"AURKB","dCERES":-0.140,"log2FC":1.15,"Drug_status":"Clinical","siRNA":"CCUGAAGACGCUCAAGGUU"},
+        {"Gene":"CENPW","dCERES":-0.125,"log2FC":0.95,"Drug_status":"Novel","siRNA":"GCAGAAGCACUGAAGAUUU"},
+        {"Gene":"RFC2", "dCERES":-0.136,"log2FC":0.50,"Drug_status":"Novel","siRNA":"GCAAGAUGCAGAAGCACUU"},
+        {"Gene":"TYMS", "dCERES":-0.131,"log2FC":0.72,"Drug_status":"Approved","siRNA":"GGACGCUCAAGAUGCAGAU"},
+    ],
+    "COAD": [
+        {"Gene":"KRAS", "dCERES":-0.210,"log2FC":0.80,"Drug_status":"Clinical","siRNA":"GCUGGAGCUGGUGGUAGUU"},
+        {"Gene":"WEE1", "dCERES":-0.180,"log2FC":1.05,"Drug_status":"Clinical","siRNA":"GCAGCUGAAGAAACUGAAU"},
+        {"Gene":"CHEK1","dCERES":-0.155,"log2FC":0.90,"Drug_status":"Clinical","siRNA":"CCAAAGAGCUGAAGAACAU"},
+        {"Gene":"RFC2", "dCERES":-0.130,"log2FC":0.55,"Drug_status":"Novel","siRNA":"GCAUCAAGAUGAAGGAGAU"},
+        {"Gene":"PKMYT1","dCERES":-0.122,"log2FC":1.07,"Drug_status":"Clinical","siRNA":"GACGCUCAAGAUGCAGAUU"},
+    ],
+}
+
+def predict_sirna(cancer):
+    return pd.DataFrame(SIRNA_DB.get(cancer, []))
+
+# ── TAB 3 — lncRNA-TREM2 ─────────────────────────────────────────────────────
+CERNA = [
+    {"lncRNA":"CYTOR","miRNA":"hsa-miR-138-5p","target":"AKT1","pathway":"TREM2 core signaling"},
+    {"lncRNA":"CYTOR","miRNA":"hsa-miR-138-5p","target":"NFKB1","pathway":"Neuroinflammation"},
+    {"lncRNA":"GAS5","miRNA":"hsa-miR-21-5p","target":"PTEN","pathway":"Neuroinflammation"},
+    {"lncRNA":"GAS5","miRNA":"hsa-miR-222-3p","target":"IL1B","pathway":"Neuroinflammation"},
+    {"lncRNA":"HOTAIRM1","miRNA":"hsa-miR-9-5p","target":"TREM2","pathway":"Direct TREM2 regulation"},
+]
+ASO = [
+    {"lncRNA":"GAS5","position":119,"accessibility":0.653,"GC_pct":50,"Tm":47.2,"priority":"HIGH"},
+    {"lncRNA":"CYTOR","position":507,"accessibility":0.653,"GC_pct":50,"Tm":46.8,"priority":"HIGH"},
+    {"lncRNA":"HOTAIRM1","position":234,"accessibility":0.621,"GC_pct":44,"Tm":44.1,"priority":"MEDIUM"},
+    {"lncRNA":"LINC00847","position":89,"accessibility":0.598,"GC_pct":56,"Tm":48.3,"priority":"MEDIUM"},
+    {"lncRNA":"ZFAS1","position":312,"accessibility":0.571,"GC_pct":48,"Tm":45.5,"priority":"MEDIUM"},
+]
+
+def get_lncrna():
+    return pd.DataFrame(CERNA), pd.DataFrame(ASO)
+
+# ── TAB 4 — FGFR3 Drug ───────────────────────────────────────────────────────
+FGFR3 = {
+    "P1 (hairpin loop)": [
+        {"Compound":"CHEMBL1575701","RNA_score":0.809,"Toxicity":0.01,"Final_score":0.793},
+        {"Compound":"CHEMBL15727",  "RNA_score":0.805,"Toxicity":0.00,"Final_score":0.789},
+        {"Compound":"Thioguanine",  "RNA_score":0.888,"Toxicity":32.5,"Final_score":0.742},
+        {"Compound":"Deazaguanine", "RNA_score":0.888,"Toxicity":35.0,"Final_score":0.735},
+        {"Compound":"CHEMBL441",    "RNA_score":0.775,"Toxicity":5.2, "Final_score":0.721},
+    ],
+    "P10 (G-quadruplex)": [
+        {"Compound":"CHEMBL15727",  "RNA_score":0.805,"Toxicity":0.00,"Final_score":0.789},
