{ "cells": [ { "cell_type": "markdown", "metadata": {}, "source": [ "## 数据处理：SMILES信息在Gene里，CP里面没有，所以首先在这里实现映射和对齐" ] }, { "cell_type": "code", "execution_count": 1, "metadata": {}, "outputs": [], "source": [ "import numpy as np\n", "import pandas as pd\n", "import matplotlib.pyplot as plt\n", "import seaborn as sns\n", "import os\n", "import h5py\n", "\n", "def load_from_HDF(fname):\n", " \"\"\"Load data from a HDF5 file to a dictionary.\"\"\"\n", " data = dict()\n", " with h5py.File(fname, 'r') as f:\n", " for key in f:\n", " data[key] = np.asarray(f[key])\n", " if isinstance(data[key][0], np.bytes_):\n", " data[key] = data[key].astype(str)\n", " return data" ] }, { "cell_type": "code", "execution_count": 2, "metadata": {}, "outputs": [ { "name": "stdout", "output_type": "stream", "text": [ "(113416, 898) (61313, 980)\n" ] }, { "data": { "text/html": [ "
\n", "\n", "\n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", "
pert_dosedet_plateCPD_SMILES221227_x_at212345_s_at218597_s_at217140_s_at209253_at214404_x_at219888_at...218397_at202996_at204608_at211071_s_at203341_at202801_at206414_s_at204978_at205379_at203897_at
050.0PAC001_U2OS_6H_X1_B1_UNI4445LCC(C)[C@H](CO)Nc1nc(Nc2cc(N)cc(Cl)c2)c2ncn(C(C...-0.118505-0.293445-0.2940880.292745-0.1458960.973580.247915...-0.5299050.8049920.430269-0.97559-1.7406050.3403820.4987780.3521520.535910.52484
112.7PAC001_U2OS_6H_X1_B1_UNI4445LIc1cccc(CSc2nnc(o2)-c2ccncc2)c10.043264-0.264945-0.0006330.0009350.177604-0.504720.214315...0.228895-0.3813080.5569690.25051-0.025625-0.024428-0.2251220.533052-0.19329-0.25436
250.0PAC001_U2OS_6H_X1_B1_UNI4445LClc1cc(Cl)c(NC(=O)Nc2ccncc2)c(Cl)c1-0.0708050.1957550.004606-0.040855-0.067346-0.19872-0.053195...-0.031795-0.051728-0.0177410.056810.1306950.005919-0.408722-0.057177-0.256990.31004
316.2PAC001_U2OS_6H_X1_B1_UNI4445LCOc1ccc(CNC(=O)Nc2ncc(s2)[N+]([O-])=O)cc10.027165-0.1495450.173113-0.100695-0.309296-0.20782-0.323985...-0.013787-0.1776080.219869-0.161290.365195-0.040848-0.2511220.0407320.17091-0.05144
412.5PAC001_U2OS_6H_X1_B1_UNI4445LOC(=O)c1ccc2c3nc4ccccc4n3c(=O)c3cccc1c230.3032940.254455-0.055418-0.0536350.000455-0.109080.031425...-0.052585-0.1961080.2454690.007840.093595-0.3034180.474878-0.022947-0.285890.06423
\n", "

5 rows × 980 columns

\n", "
" ], "text/plain": [ " pert_dose det_plate \\\n", "0 50.0 PAC001_U2OS_6H_X1_B1_UNI4445L \n", "1 12.7 PAC001_U2OS_6H_X1_B1_UNI4445L \n", "2 50.0 PAC001_U2OS_6H_X1_B1_UNI4445L \n", "3 16.2 PAC001_U2OS_6H_X1_B1_UNI4445L \n", "4 12.5 PAC001_U2OS_6H_X1_B1_UNI4445L \n", "\n", " CPD_SMILES 221227_x_at \\\n", "0 CC(C)[C@H](CO)Nc1nc(Nc2cc(N)cc(Cl)c2)c2ncn(C(C... -0.118505 \n", "1 Ic1cccc(CSc2nnc(o2)-c2ccncc2)c1 0.043264 \n", "2 Clc1cc(Cl)c(NC(=O)Nc2ccncc2)c(Cl)c1 -0.070805 \n", "3 COc1ccc(CNC(=O)Nc2ncc(s2)[N+]([O-])=O)cc1 0.027165 \n", "4 OC(=O)c1ccc2c3nc4ccccc4n3c(=O)c3cccc1c23 0.303294 \n", "\n", " 212345_s_at 218597_s_at 217140_s_at 209253_at 214404_x_at 219888_at \\\n", "0 -0.293445 -0.294088 0.292745 -0.145896 0.97358 0.247915 \n", "1 -0.264945 -0.000633 0.000935 0.177604 -0.50472 0.214315 \n", "2 0.195755 0.004606 -0.040855 -0.067346 -0.19872 -0.053195 \n", "3 -0.149545 0.173113 -0.100695 -0.309296 -0.20782 -0.323985 \n", "4 0.254455 -0.055418 -0.053635 0.000455 -0.10908 0.031425 \n", "\n", " ... 218397_at 202996_at 204608_at 211071_s_at 203341_at 202801_at \\\n", "0 ... -0.529905 0.804992 0.430269 -0.97559 -1.740605 0.340382 \n", "1 ... 0.228895 -0.381308 0.556969 0.25051 -0.025625 -0.024428 \n", "2 ... -0.031795 -0.051728 -0.017741 0.05681 0.130695 0.005919 \n", "3 ... -0.013787 -0.177608 0.219869 -0.16129 0.365195 -0.040848 \n", "4 ... -0.052585 -0.196108 0.245469 0.00784 0.093595 -0.303418 \n", "\n", " 206414_s_at 204978_at 205379_at 203897_at \n", "0 0.498778 0.352152 0.53591 0.52484 \n", "1 -0.225122 0.533052 -0.19329 -0.25436 \n", "2 -0.408722 -0.057177 -0.25699 0.31004 \n", "3 -0.251122 0.040732 0.17091 -0.05144 \n", "4 0.474878 -0.022947 -0.28589 0.06423 \n", "\n", "[5 rows x 980 columns]" ] }, "execution_count": 2, "metadata": {}, "output_type": "execute_result" } ], "source": [ "CP_path = './CP_data.parquet'\n", "GE_path = './GE_data.parquet'\n", "\n", "CP_CSV = pd.read_parquet(CP_path)\n", "GE_CSV = pd.read_parquet(GE_path)\n", "print(CP_CSV.shape, GE_CSV.shape)\n", "GE_CSV.head()" ] }, { "cell_type": "code", "execution_count": 4, "metadata": {}, "outputs": [ { "data": { "text/html": [ "
