{"metadata":{"kernelspec":{"language":"python","display_name":"Python 3","name":"python3"},"language_info":{"name":"python","version":"3.10.12","mimetype":"text/x-python","codemirror_mode":{"name":"ipython","version":3},"pygments_lexer":"ipython3","nbconvert_exporter":"python","file_extension":".py"},"kaggle":{"accelerator":"gpu","dataSources":[{"sourceId":87793,"databundleVersionId":12024591,"sourceType":"competition"},{"sourceId":7292242,"sourceType":"datasetVersion","datasetId":4229373},{"sourceId":10855324,"sourceType":"datasetVersion","datasetId":6742586},{"sourceId":11118830,"sourceType":"datasetVersion","datasetId":6933267},{"sourceId":11534041,"sourceType":"datasetVersion","datasetId":7234091},{"sourceId":157966080,"sourceType":"kernelVersion"},{"sourceId":224830487,"sourceType":"kernelVersion"}],"dockerImageVersionId":30918,"isInternetEnabled":true,"language":"python","sourceType":"notebook","isGpuEnabled":true}},"nbformat_minor":4,"nbformat":4,"cells":[{"cell_type":"code","source":"MODEL_TYPE='protenix'\nMODE = 'validation'","metadata":{"trusted":true,"execution":{"iopub.status.busy":"2025-05-02T05:46:22.575206Z","iopub.execute_input":"2025-05-02T05:46:22.575531Z","iopub.status.idle":"2025-05-02T05:46:22.579877Z","shell.execute_reply.started":"2025-05-02T05:46:22.575506Z","shell.execute_reply":"2025-05-02T05:46:22.578724Z"}},"outputs":[],"execution_count":null},{"cell_type":"code","source":"if MODEL_TYPE=='protenix' and MODE == 'validation':\n    !pip install --no-deps protenix\n    !pip install biopython\n    !pip install ml-collections\n    !pip install biotite==1.0.1\n    !pip install rdkit\n!export PROTENIX_DATA_ROOT_DIR=/kaggle/input/protenix-checkpoints","metadata":{"trusted":true,"execution":{"iopub.status.busy":"2025-05-02T05:46:22.591885Z","iopub.execute_input":"2025-05-02T05:46:22.592164Z","iopub.status.idle":"2025-05-02T05:46:48.105054Z","shell.execute_reply.started":"2025-05-02T05:46:22.592143Z","shell.execute_reply":"2025-05-02T05:46:48.104018Z"},"_kg_hide-output":true},"outputs":[],"execution_count":null},{"cell_type":"code","source":"! mkdir /af3-dev \n! ln -s /kaggle/input/protenix-checkpoints /af3-dev/release_data\n! ls /af3-dev/release_data/","metadata":{"trusted":true,"execution":{"iopub.status.busy":"2025-05-02T05:46:48.106700Z","iopub.execute_input":"2025-05-02T05:46:48.107100Z","iopub.status.idle":"2025-05-02T05:46:48.466836Z","shell.execute_reply.started":"2025-05-02T05:46:48.107067Z","shell.execute_reply":"2025-05-02T05:46:48.465837Z"},"_kg_hide-output":true},"outputs":[],"execution_count":null},{"cell_type":"markdown","source":"## Helper scripts","metadata":{}},{"cell_type":"code","source":"%%writefile proteinx_predict.py\n# scripts/nufold_predict.py\n\nimport Bio\n\nfrom copy import deepcopy\n\nimport pandas as pd\nfrom Bio.PDB import Atom, Model, Chain, Residue, Structure, PDBParser\nfrom Bio import SeqIO\nimport os, sys\nimport re\nimport numpy as np\nimport torch\n\nimport matplotlib\nimport matplotlib.pyplot as plt\nfrom tqdm import tqdm\nimport time\ntime0=time.time()\n\nprint('IMPORT OK !!!!')\n\nPYTHON = sys.executable\nprint('PYTHON',PYTHON)\n\nRHONET_DIR=\\\n'/kaggle/input/data-for-demo-for-rhofold-plus-with-kaggle-msa/RhoFold-main'\n#'<your downloaded rhofold repo>/RhoFold-main'\n\nUSALIGN = \\\n'/kaggle/working//USalign'\n#'<your us align path>/USalign'\n\nos.system('cp /kaggle/input/usalign/USalign /kaggle/working/')\nos.system('sudo chmod u+x /kaggle/working//USalign')\nsys.path.append(RHONET_DIR)\n\n\nDATA_KAGGLE_DIR = '/kaggle/input/stanford-rna-3d-folding'\n\n\n# helper ----\nclass dotdict(dict):\n\t__setattr__ = dict.