+        {"Compound":"CHEMBL5411515","RNA_score":0.945,"Toxicity":37.1,"Final_score":0.761},
+        {"Compound":"CHEMBL90",     "RNA_score":0.760,"Toxicity":2.1, "Final_score":0.745},
+        {"Compound":"CHEMBL102",    "RNA_score":0.748,"Toxicity":8.4, "Final_score":0.712},
+        {"Compound":"Berberine",    "RNA_score":0.735,"Toxicity":3.2, "Final_score":0.708},
+    ],
+}
+
+def predict_drug(pocket):
+    df = pd.DataFrame(FGFR3.get(pocket, []))
+    fig, ax = plt.subplots(figsize=(6, 4), facecolor=CARD)
+    ax.set_facecolor(CARD)
+    ax.barh(df["Compound"], df["Final_score"], color=ACC)
+    ax.set_xlabel("Final Score", color=TXT)
+    ax.tick_params(colors=TXT)
+    for sp in ax.spines.values(): sp.set_edgecolor("#334155")
+    ax.set_title(f"Top compounds — {pocket}", color=TXT, fontsize=10)
+    plt.tight_layout()
+    buf = BytesIO(); plt.savefig(buf, format="png", dpi=120, facecolor=CARD); plt.close()
+    buf.seek(0)
+    return df, Image.open(buf)
+
+# ── TAB 5 — OpenVariant ──────────────────────────────────────────────────────
+VARIANT_DB = {
+    "BRCA1:p.R1699Q": {"score":0.03, "cls":"Benign",     "conf":"High"},
+    "BRCA1:p.R1699W": {"score":0.97, "cls":"Pathogenic", "conf":"High"},
+    "BRCA2:p.D2723A": {"score":0.999,"cls":"Pathogenic", "conf":"High"},
+    "TP53:p.R248W":   {"score":0.998,"cls":"Pathogenic", "conf":"High"},
+    "TP53:p.R248Q":   {"score":0.995,"cls":"Pathogenic", "conf":"High"},
+    "EGFR:p.L858R":   {"score":0.96, "cls":"Pathogenic", "conf":"High"},
+    "ALK:p.F1174L":   {"score":0.94, "cls":"Pathogenic", "conf":"High"},
+}
+PLAIN = {
+    "Pathogenic":       "This variant is likely to cause disease. Clinical follow-up is strongly recommended.",
+    "Likely Pathogenic":"This variant is probably harmful. Discuss with your doctor.",
+    "Benign":           "This variant is likely harmless. Common in the general population.",
+    "Likely Benign":    "This variant is probably harmless. No strong reason for concern.",
+}
+
+def predict_variant(hgvs, sift, polyphen, gnomad):
+    hgvs = hgvs.strip()
+    if hgvs in VARIANT_DB:
+        r = VARIANT_DB[hgvs]
+        cls, conf, score = r["cls"], r["conf"], r["score"]
+    else:
+        score = 0.0
+        if sift < 0.05: score += 0.4
+        if polyphen > 0.85: score += 0.35
+        if gnomad < 0.0001: score += 0.25
+        score = round(score, 3)
+        cls  = "Pathogenic" if score > 0.6 else ("Likely Pathogenic" if score > 0.4 else "Benign")
+        conf = "High" if (sift < 0.01 or sift > 0.9) else "Moderate"
+    colour = "#ef4444" if "Pathogenic" in cls else "#22c55e"
+    icon   = "WARNING" if "Pathogenic" in cls else "OK"
+    bar_w  = int(score * 100)
+    explanation = PLAIN.get(cls, "")
+    html = (
+        f"<div style='background:{CARD};padding:16px;border-radius:8px;"
+        f"font-family:sans-serif;color:{TXT}'>"
+        f"<h3 style='color:{colour}'>[{icon}] {cls}</h3>"
+        f"<p>Score: <b>{score:.3f}</b> &nbsp;|&nbsp; Confidence: <b>{conf}</b></p>"