\n", "\n", "\n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", "
Metadata_moaMetadata_mmoles_per_liter2SMILESMetadata_PlateCells_AreaShape_CompactnessCells_AreaShape_EulerNumberCells_AreaShape_ExtentCells_AreaShape_FormFactorCells_AreaShape_MaxFeretDiameterCells_AreaShape_MaximumRadius...Nuclei_Texture_SumEntropy_DNA_5_0Nuclei_Texture_SumEntropy_ER_3_0Nuclei_Texture_SumEntropy_Mito_3_0Nuclei_Texture_SumEntropy_RNA_5_0Nuclei_Texture_SumVariance_ER_3_0Nuclei_Texture_SumVariance_Mito_5_0Nuclei_Texture_SumVariance_RNA_10_0Nuclei_Texture_Variance_AGP_5_0Nuclei_Texture_Variance_ER_5_0Nuclei_Texture_Variance_RNA_5_0
0local anesthetic6.05CCCOc1cc(N)ccc1C(=O)OCCN(CC)CC242770.256710NaN-0.114990-0.2370900.6155940.550434...0.0900210.375747-0.0104270.118997-0.156442-0.018014-0.024573-0.383890-0.199967-0.075658
1cannabinoid receptor antagonist10.00COc1cc(C)cc(OC)c1[C@@H]1C=C(C)CC[C@H]1C(C)=C242770.050962NaN0.051053-0.0337330.2719530.172714...-0.0303380.390687-0.0585600.065822-0.113721-0.089807-0.041392-0.248967-0.1261830.018563
2None10.00COc1cc(O)cc(\\C=C\\c2ccccc2)c124277-0.098185NaN0.0442120.1646680.0119850.127098...0.1029460.2791480.0298910.079777-0.0079790.081155-0.0896330.043140-0.0343550.052914
3interleukin inhibitor10.00COc1c(O)cc2C(=O)O[C@H]3[C@@H](O)[C@H](O)[C@@H]...242770.410722NaN-0.236208-0.0550840.6946480.516785...0.4705160.2147920.0154080.339100-0.034251-0.0475970.103611-0.139579-0.0422410.249977
4bacterial cell wall synthesis inhibitor4.39CCOc1ccc2ccccc2c1C(=O)N[C@H]1[C@H]2SC(C)(C)[C@...242770.248559NaN-0.0200580.0857720.6787420.671897...0.6717900.4241860.1188730.4156230.2575980.1933700.238471-0.0337970.0800180.335890
\n", "

5 rows × 898 columns

\n", "
" ], "text/plain": [ " Metadata_moa Metadata_mmoles_per_liter2 \\\n", "0 local anesthetic 6.05 \n", "1 cannabinoid receptor antagonist 10.00 \n", "2 None 10.00 \n", "3 interleukin inhibitor 10.00 \n", "4 bacterial cell wall synthesis inhibitor 4.39 \n", "\n", " SMILES Metadata_Plate \\\n", "0 CCCOc1cc(N)ccc1C(=O)OCCN(CC)CC 24277 \n", "1 COc1cc(C)cc(OC)c1[C@@H]1C=C(C)CC[C@H]1C(C)=C 24277 \n", "2 COc1cc(O)cc(\\C=C\\c2ccccc2)c1 24277 \n", "3 COc1c(O)cc2C(=O)O[C@H]3[C@@H](O)[C@H](O)[C@@H]... 24277 \n", "4 CCOc1ccc2ccccc2c1C(=O)N[C@H]1[C@H]2SC(C)(C)[C@... 24277 \n", "\n", " Cells_AreaShape_Compactness Cells_AreaShape_EulerNumber \\\n", "0 0.256710 NaN \n", "1 0.050962 NaN \n", "2 -0.098185 NaN \n", "3 0.410722 NaN \n", "4 0.248559 NaN \n", "\n", " Cells_AreaShape_Extent Cells_AreaShape_FormFactor \\\n", "0 -0.114990 -0.237090 \n", "1 0.051053 -0.033733 \n", "2 0.044212 0.164668 \n", "3 -0.236208 -0.055084 \n", "4 -0.020058 0.085772 \n", "\n", " Cells_AreaShape_MaxFeretDiameter Cells_AreaShape_MaximumRadius ... \\\n", "0 0.615594 0.550434 ... \n", "1 0.271953 0.172714 ... \n", "2 0.011985 0.127098 ... \n", "3 0.694648 0.516785 ... \n", "4 0.678742 0.671897 ... \n", "\n", " Nuclei_Texture_SumEntropy_DNA_5_0 Nuclei_Texture_SumEntropy_ER_3_0 \\\n", "0 0.090021 0.375747 \n", "1 -0.030338 0.390687 \n", "2 0.102946 0.279148 \n", "3 0.470516 0.214792 \n", "4 0.671790 0.424186 \n", "\n", " Nuclei_Texture_SumEntropy_Mito_3_0 Nuclei_Texture_SumEntropy_RNA_5_0 \\\n", "0 -0.010427 0.118997 \n", "1 -0.058560 0.065822 \n", "2 0.029891 0.079777 \n", "3 0.015408 0.339100 \n", "4 0.118873 0.415623 \n", "\n", " Nuclei_Texture_SumVariance_ER_3_0 Nuclei_Texture_SumVariance_Mito_5_0 \\\n", "0 -0.156442 -0.018014 \n", "1 -0.113721 -0.089807 \n", "2 -0.007979 0.081155 \n", "3 -0.034251 -0.047597 \n", "4 0.257598 0.193370 \n", "\n", " Nuclei_Texture_SumVariance_RNA_10_0 Nuclei_Texture_Variance_AGP_5_0 \\\n", "0 -0.024573 -0.383890 \n", "1 -0.041392 -0.248967 \n", "2 -0.089633 0.043140 \n", "3 0.103611 -0.139579 \n", "4 0.238471 -0.033797 \n", "\n", " Nuclei_Texture_Variance_ER_5_0 Nuclei_Texture_Variance_RNA_5_0 \n", "0 -0.199967 -0.075658 \n", "1 -0.126183 0.018563 \n", "2 -0.034355 0.052914 \n", "3 -0.042241 0.249977 \n", "4 0.080018 0.335890 \n", "\n", "[5 rows x 898 