Upload app.py

app.py

@@ -0,0 +1,197 @@
+#!/usr/bin/env python3
+"""
+Flexynesis Tissue VAE – Web Application
+========================================
+Upload a gene expression matrix → get UBERON tissue classification + embeddings.
+
+Run locally:
+    streamlit run app.py
+
+Deploy on HuggingFace Spaces:
+    See setup_webapp.py for file preparation.
+
+Author: Amit Pande, MDC Berlin/BIMSB
+"""
+
+import streamlit as st
+import pandas as pd
+import numpy as np
+import torch
+import joblib
+from pathlib import Path
+from sklearn.neighbors import KNeighborsClassifier
+from collections import Counter
+
+# ── Config ──
+MODEL_DIR = Path("model")
+K = 5
+
+st.set_page_config(page_title="Flexynesis Tissue VAE", page_icon="🧬", layout="wide")
+st.title("🧬 Flexynesis Tissue VAE")
+st.markdown(
+    "Upload a bulk RNA-seq gene expression matrix to classify tissue-of-origin "
+    "using a supervised VAE trained on **75,619 samples** from TCGA, GTEx, DepMap, and ARCHS4 "
+    "across **43 UBERON tissue categories** (90.7% balanced accuracy, 121-dim latent space)."
+)
+
+# ── Load model + artifacts ──
+@st.cache_resource
+def load_all():
+    model = torch.load(MODEL_DIR / "vae_tissue.final_model.pth",
+                       map_location='cpu', weights_only=False)
+    model.eval()
+
+    art = joblib.load(MODEL_DIR / "vae_tissue.artifacts.joblib")
+    gene_list  = list(art['feature_lists']['gex'])
+    scaler     = art['transforms']['gex']
+    label_enc  = art['label_encoders']['uberon_tissue']
+
+    train_emb  = pd.read_csv(MODEL_DIR / "embeddings_train.csv", index_col=0)
+    test_emb   = pd.read_csv(MODEL_DIR / "embeddings_test.csv",  index_col=0)
+    train_clin = pd.read_csv(MODEL_DIR / "train_clin.csv", index_col=0)
+    test_clin  = pd.read_csv(MODEL_DIR / "test_clin.csv",  index_col=0)
+
+    all_emb  = pd.concat([train_emb, test_emb])
+    all_clin = pd.concat([train_clin, test_clin])
+    idx      = all_emb.index.intersection(all_clin.index)
+    ref_emb  = all_emb.loc[idx]
+    ref_clin = all_clin.loc[idx]
+    mask = (ref_clin['uberon_tissue'].notna() &
+            ~ref_clin['uberon_tissue'].isin(['unknown','other','unmapped','nan','']))
+    ref_emb  = ref_emb[mask]
+    ref_clin = ref_clin[mask]
+
+    knn = KNeighborsClassifier(n_neighbors=K, metric='cosine', n_jobs=-1)
+    knn.fit(ref_emb.values, ref_clin['uberon_tissue'].values)
+
+    classes = list(label_enc.categories_[0]) if hasattr(label_enc, 'categories_') else None
+
+    return model, gene_list, scaler, knn, ref_emb, ref_clin, classes
+
+try:
+    model, gene_list, scaler, knn_ref, ref_emb, ref_clin, classes = load_all()
+    st.success(
+        f"✅ {len(gene_list):,} genes · "
+        f"{len(ref_emb):,} reference samples · "
+        f"{ref_clin['uberon_tissue'].nunique()} tissues"
+    )
+except Exception as e:
+    st.error(f"Load error: {e}")
+    st.info("Place model files in `model/` directory. Run `python setup_webapp.py` first.")