+        f"<div style='background:#334155;border-radius:4px;height:16px;width:100%'>"
+        f"<div style='background:{colour};height:16px;border-radius:4px;width:{bar_w}%'></div>"
+        f"</div>"
+        f"<p style='margin-top:12px'>{explanation}</p>"
+        f"<p style='font-size:11px;color:#64748b;margin-top:8px'>"
+        f"Research tool only. Not for clinical use.</p></div>"
+    )
+    return html
+
+# ── TAB 6 — LNP Corona ────────────���──────────────────────────────────────────
+def predict_corona(size, zeta, peg, lipid):
+    score = 0
+    if lipid == "Ionizable": score += 2
+    elif lipid == "Cationic": score += 1
+    if abs(zeta) < 10: score += 1
+    if peg > 1.5: score += 2
+    if size < 100: score += 1
+    proteins  = ["ApoE","Albumin","Fibrinogen","Vitronectin","ApoA-I"]
+    dominant  = proteins[min(score, 4)]
+    efficacy  = "High" if score >= 4 else "Medium" if score >= 2 else "Low"
+    return f"**Dominant corona protein:** {dominant}\n\n**Predicted efficacy class:** {efficacy}\n\n**Composite score:** {score}/6"
+
+# ── TAB 7 — Liquid Biopsy ────────────────────────────────────────────────────
+BM_W = {"CTHRC1":0.18,"FHL2":0.15,"LDHA":0.14,"P4HA1":0.13,"SERPINH1":0.12,
+        "ABCA8":-0.11,"CA4":-0.10,"CKB":-0.09,"NNMT":0.08,"CACNA2D2":-0.07}
+
+def predict_cancer(c1,c2,c3,c4,c5,c6,c7,c8,c9,c10):
+    vals    = [c1,c2,c3,c4,c5,c6,c7,c8,c9,c10]
+    names   = list(BM_W.keys())
+    weights = list(BM_W.values())
+    raw     = sum(v*w for v,w in zip(vals,weights))
+    prob    = 1 / (1 + np.exp(-raw * 2))
+    label   = "CANCER" if prob > 0.5 else "HEALTHY"
+    colour  = "#ef4444" if prob > 0.5 else "#22c55e"
+    contribs = [v*w for v,w in zip(vals,weights)]
+    cols     = [ACC if c > 0 else ACC2 for c in contribs]
+    fig, ax  = plt.subplots(figsize=(6, 3.5), facecolor=CARD)
+    ax.set_facecolor(CARD)
+    ax.barh(names, contribs, color=cols)
+    ax.axvline(0, color=TXT, linewidth=0.8)
+    ax.set_xlabel("Contribution to cancer score", color=TXT)
+    ax.tick_params(colors=TXT, labelsize=8)
+    for sp in ax.spines.values(): sp.set_edgecolor("#334155")
+    ax.set_title("Protein contributions", color=TXT, fontsize=10)
+    plt.tight_layout()
+    buf = BytesIO(); plt.savefig(buf, format="png", dpi=120, facecolor=CARD); plt.close()
+    buf.seek(0)
+    return (f"<div style='background:{CARD};padding:12px;border-radius:8px;"
+            f"color:{colour};font-size:20px;font-family:sans-serif'><b>{label}</b><br>"
+            f"<span style='color:{TXT};font-size:14px'>Probability: {prob:.2f}</span></div>"), \
+           Image.open(buf)
+
+# ── TAB 8 — Flow Corona ──────────────────────────────────────────────────────
+def predict_flow(size, zeta, peg, charge, flow_rate):
+    csi = (flow_rate/40)*0.6 + (peg/5)*0.2 + (1 if charge=="Cationic" else 0)*0.2
+    csi = round(min(csi, 1.0), 3)
+    stability = "High remodeling" if csi > 0.6 else "Medium" if csi > 0.3 else "Stable"
+    t = np.linspace(0, 60, 200)
+    kf = 0.03 * (1 + flow_rate/40)
+    ks = 0.005 * (1 + flow_rate/80)