columns]" ] }, "execution_count": 4, "metadata": {}, "output_type": "execute_result" } ], "source": [ "CP_CSV.head()" ] }, { "cell_type": "code", "execution_count": 5, "metadata": {}, "outputs": [ { "data": { "text/plain": [ "((20342,), (21782,))" ] }, "execution_count": 5, "metadata": {}, "output_type": "execute_result" } ], "source": [ "CP_CSV['SMILES'].unique().shape, GE_CSV['CPD_SMILES'].unique().shape # 包含 NaN的" ] }, { "cell_type": "code", "execution_count": 6, "metadata": {}, "outputs": [ { "data": { "text/plain": [ "((20342,), (21782,))" ] }, "execution_count": 6, "metadata": {}, "output_type": "execute_result" } ], "source": [ "# 填充nan为 DMSO\n", "CP_CSV['SMILES'] = CP_CSV['SMILES'].fillna('DMSO')\n", "GE_CSV['CPD_SMILES'] = GE_CSV['CPD_SMILES'].fillna('DMSO')\n", "\n", "CP_CSV['SMILES'].unique().shape, GE_CSV['CPD_SMILES'].unique().shape # 包含 NaN的" ] }, { "cell_type": "markdown", "metadata": {}, "source": [ "# 确保这两个CSV包含相同的SMILES" ] }, { "cell_type": "code", "execution_count": null, "metadata": {}, "outputs": [ { "name": "stdout", "output_type": "stream", "text": [ "交集数量: 20342\n", "示例交集: ['CCOC(=O)c1nc(Nc2ccc(Nc3ccccc3)cc2)c2ccccc2n1', 'COC(=O)C[C@@H]1C[C@@H]2[C@@H](Oc3ccc(NC(=O)C4CCOCC4)cc23)[C@H](CO)O1', 'CN(C)c1ccc2O[C@@H]3[C@@H](C[C@@H](CC(=O)NCc4ccccn4)O[C@@H]3CO)c2c1', 'CCN(CC)CCOc1ccc2-c3ccc(OCCN(CC)CC)cc3C(=O)c2c1', 'CO[C@@H]1CN(C)C(=O)c2ccc(NC(C)=O)cc2OC[C@@H](C)N(C[C@H]1C)C(=O)CN(C)C', 'COC(=O)[C@@H]1[C@@H](CO)[C@@H]2Cn3c(=O)c(ccc3[C@@H]2N1C(=O)C1CCC1)-c1ccc(F)cc1', 'OC[C@H]1O[C@H](CCn2cc(nn2)C2CC2)CC[C@H]1NC(=O)c1cccnc1', 'CCCC(=O)Nc1ccc2c(OC[C@H](C)N(Cc3ccccn3)C[C@@H](C)[C@H](CN(C)C2=O)OC)c1', 'Cc1cccc(C)c1OC(=O)CSc1nnc(o1)-c1ccccc1O', 'C(CN1CCOCC1)Nc1cc2CCCc3ccccc3-c2nn1']\n" ] } ], "source": [ "# 转换成集合\n", "set_cp = set(CP_CSV['SMILES'].unique())\n", "set_ge = set(GE_CSV['CPD_SMILES'].dropna().unique())\n", "\n", "# 求交集\n", "common_smiles = set_cp & set_ge\n", "\n", "print(f\"交集数量: {len(common_smiles)}\")\n", "print(\"示例交集:\", list(common_smiles)[:3]) # 只看前10个" ] }, { "cell_type": "code", "execution_count": null, "metadata": {}, "outputs": [ { "name": "stdout", "output_type": "stream", "text": [ "(61313, 980) (113416, 898)\n" ] } ], "source": [ "# 删除Gene中不在交集的行\n", "GE_CSV_matched = GE_CSV[GE_CSV['CPD_SMILES'].isin(common_smiles)].copy()\n", "# 删除CellPainting中不在交集的行\n", "CP_CSV_matched = CP_CSV[CP_CSV['SMILES'].isin(common_smiles)].copy()\n", "print(GE_CSV_matched.shape, CP_CSV_matched.shape) # (113416, 898) (68120, 980)" ] }, { "cell_type": "code", "execution_count": 10, "metadata": {}, "outputs": [], "source": [ "# 更新文件， 后续可以用来做数据填补\n", "\n", "CP_CSV_matched.to_parquet('./CP_data.parquet', index=False)\n", "GE_CSV_matched.to_parquet('./GE_data.parquet', index=False)" ] }, { "cell_type": "markdown", "metadata": {}, "source": [ "## 数据对齐\n", "\n", "### 对齐逻辑：数据清洗，各自生成对齐数据即可" ] }, { "cell_type": "code", "execution_count": 9, "metadata": {}, "outputs": [ { "name": "stdout", "output_type": "stream", "text": [ "DataFrame 中存在 NaN 值\n", "包含 NaN 的列: 194\n", "包含 NaN 的行:\n", " 113416\n", "整个 DataFrame 中 NaN 的总数: 10681272\n" ] } ], "source": [ "# 检查整个 DataFrame 中是否存在 NaN\n", "if CP_CSV_matched.isna().any().any():\n", " print(\"DataFrame 中存在 NaN 值\")\n", "else:\n", " print(\"DataFrame 中不存在 NaN 值\")\n", "\n", "# 检查每一列是否存在 NaN\n", "nan_columns = CP_CSV_matched.columns[CP_CSV_matched.isna().any()].tolist()\n", "print(\"包含 NaN 的列:\", len(nan_columns))\n", "\n", "# 检查每一行是否存在 NaN\n", "nan_rows = CP_CSV_matched[CP_CSV_matched.isna().any(axis=1)]\n", "print(\"包含 NaN 的行:\\n\", len(nan_rows))\n", "\n", "# 检查整个 DataFrame 中 NaN 的总数\n", "total_nans = CP_CSV_matched.isna().sum().sum()\n", "print(f\"整个 DataFrame 中 NaN 的总数: {total_nans}\")" ] }, { "cell_type": "code", "execution_count": 11, "metadata": {}, "outputs": [ { "name": "stdout", "output_type": "stream", "text": [ "清洗后 DataFrame 形状: (113416, 778)\n", "NaN 剩余数量: 666550\n" ] } ], "source": [ "import pandas as pd\n", "\n", "# 假设 CP_CSV_matched 已经加载好了\n", "df = CP_CSV_matched.copy()\n", "\n", "# 1. 