{"metadata":{"kernelspec":{"display_name":"Python 3","language":"python","name":"python3"},"language_info":{"codemirror_mode":{"name":"ipython","version":3},"file_extension":".py","mimetype":"text/x-python","name":"python","nbconvert_exporter":"python","pygments_lexer":"ipython3","version":"3.12.12"},"kaggle":{"accelerator":"none","dataSources":[{"sourceId":118765,"databundleVersionId":15231210,"sourceType":"competition"},{"sourceId":14604295,"sourceType":"datasetVersion","datasetId":9328538}],"dockerImageVersionId":31260,"isInternetEnabled":false,"language":"python","sourceType":"notebook","isGpuEnabled":false},"papermill":{"default_parameters":{},"duration":2354.045049,"end_time":"2026-01-13T06:54:44.524542","environment_variables":{},"exception":null,"input_path":"__notebook__.ipynb","output_path":"__notebook__.ipynb","parameters":{},"start_time":"2026-01-13T06:15:30.479493","version":"2.6.0"}},"nbformat_minor":4,"nbformat":4,"cells":[{"cell_type":"code","source":"import pandas as pd\nimport numpy as np\nimport random\nimport time\nimport warnings\nimport os, sys\nimport subprocess\n\nwarnings.filterwarnings('ignore')\n\nDATA_PATH = '/kaggle/input/stanford-rna-3d-folding-2/'\ntrain_seqs = pd.read_csv(DATA_PATH + 'train_sequences.csv')\ntest_seqs = pd.read_csv(DATA_PATH + 'test_sequences.csv')\ntrain_labels = pd.read_csv(DATA_PATH + 'train_labels.csv')\n\nsys.path.append(os.path.join(DATA_PATH, \"extra\"))\n\n# --- Robust import for Kaggle's extra/parse_fasta_py.py ---\ntry:\n    import typing as _typing\n    import builtins as _builtins\n    _builtins.Dict  = getattr(_typing, \"Dict\")\n    _builtins.Tuple = getattr(_typing, \"Tuple\")\n    _builtins.List  = getattr(_typing, \"List\")\n    from parse_fasta_py import parse_fasta as _parse_fasta_raw\n\n    def parse_fasta(fasta_content: str):\n        d = _parse_fasta_raw(fasta_content)\n        out = {}\n        for k, v in d.items():\n            out[k] = v[0] if isinstance(v, tuple) else v\n        return out\nexcept Exception:\n    def parse_fasta(fasta_content: str):\n        out = {}\n        cur = None\n        seq_parts = []\n        for line in str(fasta_content).splitlines():\n            line = line.strip()\n            if not line:\n                continue\n            if line.startswith(\">\"):\n                if cur is not None:\n                    out[cur] = \"\".join(seq_parts)\n                header = line[1:]\n                cur = header.split()[0]\n                seq_parts = []\n            else:\n                seq_parts.append(line.replace(\" \", \"\"))\n        if cur is not None:\n            out[cur] = \"\".join(seq_parts)\n        return out\n\n\ndef parse_stoichiometry(stoich: str):\n    if pd.isna(stoich) or str(stoich).strip() == \"\":\n        return []\n    out = []\n    for part in str(stoich).split(';'):\n        ch, cnt = part.split(':')\n        out.append((ch.strip(), int(cnt)))\n    return out\n\n\ndef get_chain_segments(row):\n    \"\"\"\n    Returns list of (start,end) segments in row['sequence'] corresponding\n    to chain copies in stoichiometry order. Falls back to single segment\n    if parsing fails.\n    \"\"\"\n    seq = row['sequence']\n    stoich = row.get('stoichiometry', '')\n    all_seq = row.get('all_sequences', '')\n\n    if pd.isna(stoich) or pd.isna(all_seq) or str(stoich).strip()==\"\" or str(all_seq).strip()==\"\":\n        return [(0, len(seq))]\n\n    try:\n        chain_dict = parse_fasta(all_seq)\n        order = parse_stoichiometry(stoich)\n        segs = []\n        pos = 0\n        for ch, cnt in order:\n            base = chain_dict.get(ch)\n            if base is None:\n                return [(0, len(seq))]\n            for _ in range(cnt):\n                L = len(base)\n                segs.append((pos, pos + L))\n                pos += L\n        if pos != len(seq):\n            return [(0, len(seq))]\n        return segs\n    except Exception:\n        return [(0, len(seq))]\n\n\ndef build_segments_map(df):\n    seg_map = {}\n    stoich_map = {}\n    for _, r in df.iterrows():\n        tid = r['target_id']\n        seg_map[tid] = get_chain_segments(r)\n        