+    fig, ax = plt.subplots(figsize=(6, 3.5), facecolor=CARD)
+    ax.set_facecolor(CARD)
+    ax.plot(t, 60*np.exp(-0.03*t)+20, color="#60a5fa", ls="--", label="Albumin (static)")
+    ax.plot(t, 60*np.exp(-kf*t)+10,   color="#60a5fa",           label="Albumin (flow)")
+    ax.plot(t, 5*(1-np.exp(-0.005*t))+5, color=ACC, ls="--",    label="ApoE (static)")
+    ax.plot(t, 5*(1-np.exp(-ks*t))+10,   color=ACC,             label="ApoE (flow)")
+    ax.set_xlabel("Time (min)", color=TXT)
+    ax.set_ylabel("% Corona",   color=TXT)
+    ax.tick_params(colors=TXT)
+    ax.legend(fontsize=7, labelcolor=TXT, facecolor=CARD)
+    for sp in ax.spines.values(): sp.set_edgecolor("#334155")
+    ax.set_title("Vroman Effect: Albumin to ApoE exchange", color=TXT, fontsize=9)
+    plt.tight_layout()
+    buf = BytesIO(); plt.savefig(buf, format="png", dpi=120, facecolor=CARD); plt.close()
+    buf.seek(0)
+    return f"**Corona Shift Index: {csi}** — {stability}", Image.open(buf)
+
+# ── TAB 9 — LNP Brain ────────────────────────────────────────────────────────
+def predict_bbb(smiles, pka, zeta):
+    logp     = smiles.count("C")*0.3 - smiles.count("O")*0.5 + 1.5
+    apoe_pct = max(0, min(40, (7.0 - pka)*8 + abs(zeta)*0.5 + logp*0.8))
+    bbb_prob = min(0.95, apoe_pct / 30)
+    tier     = "HIGH (>20%)" if apoe_pct > 20 else "MEDIUM (10-20%)" if apoe_pct > 10 else "LOW (<10%)"
+    cats     = ["ApoE%","BBB","logP","pKa","Zeta"]
+    vals     = [apoe_pct/40, bbb_prob, min(logp/5,1), (7-abs(pka-6.5))/7, (10-abs(zeta))/10]
+    angles   = np.linspace(0, 2*np.pi, len(cats), endpoint=False).tolist()
+    v2       = vals + [vals[0]]
+    a2       = angles + [angles[0]]
+    fig, ax  = plt.subplots(figsize=(5, 4), subplot_kw={"polar": True}, facecolor=CARD)
+    ax.set_facecolor(CARD)
+    ax.plot(a2, v2, color=ACC, linewidth=2)
+    ax.fill(a2, v2, color=ACC, alpha=0.2)
+    ax.set_xticks(angles); ax.set_xticklabels(cats, color=TXT, fontsize=8)
+    ax.tick_params(colors=TXT)
+    plt.tight_layout()
+    buf = BytesIO(); plt.savefig(buf, format="png", dpi=120, facecolor=CARD); plt.close()
+    buf.seek(0)
+    return f"**Predicted ApoE:** {apoe_pct:.1f}% — {tier}\n\n**BBB Probability:** {bbb_prob:.2f}", \
+           Image.open(buf)
+
+# ── TAB 10 — AutoCorona NLP ──────────────────────────────────────────────────
+PROTEINS = ["albumin","apolipoprotein","fibrinogen","vitronectin","clusterin",
+            "igm","iga","igg","complement","transferrin","alpha-2-macroglobulin"]
+
+def extract_corona(text):
+    out = {"nanoparticle_composition":"","size_nm":None,"zeta_mv":None,
+           "PDI":None,"protein_source":"","corona_proteins":[],"confidence":{}}
+    m = re.search(r"(\d+\.?\d*)\s*(?:nm|nanometer)", text, re.I)
+    if m: out["size_nm"] = float(m.group(1)); out["confidence"]["size_nm"] = "HIGH"
+    m = re.search(r"([+-]?\d+\.?\d*)\s*mV", text, re.I)
+    if m: out["zeta_mv"] = float(m.group(1)); out["confidence"]["zeta_mv"] = "HIGH"
+    m = re.search(r"PDI\s*[=:of]*\s*(\d+\.?\d*)", text, re.I)