删除 NaN 占比 > 20% 的列\n", "col_thresh = 0.2\n", "row_thresh = 0.2\n", "\n", "# 列筛选\n", "valid_cols = df.columns[df.isnull().mean() <= col_thresh]\n", "\n", "# 行筛选\n", "valid_rows = df.index[df.isnull().mean(axis=1) <= row_thresh]\n", "\n", "# 保留行和列\n", "df = df.loc[valid_rows, valid_cols]\n", "\n", "print(f\"清洗后 DataFrame 形状: {df.shape}\")\n", "print(\"NaN 剩余数量:\", df.isnull().sum().sum())" ] }, { "cell_type": "code", "execution_count": 12, "metadata": {}, "outputs": [ { "name": "stdout", "output_type": "stream", "text": [ "(113416, 778) (113416, 898)\n" ] } ], "source": [ "print(df.shape, CP_CSV_matched.shape)" ] }, { "cell_type": "code", "execution_count": 13, "metadata": {}, "outputs": [ { "name": "stdout", "output_type": "stream", "text": [ "NaN 剩余数量: 666550\n" ] } ], "source": [ "print(\"NaN 剩余数量:\", df.isnull().sum().sum())" ] }, { "cell_type": "code", "execution_count": 41, "metadata": {}, "outputs": [ { "data": { "text/html": [ "
\n", "\n", "\n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", "
Metadata_moaMetadata_mmoles_per_liter2SMILESMetadata_PlateCells_AreaShape_CompactnessCells_AreaShape_EulerNumberCells_AreaShape_ExtentCells_AreaShape_FormFactorCells_AreaShape_MaxFeretDiameterCells_AreaShape_MaximumRadius...Nuclei_Texture_SumEntropy_DNA_5_0Nuclei_Texture_SumEntropy_ER_3_0Nuclei_Texture_SumEntropy_Mito_3_0Nuclei_Texture_SumEntropy_RNA_5_0Nuclei_Texture_SumVariance_ER_3_0Nuclei_Texture_SumVariance_Mito_5_0Nuclei_Texture_SumVariance_RNA_10_0Nuclei_Texture_Variance_AGP_5_0Nuclei_Texture_Variance_ER_5_0Nuclei_Texture_Variance_RNA_5_0
0local anesthetic6.05CCCOc1cc(N)ccc1C(=O)OCCN(CC)CC242770.256710NaN-0.114990-0.2370900.6155940.550434...0.0900210.375747-0.0104270.118997-0.156442-0.018014-0.024573-0.383890-0.199967-0.075658
1cannabinoid receptor antagonist10.00COc1cc(C)cc(OC)c1[C@@H]1C=C(C)CC[C@H]1C(C)=C242770.050962NaN0.051053-0.0337330.2719530.172714...-0.0303380.390687-0.0585600.065822-0.113721-0.089807-0.041392-0.248967-0.1261830.018563
2None10.00COc1cc(O)cc(\\C=C\\c2ccccc2)c124277-0.098185NaN0.0442120.1646680.0119850.127098...0.1029460.2791480.0298910.079777-0.0079790.081155-0.0896330.043140-0.0343550.052914
3interleukin inhibitor10.00COc1c(O)cc2C(=O)O[C@H]3[C@@H](O)[C@H](O)[C@@H]...242770.410722NaN-0.236208-0.0550840.6946480.516785...0.4705160.2147920.0154080.339100-0.034251-0.0475970.103611-0.139579-0.0422410.249977
4bacterial cell wall synthesis inhibitor4.39CCOc1ccc2ccccc2c1C(=O)N[C@H]1[C@H]2SC(C)(C)[C@...242770.248559NaN-0.0200580.0857720.6787420.671897...0.6717900.4241860.1188730.4156230.2575980.1933700.238471-0.0337970.0800180.335890
\n", "

5 rows × 898 columns

\n", "
" ], "text/plain": [ " Metadata_moa Metadata_mmoles_per_liter2 \\\n", "0 local anesthetic 6.05 \n", "1 cannabinoid receptor antagonist 10.00 \n", "2 None 10.00 \n", "3 interleukin inhibitor 10.00 \n", "4 bacterial cell wall synthesis inhibitor 4.39 \n", "\n", " SMILES Metadata_Plate \\\n", "0 CCCOc1cc(N)ccc1C(=O)OCCN(CC)CC 24277 \n", "1 COc1cc(C)cc(OC)c1[C@@H]1C=C(C)CC[C@H]1C(C)=C 24277 \n", "2 COc1cc(O)cc(\\C=C\\c2ccccc2)c1 24277 \n", "3 COc1c(O)cc2C(=O)O[C@H]3[C@@H](O)[C@H](O)[C@@H]... 24277 \n", "4 CCOc1ccc2ccccc2c1C(=O)N[C@H]1[C@H]2SC(C)(C)[C@... 24277 \n", "\n", " Cells_AreaShape_Compactness Cells_AreaShape_EulerNumber \\\n", "0 0.256710 NaN \n", "1 0.050962 NaN \n", "2 -0.098185 NaN \n", "3 0.410722 NaN \n", "4 0.248559 NaN \n", "\n", " Cells_AreaShape_Extent Cells_AreaShape_FormFactor \\\n", "0 -0.114990 -0.237090 \n", "1 0.051053 -0.033733 \n", "2 0.044212 0.164668 \n", "3 -0.236208 -0.055084 \n", "4 -0.020058 0.085772 \n", "\n", " Cells_AreaShape_MaxFeretDiameter Cells_AreaShape_MaximumRadius ... \\\n", "0 0.615594 0.550434 ... \n", "1 0.271953 0.172714 ... \n", "2 0.011985 0.127098 ... \n", "3 0.694648 0.516785 ... \n", "4 0.678742 0.671897 ... \n", "\n", " Nuclei_Texture_SumEntropy_DNA_5_0 Nuclei_Texture_SumEntropy_ER_3_0 \\\n", "0 0.090021 0.375747 \n", "1 -0.030338 0.390687 \n", "2 0.102946 0.279148 \n", "3 0.470516 0.214792 \n", "4 0.671790 0.424186 \n", "\n", " Nuclei_Texture_SumEntropy_Mito_3_0 Nuclei_Texture_SumEntropy_RNA_5_0 \\\n", "0 -0.010427 0.118997 \n", "1 -0.058560 0.065822 \n", "2 0.029891 0.079777 \n", "3 0.015408 0.339100 \n", "4 0.118873 0.415623 \n", "\n", " Nuclei_Texture_SumVariance_ER_3_0 Nuclei_Texture_SumVariance_Mito_5_0 \\\n", "0 -0.156442 -0.018014 \n", "1 -0.113721 -0.089807 \n", "2 -0.007979 0.081155 \n", "3 -0.034251 -0.047597 \n", "4 0.257598 0.193370 \n", "\n", " Nuclei_Texture_SumVariance_RNA_10_0 Nuclei_Texture_Variance_AGP_5_0 \\\n", "0 -0.024573 -0.383890 \n", "1 -0.041392 -0.248967 \n", "2 -0.089633 0.043140 \n", "3 0.103611 -0.139579 \n", "4 0.238471 -0.033797 \n", "\n", " Nuclei_Texture_Variance_ER_5_0 Nuclei_Texture_Variance_RNA_5_0 \n", "0 -0.199967 -0.075658 \n", "1 -0.126183 0.018563 \n", "2 -0.034355 0.052914 \n", "3 -0.042241 0.249977 \n", "4 0.080018 0.335890 \n", "\n", "[5 rows x 898 columns]" ] }, "execution_count": 41, "metadata": {}, "output_type": "execute_result" } ], "source": [ "CP_CSV_matched.head()\n", "\n", "# nan_count = CP_CSV_matched['Metadata_mmoles_per_liter2'].isna().sum()\n", "# print(f\"NaN count in SMILES: {nan_count}\")" ] }, { "cell_type": "code", "execution_count": 42, "metadata": {}, "outputs": [ { "name": "stdout", "output_type": "stream", "text": [ "清洗后 DataFrame 形状: (113416, 778)\n", "NaN 剩余数量 (第一轮清洗后): 666550\n", "DataFrame 最终填充完成。\n", "NaN 剩余数量 (最终): 0\n", "最终 DataFrame 形状: (113416, 778)\n" ] } ], "source": [ "import pandas as pd\n", "import numpy as np\n", "\n", "# 你的初始数据清洗步骤\n", "# 假设 CP_CSV_matched 已经加载好了\n", "df = CP_CSV_matched.copy()\n", "\n", "# 1. 删除 NaN 占比 > 20% 的列和行\n", "col_thresh = 0.2\n", "row_thresh = 0.2\n", "\n", "# 列筛选\n", "valid_cols = df.columns[df.isnull().mean() <= col_thresh]\n", "df = df[valid_cols]\n", "\n", "# 行筛选\n", "valid_rows = df.index[df.isnull().mean(axis=1) <= row_thresh]\n", "df = df.loc[valid_rows]\n", "\n", "print(f\"清洗后 DataFrame 形状: {df.shape}\")\n", "print(\"NaN 剩余数量 (第一轮清洗后):\", df.isnull().sum().sum())\n", "\n", "# --- 最终填充方案 ---\n", "\n", "# 2. 强制转换所有列为数值类型\n", "# `errors='coerce'` 会将任何非数值数据（如空字符串、'-'）强制转换为 NaN\n", "# 这一步是关键，它将所有需要填充的数据统一为 NaN\n", "# 我们需要保留 SMILES 列，因为它不是数值\n", "smiles_col = 'SMILES' # 假设 SMILES 列名为 canonical_smiles\n", "if smiles_col in df.columns:\n", " df_numeric = df.drop(columns=[smiles_col]).apply(pd.to_numeric, errors='coerce')\n", " df_numeric[smiles_col] = df[smiles_col]\n", "else:\n", " df_numeric = df.apply(pd.to_numeric, errors='coerce')\n", "\n", "# 3. 用中位数填充剩余的 NaN\n", "# 中位数对异常值不敏感，是填充数据的好选择\n", "# fillna() 会用 Series 的中位数来填充其 NaN 值\n", "# .median() 会自动跳过 NaN 值进行计算\n", "df_filled = df_numeric.fillna(df_numeric.median(numeric_only=True))\n", "\n", "# 4. 再次检查，确保没有剩余的 NaN\n", "# 如果有某一列在转换后全变成了 NaN（比如该列原本都是非数值），那么它的中位数也会是 NaN\n", "# 我们可以将这些剩余的 NaN 填充为 0\n", "df_filled = df_filled.fillna(0)\n", "\n", "print(\"DataFrame 最终填充完成。\")\n", "print(\"NaN 剩余数量 (最终):\", df_filled.isnull().sum().sum())\n", "print(\"最终 DataFrame 形状:\", df_filled.shape)" ] }, { "cell_type": "code", "execution_count": 43, "metadata": {}, "outputs": [ { "data": { "text/html": [ "
\n", "\n", "\n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", "
Metadata_mmoles_per_liter2Metadata_PlateCells_AreaShape_CompactnessCells_AreaShape_ExtentCells_AreaShape_FormFactorCells_AreaShape_MaxFeretDiameterCells_AreaShape_MaximumRadiusCells_AreaShape_SolidityCells_AreaShape_Zernike_0_0Cells_AreaShape_Zernike_1_1...Nuclei_Texture_SumEntropy_ER_3_0Nuclei_Texture_SumEntropy_Mito_3_0Nuclei_Texture_SumEntropy_RNA_5_0Nuclei_Texture_SumVariance_ER_3_0Nuclei_Texture_SumVariance_Mito_5_0Nuclei_Texture_SumVariance_RNA_10_0Nuclei_Texture_Variance_AGP_5_0Nuclei_Texture_Variance_ER_5_0Nuclei_Texture_Variance_RNA_5_0SMILES
06.05242770.256710-0.114990-0.2370900.6155940.5504340.072795-0.2109080.008488...0.375747-0.0104270.118997-0.156442-0.018014-0.024573-0.383890-0.199967-0.075658CCCOc1cc(N)ccc1C(=O)OCCN(CC)CC
110.00242770.0509620.051053-0.0337330.2719530.1727140.145941-0.039057-0.070826...0.390687-0.0585600.065822-0.113721-0.089807-0.041392-0.248967-0.1261830.018563COc1cc(C)cc(OC)c1[C@@H]1C=C(C)CC[C@H]1C(C)=C
210.0024277-0.0981850.0442120.1646680.0119850.1270980.0947360.0782570.119333...0.2791480.0298910.079777-0.0079790.081155-0.0896330.043140-0.0343550.052914COc1cc(O)cc(\\C=C\\c2ccccc2)c1
310.00242770.410722-0.236208-0.0550840.6946480.5167850.154239-0.347572-0.172400...0.2147920.0154080.339100-0.034251-0.0475970.103611-0.139579-0.0422410.249977COc1c(O)cc2C(=O)O[C@H]3[C@@H](O)[C@H](O)[C@@H]...