__setitem__\n\t__delattr__ = dict.__delitem__\n\n\tdef __getattr__(self, name):\n\t\ttry:\n\t\t\treturn self[name]\n\t\texcept KeyError:\n\t\t\traise AttributeError(name)\n\n# visualisation helper ----\ndef set_aspect_equal(ax):\n\tx_limits = ax.get_xlim()\n\ty_limits = ax.get_ylim()\n\tz_limits = ax.get_zlim()\n\n\t# Compute the mean of each axis\n\tx_middle = np.mean(x_limits)\n\ty_middle = np.mean(y_limits)\n\tz_middle = np.mean(z_limits)\n\n\t# Compute the max range across all axes\n\tmax_range = max(x_limits[1] - x_limits[0],\n\t\t\t\t\ty_limits[1] - y_limits[0],\n\t\t\t\t\tz_limits[1] - z_limits[0]) / 2.0\n\n\t# Set the new limits to ensure equal scaling\n\tax.set_xlim(x_middle - max_range, x_middle + max_range)\n\tax.set_ylim(y_middle - max_range, y_middle + max_range)\n\tax.set_zlim(z_middle - max_range, z_middle + max_range)\n\n\n\n\n# xyz df helper --------------------\ndef get_truth_df(target_id):\n    truth_df = LABEL_DF[LABEL_DF['target_id'] == target_id]\n    truth_df = truth_df.reset_index(drop=True)\n    return truth_df\n\ndef parse_output_to_df(output, seq, target_id):\n    df = []\n    chain_data = []\n    for i, res in enumerate(seq):\n        d=dict(ID = target_id,\n                    resname=res,\n                    resid=i+1)\n        for n in range(len(output)):\n            d={**d, f'x_{n+1}': round(output[n,i,0].item(),3),\n                     f'y_{n+1}': round(output[n,i,1].item(),3),\n                     f'z_{n+1}': round(output[n,i,2].item(),3)}\n        chain_data.append(d)\n\n    if len(chain_data)!=0:\n        chain_df = pd.DataFrame(chain_data)\n        df.append(chain_df)\n        ##print(chain_df)\n    return df\n\ndef parse_pdb_to_df(pdb_file, target_id):\n    parser = PDBParser()\n    structure = parser.get_structure('', pdb_file)\n\n    df = []\n    for model in structure:\n        for chain in model:\n            print(chain)\n            chain_data = []\n            for residue in chain:\n                # print(residue)\n                if residue.get_resname() in ['A', 'U', 'G', 'C']:\n                    # Check if the residue has a C1' atom\n                    if 'C1\\'' in residue:\n                        atom = residue['C1\\'']\n                        xyz = atom.get_coord()\n                        resname = residue.get_resname()\n                        resid = residue.get_id()[1]\n\n                        #todo detect discontinous: resid = prev_resid+1\n                        #ID\tresname\tresid\tx_1\ty_1\tz_1\n                        chain_data.append(dict(\n                            ID = target_id+'_'+str(resid),\n                            resname=resname,\n                            resid=resid,\n                            x_1=xyz[0],\n                            y_1=xyz[1],\n                            z_1=xyz[2],\n                        ))\n                        ##print(f\"Residue {resname} {resid}, Atom: {atom.get_name()}, xyz: {xyz}\")\n\n            if len(chain_data)!=0:\n                chain_df = pd.DataFrame(chain_data)\n                df.append(chain_df)\n                ##print(chain_df)\n    return df\n\n# usalign helper --------------------\ndef write_target_line(\n    atom_name, atom_serial, residue_name, chain_id, residue_num, x_coord, y_coord, z_coord, occupancy=1.0, b_factor=0.0, atom_type='P'\n):\n    \"\"\"\n    Writes a single line of PDB format based on provided atom information.\n\n    Args:\n        atom_name (str): Name of the atom (e.g., \"N\", \"CA\").\n        atom_serial (int): Atom serial number.\n        residue_name (str): Residue name (e.g., \"ALA\").