+    if m: out["PDI"] = float(m.group(1)); out["confidence"]["PDI"] = "HIGH"
+    for src in ["human plasma","human serum","fetal bovine serum","FBS","PBS"]:
+        if src.lower() in text.lower():
+            out["protein_source"] = src; out["confidence"]["protein_source"] = "HIGH"; break
+    out["corona_proteins"] = [
+        {"name":p,"confidence":"MEDIUM"} for p in PROTEINS if p in text.lower()
+    ]
+    for lip in ["DSPC","DOPE","MC3","DLin","cholesterol","PEG","DOTAP"]:
+        if lip in text: out["nanoparticle_composition"] += lip + " "
+    out["nanoparticle_composition"] = out["nanoparticle_composition"].strip()
+    flags = []
+    if not out["size_nm"]:        flags.append("size_nm not found")
+    if not out["zeta_mv"]:        flags.append("zeta_mv not found")
+    if not out["corona_proteins"]:flags.append("no corona proteins detected")
+    summary = "All key fields extracted" if not flags else " | ".join(flags)
+    return json.dumps(out, indent=2), summary
+
+# ── CSS & BUILD ───────────────────────────────────────────────────────────────
+css = (
+    f"body,.gradio-container{{background:{BG}!important;color:{TXT}!important}}"
+    f".tab-nav button{{color:{TXT}!important;background:{CARD}!important}}"
+    f".tab-nav button.selected{{border-bottom:2px solid {ACC}!important;color:{ACC}!important}}"
+    f"h1,h2,h3{{color:{ACC}!important}}"
+    f".gr-button-primary{{background:{ACC}!important;border:none!important}}"
+    f"footer{{display:none!important}}"
+)
+
+with gr.Blocks(css=css, title="K R&D Lab") as demo:
+    gr.Markdown(
+        "# K R&D Lab — Computational Biology Suite\n"
+        "**Oksana Kolisnyk** · ML Engineer @ [FAZENA](https://fazena.org) · "
+        "[kosatiks-group.pp.ua](https://kosatiks-group.pp.ua)  \n"
+        "> 10 open-source computational biology tools. Hypothesis testing across disciplines."
+    )
+
+    with gr.Tabs():
+
+        with gr.TabItem("BRCA2 miRNA"):
+            gr.Markdown("### Tumor Suppressor miRNAs\nDownregulated miRNAs as miRNA-mimic therapy candidates.")
+            g1 = gr.Dropdown(["BRCA2","BRCA1","TP53"], value="BRCA2", label="Gene")
+            b1 = gr.Button("Find miRNAs", variant="primary")
+            o1 = gr.Dataframe(label="Top 5 downregulated miRNAs")
+            gr.Examples([["BRCA2"],["TP53"]], inputs=[g1])
+            b1.click(predict_mirna, g1, o1)
+
+        with gr.TabItem("TP53 siRNA"):
+            gr.Markdown("### Synthetic Lethal siRNA Targets\nDepMap essentiality + siRNA sequences.")
+            g2 = gr.Dropdown(["LUAD","BRCA","COAD"], value="LUAD", label="Cancer type")
+            b2 = gr.Button("Find Targets", variant="primary")
+            o2 = gr.Dataframe(label="Top 5 siRNA targets")
+            gr.Examples([["LUAD"],["BRCA"]], inputs=[g2])
+            b2.click(predict_sirna, g2, o2)
+
+        with gr.TabItem("lncRNA-TREM2"):
+            gr.Markdown("### lncRNA Networks in TREM2-Dependent Inflammation\nAlzheimer's therapy: ceRNA axes and ASO design.")