44.39242770.248559-0.0200580.0857720.6787420.6718970.214587-0.221396-0.086168...0.4241860.1188730.4156230.2575980.1933700.238471-0.0337970.0800180.335890CCOc1ccc2ccccc2c1C(=O)N[C@H]1[C@H]2SC(C)(C)[C@...
\n", "

5 rows × 778 columns

\n", "
" ], "text/plain": [ " Metadata_mmoles_per_liter2 Metadata_Plate Cells_AreaShape_Compactness \\\n", "0 6.05 24277 0.256710 \n", "1 10.00 24277 0.050962 \n", "2 10.00 24277 -0.098185 \n", "3 10.00 24277 0.410722 \n", "4 4.39 24277 0.248559 \n", "\n", " Cells_AreaShape_Extent Cells_AreaShape_FormFactor \\\n", "0 -0.114990 -0.237090 \n", "1 0.051053 -0.033733 \n", "2 0.044212 0.164668 \n", "3 -0.236208 -0.055084 \n", "4 -0.020058 0.085772 \n", "\n", " Cells_AreaShape_MaxFeretDiameter Cells_AreaShape_MaximumRadius \\\n", "0 0.615594 0.550434 \n", "1 0.271953 0.172714 \n", "2 0.011985 0.127098 \n", "3 0.694648 0.516785 \n", "4 0.678742 0.671897 \n", "\n", " Cells_AreaShape_Solidity Cells_AreaShape_Zernike_0_0 \\\n", "0 0.072795 -0.210908 \n", "1 0.145941 -0.039057 \n", "2 0.094736 0.078257 \n", "3 0.154239 -0.347572 \n", "4 0.214587 -0.221396 \n", "\n", " Cells_AreaShape_Zernike_1_1 ... Nuclei_Texture_SumEntropy_ER_3_0 \\\n", "0 0.008488 ... 0.375747 \n", "1 -0.070826 ... 0.390687 \n", "2 0.119333 ... 0.279148 \n", "3 -0.172400 ... 0.214792 \n", "4 -0.086168 ... 0.424186 \n", "\n", " Nuclei_Texture_SumEntropy_Mito_3_0 Nuclei_Texture_SumEntropy_RNA_5_0 \\\n", "0 -0.010427 0.118997 \n", "1 -0.058560 0.065822 \n", "2 0.029891 0.079777 \n", "3 0.015408 0.339100 \n", "4 0.118873 0.415623 \n", "\n", " Nuclei_Texture_SumVariance_ER_3_0 Nuclei_Texture_SumVariance_Mito_5_0 \\\n", "0 -0.156442 -0.018014 \n", "1 -0.113721 -0.089807 \n", "2 -0.007979 0.081155 \n", "3 -0.034251 -0.047597 \n", "4 0.257598 0.193370 \n", "\n", " Nuclei_Texture_SumVariance_RNA_10_0 Nuclei_Texture_Variance_AGP_5_0 \\\n", "0 -0.024573 -0.383890 \n", "1 -0.041392 -0.248967 \n", "2 -0.089633 0.043140 \n", "3 0.103611 -0.139579 \n", "4 0.238471 -0.033797 \n", "\n", " Nuclei_Texture_Variance_ER_5_0 Nuclei_Texture_Variance_RNA_5_0 \\\n", "0 -0.199967 -0.075658 \n", "1 -0.126183 0.018563 \n", "2 -0.034355 0.052914 \n", "3 -0.042241 0.249977 \n", "4 0.080018 0.335890 \n", "\n", " SMILES \n", "0 CCCOc1cc(N)ccc1C(=O)OCCN(CC)CC \n", "1 COc1cc(C)cc(OC)c1[C@@H]1C=C(C)CC[C@H]1C(C)=C \n", "2 COc1cc(O)cc(\\C=C\\c2ccccc2)c1 \n", "3 COc1c(O)cc2C(=O)O[C@H]3[C@@H](O)[C@H](O)[C@@H]... \n", "4 CCOc1ccc2ccccc2c1C(=O)N[C@H]1[C@H]2SC(C)(C)[C@... \n", "\n", "[5 rows x 778 columns]" ] }, "execution_count": 43, "metadata": {}, "output_type": "execute_result" } ], "source": [ "df_filled.head()" ] }, { "cell_type": "code", "execution_count": 44, "metadata": {}, "outputs": [ { "data": { "text/html": [ "
\n", "\n", "\n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", "
SMILESMetadata_mmoles_per_liter2Metadata_PlateCells_AreaShape_CompactnessCells_AreaShape_ExtentCells_AreaShape_FormFactorCells_AreaShape_MaxFeretDiameterCells_AreaShape_MaximumRadiusCells_AreaShape_SolidityCells_AreaShape_Zernike_0_0...Nuclei_Texture_SumEntropy_DNA_5_0Nuclei_Texture_SumEntropy_ER_3_0Nuclei_Texture_SumEntropy_Mito_3_0Nuclei_Texture_SumEntropy_RNA_5_0Nuclei_Texture_SumVariance_ER_3_0Nuclei_Texture_SumVariance_Mito_5_0Nuclei_Texture_SumVariance_RNA_10_0Nuclei_Texture_Variance_AGP_5_0Nuclei_Texture_Variance_ER_5_0Nuclei_Texture_Variance_RNA_5_0
0CCCOc1cc(N)ccc1C(=O)OCCN(CC)CC6.05242770.256710-0.114990-0.2370900.6155940.5504340.072795-0.210908...0.0900210.375747-0.0104270.118997-0.156442-0.018014-0.024573-0.383890-0.199967-0.075658
1COc1cc(C)cc(OC)c1[C@@H]1C=C(C)CC[C@H]1C(C)=C10.00242770.0509620.051053-0.0337330.2719530.1727140.145941-0.039057...-0.0303380.390687-0.0585600.065822-0.113721-0.089807-0.041392-0.248967-0.1261830.018563
2COc1cc(O)cc(\\C=C\\c2ccccc2)c110.0024277-0.0981850.0442120.1646680.0119850.1270980.0947360.078257...0.1029460.2791480.0298910.079777-0.0079790.081155-0.0896330.043140-0.0343550.052914
3COc1c(O)cc2C(=O)O[C@H]3[C@@H](O)[C@H](O)[C@@H]...10.00242770.410722-0.236208-0.0550840.6946480.5167850.154239-0.347572...0.4705160.2147920.0154080.339100-0.034251-0.0475970.103611-0.139579-0.0422410.249977