\n        chain_id (str): Chain identifier.\n        residue_num (int): Residue number.\n        x_coord (float): X coordinate.\n        y_coord (float): Y coordinate.\n        z_coord (float): Z coordinate.\n        occupancy (float, optional): Occupancy value (default: 1.0).\n        b_factor (float, optional): B-factor value (default: 0.0).\n\n    Returns:\n        str: A single line of PDB string.\n    \"\"\"\n    return f'ATOM  {atom_serial:>5d}  {atom_name:<5s} {residue_name:<3s} {residue_num:>3d}    {x_coord:>8.3f}{y_coord:>8.3f}{z_coord:>8.3f}{occupancy:>6.2f}{b_factor:>6.2f}           {atom_type}\\n'\n\ndef write_xyz_to_pdb(df, pdb_file, xyz_id = 1):\n    resolved_cnt = 0\n    with open(pdb_file, 'w') as target_file:\n        for _, row in df.iterrows():\n            x_coord = row[f'x_{xyz_id}']\n            y_coord = row[f'y_{xyz_id}']\n            z_coord = row[f'z_{xyz_id}']\n\n            if x_coord > -1e17 and y_coord > -1e17 and z_coord > -1e17:\n                resolved_cnt += 1\n                target_line = write_target_line(\n                    atom_name=\"C1'\",\n                    atom_serial=int(row['resid']),\n                    residue_name=row['resname'],\n                    chain_id='0',\n                    residue_num=int(row['resid']),\n                    x_coord=x_coord,\n                    y_coord=y_coord,\n                    z_coord=z_coord,\n                    atom_type='C',\n                )\n                target_file.write(target_line)\n    return resolved_cnt\n\ndef parse_usalign_for_tm_score(output):\n    # Extract TM-score based on length of reference structure (second)\n    tm_score_match = re.findall(r'TM-score=\\s+([\\d.]+)', output)[1]\n    if not tm_score_match:\n        raise ValueError('No TM score found')\n    return float(tm_score_match)\n\ndef parse_usalign_for_transform(output):\n    # Locate the rotation matrix section\n    matrix_lines = []\n    found_matrix = False\n\n    for line in output.splitlines():\n        if \"The rotation matrix to rotate Structure_1 to Structure_2\" in line:\n            found_matrix = True\n        elif found_matrix and re.match(r'^\\d+\\s+[-\\d.]+\\s+[-\\d.]+\\s+[-\\d.]+\\s+[-\\d.]+$', line):\n            matrix_lines.append(line)\n        elif found_matrix and not line.strip():\n            break  # Stop parsing if an empty line is encountered after the matrix\n\n    # Parse the rotation matrix values\n    rotation_matrix = []\n    for line in matrix_lines:\n        parts = line.split()\n        row_values = list(map(float, parts[1:]))  # Skip the first column (index)\n        rotation_matrix.append(row_values)\n\n    return np.array(rotation_matrix)\n\ndef make_pdb(predict_df, truth_df):\n    truth_pdb = '~truth.pdb'\n    predict_pdb = '~predict.pdb'\n    write_xyz_to_pdb(predict_df, predict_pdb, xyz_id=1)\n    write_xyz_to_pdb(truth_df, truth_pdb, xyz_id=1)\n    return truth_pdb, predict_pdb\ndef call_usalign(predict_df, truth_df, verbose=1):\n    truth_pdb, predict_pdb = make_pdb(predict_df, truth_df)\n    command = f'{USALIGN} {predict_pdb} {truth_pdb} -atom \" C1\\'\" -m -'\n    output = os.popen(command).read()\n    if verbose==1:\n        print(output)\n    tm_score = parse_usalign_for_tm_score(output)\n    transform = parse_usalign_for_transform(output)\n    return tm_score, transform\n\nprint('HELPER OK!!!')