+            b3 = gr.Button("Load Results", variant="primary")
+            o3a = gr.Dataframe(label="ceRNA Network")
+            o3b = gr.Dataframe(label="ASO Candidates")
+            b3.click(get_lncrna, [], [o3a, o3b])
+
+        with gr.TabItem("FGFR3 Drug"):
+            gr.Markdown("### RNA-Directed Drug Discovery: FGFR3\nSmall molecules targeting FGFR3 5'UTR for bladder cancer.")
+            g4 = gr.Radio(["P1 (hairpin loop)","P10 (G-quadruplex)"],
+                          value="P1 (hairpin loop)", label="Target pocket")
+            b4 = gr.Button("Screen Compounds", variant="primary")
+            o4t = gr.Dataframe(label="Top 5 candidates")
+            o4p = gr.Image(label="Binding scores")
+            gr.Examples([["P1 (hairpin loop)"],["P10 (G-quadruplex)"]], inputs=[g4])
+            b4.click(predict_drug, g4, [o4t, o4p])
+
+        with gr.TabItem("OpenVariant"):
+            gr.Markdown("### OpenVariant — Pathogenicity Classifier\nOpen-source alternative to AlphaMissense. AUC=0.939 on ClinVar 2026.")
+            hgvs = gr.Textbox(label="HGVS notation", placeholder="BRCA1:p.R1699Q")
+            gr.Markdown("**Or enter scores manually:**")
+            with gr.Row():
+                sift = gr.Slider(0, 1, 0.5, label="SIFT (0=damaging)")
+                pp   = gr.Slider(0, 1, 0.5, label="PolyPhen-2")
+                gn   = gr.Slider(0, 0.01, 0.001, label="gnomAD AF", step=0.0001)
+            b5 = gr.Button("Predict Pathogenicity", variant="primary")
+            o5 = gr.HTML(label="Result")
+            gr.Examples([["BRCA1:p.R1699Q",0.82,0.05,0.0012],
+                         ["TP53:p.R248W",0.00,1.00,0.0],
+                         ["BRCA2:p.D2723A",0.01,0.98,0.0]],
+                        inputs=[hgvs,sift,pp,gn])
+            b5.click(predict_variant, [hgvs,sift,pp,gn], o5)
+
+        with gr.TabItem("LNP Corona"):
+            gr.Markdown("### LNP Protein Corona Prediction\nDominant corona protein + transfection efficacy from formulation.")
+            with gr.Row():
+                sz = gr.Slider(50, 300, 100, label="Size (nm)")
+                zt = gr.Slider(-40, 10, -5, label="Zeta (mV)")
+            with gr.Row():
+                pg = gr.Slider(0, 5, 1.5, label="PEG mol%")
+                lp = gr.Dropdown(["Ionizable","Cationic","Anionic","Neutral"],
+                                 value="Ionizable", label="Lipid type")
+            b6 = gr.Button("Predict", variant="primary")
+            o6 = gr.Markdown()
+            gr.Examples([[100,-5,1.5,"Ionizable"],[80,5,0.5,"Cationic"]],
+                        inputs=[sz,zt,pg,lp])
+            b6.click(predict_corona, [sz,zt,pg,lp], o6)
+
+        with gr.TabItem("Liquid Biopsy"):
+            gr.Markdown("### Protein Corona Cancer Diagnostics\nClassify cancer vs healthy from plasma protein z-scores.")