4CCOc1ccc2ccccc2c1C(=O)N[C@H]1[C@H]2SC(C)(C)[C@...4.39242770.248559-0.0200580.0857720.6787420.6718970.214587-0.221396...0.6717900.4241860.1188730.4156230.2575980.1933700.238471-0.0337970.0800180.335890
\n", "

5 rows × 778 columns

\n", "
" ], "text/plain": [ " SMILES \\\n", "0 CCCOc1cc(N)ccc1C(=O)OCCN(CC)CC \n", "1 COc1cc(C)cc(OC)c1[C@@H]1C=C(C)CC[C@H]1C(C)=C \n", "2 COc1cc(O)cc(\\C=C\\c2ccccc2)c1 \n", "3 COc1c(O)cc2C(=O)O[C@H]3[C@@H](O)[C@H](O)[C@@H]... \n", "4 CCOc1ccc2ccccc2c1C(=O)N[C@H]1[C@H]2SC(C)(C)[C@... \n", "\n", " Metadata_mmoles_per_liter2 Metadata_Plate Cells_AreaShape_Compactness \\\n", "0 6.05 24277 0.256710 \n", "1 10.00 24277 0.050962 \n", "2 10.00 24277 -0.098185 \n", "3 10.00 24277 0.410722 \n", "4 4.39 24277 0.248559 \n", "\n", " Cells_AreaShape_Extent Cells_AreaShape_FormFactor \\\n", "0 -0.114990 -0.237090 \n", "1 0.051053 -0.033733 \n", "2 0.044212 0.164668 \n", "3 -0.236208 -0.055084 \n", "4 -0.020058 0.085772 \n", "\n", " Cells_AreaShape_MaxFeretDiameter Cells_AreaShape_MaximumRadius \\\n", "0 0.615594 0.550434 \n", "1 0.271953 0.172714 \n", "2 0.011985 0.127098 \n", "3 0.694648 0.516785 \n", "4 0.678742 0.671897 \n", "\n", " Cells_AreaShape_Solidity Cells_AreaShape_Zernike_0_0 ... \\\n", "0 0.072795 -0.210908 ... \n", "1 0.145941 -0.039057 ... \n", "2 0.094736 0.078257 ... \n", "3 0.154239 -0.347572 ... \n", "4 0.214587 -0.221396 ... \n", "\n", " Nuclei_Texture_SumEntropy_DNA_5_0 Nuclei_Texture_SumEntropy_ER_3_0 \\\n", "0 0.090021 0.375747 \n", "1 -0.030338 0.390687 \n", "2 0.102946 0.279148 \n", "3 0.470516 0.214792 \n", "4 0.671790 0.424186 \n", "\n", " Nuclei_Texture_SumEntropy_Mito_3_0 Nuclei_Texture_SumEntropy_RNA_5_0 \\\n", "0 -0.010427 0.118997 \n", "1 -0.058560 0.065822 \n", "2 0.029891 0.079777 \n", "3 0.015408 0.339100 \n", "4 0.118873 0.415623 \n", "\n", " Nuclei_Texture_SumVariance_ER_3_0 Nuclei_Texture_SumVariance_Mito_5_0 \\\n", "0 -0.156442 -0.018014 \n", "1 -0.113721 -0.089807 \n", "2 -0.007979 0.081155 \n", "3 -0.034251 -0.047597 \n", "4 0.257598 0.193370 \n", "\n", " Nuclei_Texture_SumVariance_RNA_10_0 Nuclei_Texture_Variance_AGP_5_0 \\\n", "0 -0.024573 -0.383890 \n", "1 -0.041392 -0.248967 \n", "2 -0.089633 0.043140 \n", "3 0.103611 -0.139579 \n", "4 0.238471 -0.033797 \n", "\n", " Nuclei_Texture_Variance_ER_5_0 Nuclei_Texture_Variance_RNA_5_0 \n", "0 -0.199967 -0.075658 \n", "1 -0.126183 0.018563 \n", "2 -0.034355 0.052914 \n", "3 -0.042241 0.249977 \n", "4 0.080018 0.335890 \n", "\n", "[5 rows x 778 columns]" ] }, "execution_count": 44, "metadata": {}, "output_type": "execute_result" } ], "source": [ "# SMILES列移动到第一列\n", "\n", "if 'SMILES' in df_filled.columns:\n", " cols = df_filled.columns.tolist()\n", " cols.insert(0, cols.pop(cols.index('SMILES')))\n", " df_filled = df_filled[cols]\n", "\n", "df_filled.head()" ] }, { "cell_type": "markdown", "metadata": {}, "source": [ "# 规范化SMILES" ] }, { "cell_type": "code", "execution_count": 45, "metadata": {}, "outputs": [ { "name": "stdout", "output_type": "stream", "text": [ "最终 SMILES 唯一值数量: 20342\n" ] } ], "source": [ "final_smiles = df_filled['SMILES'].unique()\n", "print(f\"最终 SMILES 唯一值数量: {len(final_smiles)}\")" ] }, { "cell_type": "code", "execution_count": 46, "metadata": {}, "outputs": [ { "name": "stdout", "output_type": "stream", "text": [ "最终 SMILES 唯一值数量: 20342\n" ] } ], "source": [ "import pandas as pd\n", "from rdkit import Chem\n", "\n", "def canonicalize_smiles(smiles):\n", " \"\"\"尝试将一个 SMILES 转换为规范化形式，忽略 DMSO\"\"\"\n", " if pd.isna(smiles): \n", " return smiles\n", " if str(smiles).upper() == \"DMSO\": # 特殊情况，直接跳过\n", " return \"DMSO\"\n", " try:\n", " mol = Chem.MolFromSmiles(smiles)\n", " if mol is not None:\n", " return Chem.MolToSmiles(mol)\n", " else:\n", " return smiles\n", " except:\n", " return smiles\n", "\n", "# 应用到整列\n", "df_filled['SMILES'] = df_filled['SMILES'].apply(canonicalize_smiles)\n", "\n", "# 查看结果\n", "final_smiles = df_filled['SMILES'].unique()\n", "print(f\"最终 SMILES 唯一值数量: {len(final_smiles)}\")\n" ] }, { "cell_type": "code", "execution_count": 49, "metadata": {}, "outputs": [ { "data": { "text/html": [ "
\n", "\n", "\n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", "