+    st.stop()
+
+# ── Helpers ──
+def orient_matrix(df):
+    genes_in_cols = len(set(df.columns) & set(gene_list))
+    genes_in_rows = len(set(df.index)   & set(gene_list))
+    if genes_in_rows > genes_in_cols:
+        df = df.T
+    return df
+
+def encode(model, X_tensor):
+    with torch.no_grad():
+        h  = model.encoders[0](X_tensor)
+        mu = model.FC_mean(h[0])
+    return mu
+
+def classify_from_logits(model, mu, classes):
+    with torch.no_grad():
+        logits = model.MLPs['uberon_tissue'](mu)
+        # Mask out nan class if present
+        label_mapping = model.dataset.label_mappings['uberon_tissue']
+        name_to_idx   = {v: k for k, v in label_mapping.items()}
+        nan_idx = name_to_idx.get('nan', None)
+        if nan_idx is not None:
+            logits[:, nan_idx] = -1e9
+        probs    = torch.softmax(logits, dim=1)
+        pred_idx = logits.argmax(dim=1)
+    pred_labels = [label_mapping[int(i)] for i in pred_idx]
+    return pred_labels, probs.numpy()
+
+# ── File upload ──
+st.markdown("---")
+col1, col2 = st.columns([2, 1])
+with col1:
+    uploaded = st.file_uploader(
+        "Upload CSV/TSV (genes × samples or samples × genes)",
+        type=["csv", "tsv", "txt"],
+        help="HGNC gene symbols. Log2-transformed expression values (TPM, RPKM, or counts).")
+with col2:
+    st.markdown(f"""
+    **Expected input:**
+    - HGNC gene symbols
+    - {len(gene_list):,} genes used by model
+    - Log2-transformed expression
+    """)
+
+if uploaded:
+    sep = '\t' if uploaded.name.endswith(('.tsv', '.txt')) else ','
+    df  = pd.read_csv(uploaded, index_col=0, sep=sep)
+    st.write(f"**Uploaded:** {df.shape[0]:,} × {df.shape[1]:,}")
+    st.dataframe(df.iloc[:5, :5], use_container_width=True)
+
+    if st.button("🚀 Classify Tissues", type="primary"):
+        with st.spinner("Processing..."):
+            df      = orient_matrix(df)
+            overlap = len(set(df.columns) & set(gene_list))
+            st.write(f"Gene overlap: **{overlap:,}/{len(gene_list):,}** ({100*overlap/len(gene_list):.1f}%)")
+            if overlap < 1000:
+                st.warning("Low gene overlap — results may be unreliable.")
+
+            aligned        = pd.DataFrame(0.0, index=df.index, columns=gene_list)
+            common         = [g for g in gene_list if g in df.columns]
+            aligned[common] = df[common].values
+            aligned        = aligned.fillna(0)
+
+            X_scaled = scaler.transform(aligned.values)
+            X_tensor = torch.tensor(X_scaled, dtype=torch.float32)
+
+            mu         = encode(model, X_tensor)
+            embeddings = mu.numpy()
+
+            pred_labels, probs = classify_from_logits(model, mu, classes)
+            max_probs          = probs.max(axis=1)
+
+            distances, indices = knn_ref.kneighbors(embeddings)
+            breakdowns = []
+            for i in range(len(embeddings)):
+                nn_clin = ref_clin.iloc[indices[i]]
+                src_counts = Counter(nn_clin['source'].values)
+                breakdowns.append('; '.join(f"{s}:{c}" for s, c in src_counts.most_common()))
+
+        st.markdown("---")
+        st.subheader("📊 Results")
+        results = pd.DataFrame({
+            'Sample':           df.index,
+            'Tissue':           pred_labels,
+            'Confidence':       [f"{p:.1%}" for p in max_probs],
+            'kNN Dist':         distances.mean(axis=1).round(4),
+            'Sources':          breakdowns,
+        })
+        st.dataframe(results, use_container_width=True, height=400)
+
+        col1, col2 = st.columns(2)
+        with col1:
+            st.subheader("Tissue Distribution")
+            st.bar_chart(pd.Series(pred_labels).value_counts())
+        with col2:
+            st.subheader("Confidence Distribution")
+            st.bar_chart(pd.DataFrame({'Confidence': max_probs}, index=df.index))
+
+        st.markdown("---")
+        c1, c2 = st.columns(2)
+        with c1:
+            emb_df = pd.DataFrame(embeddings, index=df.index,
+                                  columns=[f"z{i}" for i in range(121)])
+            st.download_button("📥 Embeddings (CSV)", emb_df.to_csv(),
+                               "flexynesis_embeddings.csv", "text/csv")
+        with c2:
+            st.download_button("📥 Classifications (CSV)", results.to_csv(index=False),
+                               "flexynesis_classifications.csv", "text/csv")
+
+st.markdown("---")
+st.caption(
+    "Flexynesis Tissue VAE · Akalin Lab, MDC Berlin/BIMSB · "
+    "75,619 samples · 43 UBERON tissues · 90.7% balanced accuracy · "
+    "GitHub: github.com/BIMSBbioinfo/flexynesis"
+)