stoich_map[tid] = str(r.get('stoichiometry', '') if not pd.isna(r.get('stoichiometry', '')) else '')\n    return seg_map, stoich_map\n\n\n# Currently not used downstream, but keep for compatibility with other cells\ntrain_segs_map, train_stoich_map = build_segments_map(train_seqs)\ntest_segs_map,  test_stoich_map  = build_segments_map(test_seqs)\n\n\ndef process_labels(labels_df):\n    coords_dict = {}\n    prefixes = labels_df['ID'].str.rsplit('_', n=1).str[0]\n    for id_prefix, group in labels_df.groupby(prefixes):\n        coords_dict[id_prefix] = group.sort_values('resid')[['x_1', 'y_1', 'z_1']].values\n    return coords_dict\n\n\ntrain_coords_dict = process_labels(train_labels)\n\n# ============================================================\n# FIX: INSTALL BIOPYTHON FROM /kaggle/input/biopython-cp312\n# ============================================================\nBIO_WHEEL_DIR = \"/kaggle/input/biopython-cp312\"\n\ntry:\n    from Bio.Align import PairwiseAligner\n    print(\"✅ Biopython already available\")\nexcept ModuleNotFoundError:\n    wheels = [f for f in os.listdir(BIO_WHEEL_DIR) if f.endswith(\".whl\")]\n    if not wheels:\n        raise FileNotFoundError(f\"No .whl found in {BIO_WHEEL_DIR}\")\n    wheel_path = os.path.join(BIO_WHEEL_DIR, wheels[0])\n    print(f\"Installing Biopython from local wheel: {wheel_path}\")\n    subprocess.run(\n        [\"pip\", \"install\", \"--no-index\", wheel_path],\n        check=True\n    )\n    from Bio.Align import PairwiseAligner\n    print(\"✅ Biopython installed from local wheel\")\n\n# Alignment config\naligner = PairwiseAligner()\naligner.mode = 'global'\naligner.match_score = 2\naligner.mismatch_score = -1.5\n\naligner.open_gap_score   = -8\naligner.extend_gap_score = -0.4\naligner.query_left_open_gap_score   = -8\naligner.query_left_extend_gap_score = -0.4\naligner.query_right_open_gap_score  = -8\naligner.query_right_extend_gap_score = -0.4\naligner.target_left_open_gap_score  = -8\naligner.target_left_extend_gap_score = -0.4\naligner.target_right_open_gap_score = -8\naligner.target_right_extend_gap_score = -0.4\n\n\n# ============================================================\n# FAST SEQUENCE EMBEDDINGS + NEAREST NEIGHBOR PREFILTER\n# ============================================================\n\nBASES = ['A', 'C', 'G', 'U']\nBASE_TO_IDX = {b: i for i, b in enumerate(BASES)}\nNUM_KMER = 4\nVOCAB_SIZE = len(BASES) ** NUM_KMER  # 4^4 = 256\n\n\ndef seq_to_kmer_vec(seq: str) -> np.ndarray:\n    \"\"\"Map sequence to normalized 4-mer frequency vector (length 256).\"\"\"\n    vec = np.zeros(VOCAB_SIZE, dtype=np.float32)\n    if len(seq) < NUM_KMER:\n        return vec\n    total = 0\n    for i in range(len(seq) - NUM_KMER + 1):\n        k = seq[i:i+NUM_KMER]\n        idx = 0\n        valid = True\n        for ch in k:\n            ch = ch.upper().replace('T', 'U')\n            if ch not in BASE_TO_IDX:\n                valid = False\n                break\n            idx = idx * len(BASES) + BASE_TO_IDX[ch]\n        if not valid:\n            continue\n        vec[idx] += 1.0\n        total += 1\n    if total > 0:\n        vec /= total\n    return vec\n\n\ntrain_ids = []\ntrain_seqs_filtered = []\ntrain_vecs = []\ntrain_lens = []\n\nfor _, row in train_seqs.iterrows():\n    tid = row['target_id']\n    if tid not in train_coords_dict:\n        continue\n    s = row['sequence']\n    train_ids.append(tid)\n    train_seqs_filtered.append(s)\n    train_vecs.append(seq_to_kmer_vec(s))\n    train_lens.append(len(s))\n\ntrain_vecs = np.stack(train_vecs, axis=0)\ntrain_lens = np.array(train_lens, dtype=np.int32)\n\n\ndef find_similar_sequences(query_seq, top_n=5, nn_pool_size=256, max_len_ratio=0.3):\n    q_vec = seq_to_kmer_vec(query_seq)\n    if np.all(q_vec == 0):\n        candidate_idx = np.arange(len(train_ids))\n    else:\n        sims = train_vecs @ q_vec\n        if nn_pool_size < len(sims):\n            candidate_idx = np.argpartition(-sims, nn_pool_size)[:nn_pool_size]\n        else:\n            candidate_idx = np.arange(len(sims))\n\n    similar_seqs = []\n    Lq = len(query_seq)\n    for idx in candidate_idx:\n        tid = train_ids[idx]\n        