\n\nfrom runner.batch_inference import get_default_runner\nfrom runner.inference import update_inference_configs, InferenceRunner\nfrom runner.dumper import DataDumper\n\nfrom protenix.data.infer_data_pipeline import InferenceDataset\n\nnp.random.seed(0)\ntorch.random.manual_seed(0)\ntorch.cuda.manual_seed_all(0)\n\nclass DictDataset(InferenceDataset):\n    def __init__(\n        self,\n        seq_list: list,\n        dump_dir: str,\n        id_list: list = None,\n        use_msa: bool = False,\n    ) -> None:\n\n        self.dump_dir = dump_dir\n        self.use_msa = use_msa\n        if isinstance(id_list,type(None)):\n            self.inputs = [{\"sequences\": \n                            [{\"rnaSequence\": \n                              {\"sequence\": seq, \n                               \"count\": 1}}],\n                            \"name\": \"query\"} for seq in seq_list]\n        else:\n            self.inputs = [{\"sequences\": \n                            [{\"rnaSequence\": \n                              {\"sequence\": seq, \n                               \"count\": 1}}],\n                            \"name\": i} for i, seq in zip(id_list,seq_list)]\n                \nfrom configs.configs_base import configs as configs_base\nfrom configs.configs_data import data_configs\nfrom configs.configs_inference import inference_configs\nfrom protenix.config.config import parse_configs\n\nconfigs_base[\"use_deepspeed_evo_attention\"] = (\nos.environ.get(\"USE_DEEPSPEED_EVO_ATTENTION\", False) == \"true\")\nconfigs_base[\"model\"][\"N_cycle\"] = 10 #10\nconfigs_base[\"sample_diffusion\"][\"N_sample\"] = 3\nconfigs_base[\"sample_diffusion\"][\"N_step\"] = 200\ninference_configs['load_checkpoint_path']='/kaggle/input/protenix-checkpoints/model_v0.2.0.pt'\nconfigs = {**configs_base, **{\"data\": data_configs}, **inference_configs}\n\nconfigs = parse_configs(\n        configs=configs,\n        fill_required_with_null=True,\n    )\n\nrunner=InferenceRunner(configs)\ndumper = runner.dumper  # already initialized for you\n\nimport os\nimport argparse\nimport warnings\nimport pandas as pd\nimport torch\nfrom tqdm import tqdm\nimport gc\nimport os\nimport sys\nimport argparse\nimport pandas as pd\nimport torch\nimport gc\nfrom tqdm import tqdm\nimport os\nimport argparse\nimport pandas as pd\nimport torch\nimport gc\nfrom tqdm import tqdm\n\nfrom Bio.PDB import MMCIFParser, PDBIO\n\ndef cif_to_pdb(cif_file, pdb_file):\n    parser = MMCIFParser(QUIET=True)\n    struct = parser.get_structure(os.path.basename(cif_file), cif_file)\n    io = PDBIO()\n    io.set_structure(struct)\n    io.save(pdb_file)\n\ndef main():\n    parser = argparse.ArgumentParser(description=\"ProteinX batched inference script\")\n    parser.add_argument('--input_csv',  required=True, help='CSV with [sequence,target_id]')\n    parser.add_argument('--output_dir', required=True, help='Directory for CIFs, PDBs, and CSV')\n    args = parser.parse_args()\n\n    os.makedirs(args.output_dir, exist_ok=True)\n    df = pd.read_csv(args.input_csv)\n\n    # Prepare dataset\n    dataset = DictDataset(\n        seq_list=df['sequence'].tolist(),\n        id_list=df['target_id'].tolist(),\n        use_msa=False,\n        dump_dir=args.output_dir\n    )\n\n    # Initialize runner & dumper\n    runner = InferenceRunner(configs)\n    dumper = runner.dumper\n\n    records = []\n    seed = 0\n    torch.manual_seed(seed)\n\n    for idx, seq in tqdm(enumerate(df['sequence']), total=len(df)):\n        target_id = df['target_id'].iloc[idx]\n        if len(seq) > 300:\n            print(f\"Skipping {target_id}, length {len(seq)} > 300\")\n            continue\n\n        with torch.no_grad():\n            data, atom_array, err = dataset[idx]\n            if err:\n                continue\n\n            # Update configs & predict\n            