+            with gr.Row():
+                p1=gr.Slider(-3,3,0,label="CTHRC1"); p2=gr.Slider(-3,3,0,label="FHL2")
+                p3=gr.Slider(-3,3,0,label="LDHA");   p4=gr.Slider(-3,3,0,label="P4HA1")
+                p5=gr.Slider(-3,3,0,label="SERPINH1")
+            with gr.Row():
+                p6=gr.Slider(-3,3,0,label="ABCA8");  p7=gr.Slider(-3,3,0,label="CA4")
+                p8=gr.Slider(-3,3,0,label="CKB");    p9=gr.Slider(-3,3,0,label="NNMT")
+                p10=gr.Slider(-3,3,0,label="CACNA2D2")
+            b7 = gr.Button("Classify", variant="primary")
+            o7t = gr.HTML(); o7p = gr.Image(label="Feature contributions")
+            gr.Examples([[2,2,1.5,1.8,1.6,-1,-1.2,-0.8,1.4,-1.1],
+                         [0,0,0,0,0,0,0,0,0,0]],
+                        inputs=[p1,p2,p3,p4,p5,p6,p7,p8,p9,p10])
+            b7.click(predict_cancer, [p1,p2,p3,p4,p5,p6,p7,p8,p9,p10], [o7t,o7p])
+
+        with gr.TabItem("Flow Corona"):
+            gr.Markdown("### Corona Remodeling Under Blood Flow\nVroman effect model: albumin displacement by apolipoproteins.")
+            with gr.Row():
+                s8=gr.Slider(50,300,100,label="Size (nm)")
+                z8=gr.Slider(-40,10,-5,label="Zeta (mV)")
+                pg8=gr.Slider(0,5,1.5,label="PEG mol%")
+            with gr.Row():
+                ch8=gr.Dropdown(["Ionizable","Cationic","Anionic","Neutral"],
+                                value="Ionizable",label="Charge type")
+                fl8=gr.Slider(0,40,20,label="Flow rate cm/s (aorta=40)")
+            b8=gr.Button("Model Vroman Effect", variant="primary")
+            o8t=gr.Markdown(); o8p=gr.Image(label="Kinetics plot")
+            gr.Examples([[100,-5,1.5,"Ionizable",40],[150,5,0.5,"Cationic",10]],
+                        inputs=[s8,z8,pg8,ch8,fl8])
+            b8.click(predict_flow, [s8,z8,pg8,ch8,fl8], [o8t,o8p])
+
+        with gr.TabItem("LNP Brain"):
+            gr.Markdown("### LNP Brain Delivery Predictor\nApoE corona enrichment and BBB crossing probability for GBM.")
+            smi = gr.Textbox(label="Ionizable lipid SMILES",
+                             value="CC(C)CC(=O)OCC(COC(=O)CC(C)C)OC(=O)CC(C)C")
+            with gr.Row():
+                pk = gr.Slider(4, 8, 6.5, step=0.1, label="pKa")
+                zt9 = gr.Slider(-20, 10, -3, label="Zeta (mV)")
+            b9 = gr.Button("Predict BBB Crossing", variant="primary")
+            o9t = gr.Markdown(); o9p = gr.Image(label="Radar profile")
+            gr.Examples([["CC(C)CC(=O)OCC(COC(=O)CC(C)C)OC(=O)CC(C)C", 6.5, -3]],
+                        inputs=[smi,pk,zt9])
+            b9.click(predict_bbb, [smi,pk,zt9], [o9t,o9p])
+
+        with gr.TabItem("AutoCorona NLP"):
+            gr.Markdown("### AutoCorona NLP Extraction\nPaste any paper abstract to extract structured LNP corona data.")
+            txt = gr.Textbox(lines=6, label="Paper abstract or methods",
+                             placeholder="Paste text here...")
+            b10 = gr.Button("Extract Data", variant="primary")
+            o10j = gr.Code(label="Extracted JSON", language="json")
+            o10f = gr.Textbox(label="Validation flags")
+            gr.Examples([[
+                "LNPs composed of MC3, DSPC, Cholesterol (50:10:40 mol%) with 1.5% PEG-DMG. "
+                "Hydrodynamic diameter was 98 nm, zeta potential -3.2 mV, PDI 0.12. "
+                "Incubated in human plasma. Corona: albumin, apolipoprotein E, fibrinogen."
+            ]], inputs=[txt])
+            b10.click(extract_corona, txt, [o10j, o10f])
+
+    gr.Markdown(
+        "---\n**K R&D Lab** | Research tool only — not for clinical use | "
+        "[GitHub](https://github.com/oksana-kolisnyk/K-RnD-Lab-BIO-2026) | "
+        "[fazena.org](https://fazena.org)"
+    )
+
+if __name__ == "__main__":
+    demo.launch()