tseq = train_seqs_filtered[idx]\n        Lt = train_lens[idx]\n        if abs(Lt - Lq) / max(Lt, Lq) > max_len_ratio:\n            continue\n        raw_score = aligner.score(query_seq, tseq)\n        norm_score = raw_score / (2 * min(Lq, Lt))\n        similar_seqs.append((tid, tseq, norm_score, train_coords_dict[tid]))\n\n    similar_seqs.sort(key=lambda x: x[2], reverse=True)\n    return similar_seqs[:top_n]\n\n\ndef adapt_template_to_query(query_seq, template_seq, template_coords):\n    alignment = next(iter(aligner.align(query_seq, template_seq)))\n    new_coords = np.full((len(query_seq), 3), np.nan, dtype=np.float32)\n\n    for (q_start, q_end), (t_start, t_end) in zip(*alignment.aligned):\n        t_chunk = template_coords[t_start:t_end]\n        if len(t_chunk) == (q_end - q_start):\n            new_coords[q_start:q_end] = t_chunk\n\n    for i in range(len(new_coords)):\n        if np.isnan(new_coords[i, 0]):\n            prev_v = next((j for j in range(i-1, -1, -1) if not np.isnan(new_coords[j, 0])), -1)\n            next_v = next((j for j in range(i+1, len(new_coords)) if not np.isnan(new_coords[j, 0])), -1)\n            if prev_v >= 0 and next_v >= 0:\n                w = (i - prev_v) / (next_v - prev_v)\n                new_coords[i] = (1-w)*new_coords[prev_v] + w*new_coords[next_v]\n            elif prev_v >= 0:\n                new_coords[i] = new_coords[prev_v] + [3, 0, 0]\n            elif next_v >= 0:\n                new_coords[i] = new_coords[next_v] + [3, 0, 0]\n            else:\n                new_coords[i] = [i*3, 0, 0]\n\n    return np.nan_to_num(new_coords)\n\n\ndef adaptive_rna_constraints(coordinates, sequence, confidence=1.0):\n    refined_coords = coordinates.copy()\n    n = len(sequence)\n    strength = 0.68 * (1.0 - min(confidence, 0.96))\n\n    for _ in range(2):\n        for i in range(n - 1):\n            p1, p2 = refined_coords[i], refined_coords[i+1]\n            dist = np.linalg.norm(p2 - p1)\n            if dist > 0:\n                adj = (5.95 - dist) * strength * 0.45\n                refined_coords[i+1] += (p2 - p1) / dist * adj\n\n            if i < n - 2:\n                p3 = refined_coords[i+2]\n                dist2 = np.linalg.norm(p3 - p1)\n                if dist2 > 0:\n                    adj2 = (10.2 - dist2) * strength * 0.25\n                    refined_coords[i+2] += (p3 - p1) / dist2 * adj2\n\n    return refined_coords\n\n\ndef predict_rna_structures(sequence, target_id, n_predictions=5):\n    predictions = []\n    similar_seqs = find_similar_sequences(sequence, top_n=n_predictions)\n\n    for i in range(n_predictions):\n        if i < len(similar_seqs):\n            t_id, t_seq, sim, t_coords = similar_seqs[i]\n            adapted = adapt_template_to_query(sequence, t_seq, t_coords)\n            refined = adaptive_rna_constraints(adapted, sequence, confidence=sim)\n            noise = 0.0 if i == 0 else max(0.006, (0.38 - sim) * 0.07)\n            if noise > 0:\n                refined = refined + np.random.normal(0, noise, refined.shape)\n            predictions.append(refined)\n        else:\n            n = len(sequence)\n            coords = np.zeros((n, 3), dtype=np.float32)\n            for j in range(1, n):\n                coords[j] = coords[j-1] + [4.0, 0, 0]\n            predictions.append(coords)\n\n    return predictions\n\n\nall_predictions = []\nstart_time = time.time()\nfor idx, row in test_seqs.iterrows():\n    if idx % 10 == 0:\n        print(f\"Processing {idx} | {time.time()-start_time:.1f}s\")\n    tid, seq = row['target_id'], row['sequence']\n    preds = predict_rna_structures(seq, tid)\n    for j in range(len(seq)):\n        res = {'ID': f\"{tid}_{j+1}\", 'resname': seq[j], 'resid': j+1}\n        for i in range(5):\n            res[f'x_{i+1}'], res[f'y_{i+1}'], res[f'z_{i+1}'] = preds[i][j]\n        all_predictions.append(res)\n\nsub = pd.DataFrame(all_predictions)\ncols = ['ID', 'resname', 'resid'] + [f'{c}_{i}' for i in range(1, 6) for c in ['x', 'y', 'z']]\nsub[cols].to_csv('submission.csv', index=False)\nprint(\"submission.csv! saved\")\n","metadata":{"trusted":true},"outputs":[],"execution_count":null}]}