new_cfg = update_inference_configs(runner.configs, data['N_token'].item())\n            runner.update_model_configs(new_cfg)\n            pred = runner.predict(data)\n\n        # Dump *all* 3 samples into CIF\n        dumper.dump(\n            dataset_name=\"\",\n            pdb_id=target_id,\n            seed=seed,\n            pred_dict=pred,\n            atom_array=atom_array,\n            entity_poly_type=data[\"entity_poly_type\"],\n        )\n\n        # Now convert each sample_i.cif → sample_i.pdb\n        n_samples = configs[\"sample_diffusion\"][\"N_sample\"]\n        for sample_i in range(n_samples):\n            cif_path = os.path.join(\n                '/kaggle/working/output', target_id, f\"seed_{seed}\", \"predictions\",\n                f\"{target_id}_seed_{seed}_sample_{sample_i}.cif\"\n            )\n            pdb_path = os.path.join(\n                args.output_dir,\n                f\"{target_id}_seed_{seed}_sample_{sample_i}.pdb\"\n            )\n\n            try:\n                cif_to_pdb(cif_path, pdb_path)\n                # print(f\"Converted {cif_path} → {pdb_path}\")\n            except FileNotFoundError:\n                print(f\"Warning: CIF not found at {cif_path}\")\n                continue\n\n            records.append({\n                'sequence': seq,\n                'target_id': target_id,\n                'seed': seed,\n                'sample': sample_i,\n                'predicted_model_path': pdb_path\n            })\n\n        # Cleanup\n        del pred, atom_array, data\n        torch.cuda.empty_cache()\n        gc.collect()\n\n    # Save summary CSV\n    out_csv = os.path.join('/kaggle/working/', 'proteinx_train_preds.csv')\n    pd.DataFrame(records).to_csv(out_csv, index=False)\n    print(f\"Saved predictions CSV at {out_csv}\")\n\nif __name__ == '__main__':\n    main()","metadata":{"trusted":true,"execution":{"iopub.status.busy":"2025-05-02T06:26:09.370443Z","iopub.execute_input":"2025-05-02T06:26:09.370813Z","iopub.status.idle":"2025-05-02T06:26:09.381778Z","shell.execute_reply.started":"2025-05-02T06:26:09.370786Z","shell.execute_reply":"2025-05-02T06:26:09.380739Z"},"_kg_hide-input":true},"outputs":[],"execution_count":null},{"cell_type":"code","source":"!python proteinx_predict.py --input_csv /kaggle/input/stanford-rna-3d-folding/train_sequences.csv --output_dir  'proteinx_predictions'","metadata":{"trusted":true,"execution":{"iopub.status.busy":"2025-05-02T06:26:09.383248Z","iopub.execute_input":"2025-05-02T06:26:09.383519Z","execution_failed":"2025-05-02T17:47:17.000Z"}},"outputs":[],"execution_count":null},{"cell_type":"code","source":"import os, sys\n\nINPUT_CSV  = '/kaggle/input/stanford-rna-3d-folding/train_sequences.csv'\nOUTPUT_DIR = 'proteinx_predictions'\nSCRIPTS    = ['proteinx_predict.py']\n\nfor script in SCRIPTS:\n    # cmd = (\n    #     f\"{sys.executable} -u {script} \"\n    #     f\"--input_csv {INPUT_CSV} --output_dir {OUTPUT_DIR}\"\n    # )\n    cmd = (\n        f\"python -u {script} \"\n        f\"--input_csv {INPUT_CSV} --output_dir {OUTPUT_DIR}\"\n    )\n    print(f\"▶︎ {cmd}\\n\")\n    exit_code = os.system(cmd)\n    if exit_code != 0:\n        raise RuntimeError(f\"{script} failed (exit code {exit_code})\")\n    print(f\"\\n✅ {script} done\\n\",flush=True)","metadata":{"trusted":true,"execution":{"execution_failed":"2025-05-02T17:47:17.006Z"}},"outputs":[],"execution_count":null},{"cell_type":"code","source":"!zip -r /kaggle/working/proteinx_predictions.zip /kaggle/working/proteinx_predictions","metadata":{"trusted":true},"outputs":[],"execution_count":null}]}