SMILESpert_dosePlateCells_AreaShape_CompactnessCells_AreaShape_ExtentCells_AreaShape_FormFactorCells_AreaShape_MaxFeretDiameterCells_AreaShape_MaximumRadiusCells_AreaShape_SolidityCells_AreaShape_Zernike_0_0...Nuclei_Texture_SumEntropy_DNA_5_0Nuclei_Texture_SumEntropy_ER_3_0Nuclei_Texture_SumEntropy_Mito_3_0Nuclei_Texture_SumEntropy_RNA_5_0Nuclei_Texture_SumVariance_ER_3_0Nuclei_Texture_SumVariance_Mito_5_0Nuclei_Texture_SumVariance_RNA_10_0Nuclei_Texture_Variance_AGP_5_0Nuclei_Texture_Variance_ER_5_0Nuclei_Texture_Variance_RNA_5_0
0CCCOc1cc(N)ccc1C(=O)OCCN(CC)CC6.05242770.256710-0.114990-0.2370900.6155940.5504340.072795-0.210908...0.0900210.375747-0.0104270.118997-0.156442-0.018014-0.024573-0.383890-0.199967-0.075658
1C=C(C)[C@@H]1CCC(C)=C[C@H]1c1c(OC)cc(C)cc1OC10.00242770.0509620.051053-0.0337330.2719530.1727140.145941-0.039057...-0.0303380.390687-0.0585600.065822-0.113721-0.089807-0.041392-0.248967-0.1261830.018563
2COc1cc(O)cc(/C=C/c2ccccc2)c110.0024277-0.0981850.0442120.1646680.0119850.1270980.0947360.078257...0.1029460.2791480.0298910.079777-0.0079790.081155-0.0896330.043140-0.0343550.052914
3COc1c(O)cc2c(c1O)[C@H]1O[C@H](CO)[C@@H](O)[C@H...10.00242770.410722-0.236208-0.0550840.6946480.5167850.154239-0.347572...0.4705160.2147920.0154080.339100-0.034251-0.0475970.103611-0.139579-0.0422410.249977
4CCOc1ccc2ccccc2c1C(=O)N[C@@H]1C(=O)N2[C@@H]1SC...4.39242770.248559-0.0200580.0857720.6787420.6718970.214587-0.221396...0.6717900.4241860.1188730.4156230.2575980.1933700.238471-0.0337970.0800180.335890
\n", "

5 rows × 778 columns

\n", "
" ], "text/plain": [ " SMILES pert_dose Plate \\\n", "0 CCCOc1cc(N)ccc1C(=O)OCCN(CC)CC 6.05 24277 \n", "1 C=C(C)[C@@H]1CCC(C)=C[C@H]1c1c(OC)cc(C)cc1OC 10.00 24277 \n", "2 COc1cc(O)cc(/C=C/c2ccccc2)c1 10.00 24277 \n", "3 COc1c(O)cc2c(c1O)[C@H]1O[C@H](CO)[C@@H](O)[C@H... 10.00 24277 \n", "4 CCOc1ccc2ccccc2c1C(=O)N[C@@H]1C(=O)N2[C@@H]1SC... 4.39 24277 \n", "\n", " Cells_AreaShape_Compactness Cells_AreaShape_Extent \\\n", "0 0.256710 -0.114990 \n", "1 0.050962 0.051053 \n", "2 -0.098185 0.044212 \n", "3 0.410722 -0.236208 \n", "4 0.248559 -0.020058 \n", "\n", " Cells_AreaShape_FormFactor Cells_AreaShape_MaxFeretDiameter \\\n", "0 -0.237090 0.615594 \n", "1 -0.033733 0.271953 \n", "2 0.164668 0.011985 \n", "3 -0.055084 0.694648 \n", "4 0.085772 0.678742 \n", "\n", " Cells_AreaShape_MaximumRadius Cells_AreaShape_Solidity \\\n", "0 0.550434 0.072795 \n", "1 0.172714 0.145941 \n", "2 0.127098 0.094736 \n", "3 0.516785 0.154239 \n", "4 0.671897 0.214587 \n", "\n", " Cells_AreaShape_Zernike_0_0 ... Nuclei_Texture_SumEntropy_DNA_5_0 \\\n", "0 -0.210908 ... 0.090021 \n", "1 -0.039057 ... -0.030338 \n", "2 0.078257 ... 0.102946 \n", "3 -0.347572 ... 0.470516 \n", "4 -0.221396 ... 0.671790 \n", "\n", " Nuclei_Texture_SumEntropy_ER_3_0 Nuclei_Texture_SumEntropy_Mito_3_0 \\\n", "0 0.375747 -0.010427 \n", "1 0.390687 -0.058560 \n", "2 0.279148 0.029891 \n", "3 0.214792 0.015408 \n", "4 0.424186 0.118873 \n", "\n", " Nuclei_Texture_SumEntropy_RNA_5_0 Nuclei_Texture_SumVariance_ER_3_0 \\\n", "0 0.118997 -0.156442 \n", "1 0.065822 -0.113721 \n", "2 0.079777 -0.007979 \n", "3 0.339100 -0.034251 \n", "4 0.415623 0.257598 \n", "\n", " Nuclei_Texture_SumVariance_Mito_5_0 Nuclei_Texture_SumVariance_RNA_10_0 \\\n", "0 -0.018014 -0.024573 \n", "1 -0.089807 -0.041392 \n", "2 0.081155 -0.089633 \n", "3 -0.047597 0.103611 \n", "4 0.193370 0.238471 \n", "\n", " Nuclei_Texture_Variance_AGP_5_0 Nuclei_Texture_Variance_ER_5_0 \\\n", "0 -0.383890 -0.199967 \n", "1 -0.248967 -0.126183 \n", "2 0.043140 -0.034355 \n", "3 -0.139579 -0.042241 \n", "4 -0.033797 0.080018 \n", "\n", " Nuclei_Texture_Variance_RNA_5_0 \n", "0 -0.075658 \n", "1 0.018563 \n", "2 0.052914 \n", "3 0.249977 \n", "4 0.335890 \n", "\n", "[5 rows x 778 columns]" ] }, "execution_count": 49, "metadata": {}, "output_type": "execute_result" } ], "source": [ "df_filled.head()" ] }, { "cell_type": "code", "execution_count": 48, "metadata": {}, "outputs": [], "source": [ "# 假设你想把 'old_name1' 改成 'new_name1'，'old_name2' 改成 'new_name2'\n", "df_filled = df_filled.rename(columns={\n", " 'Metadata_mmoles_per_liter2': 'pert_dose',\n", " 'Metadata_Plate': 'Plate'\n", "})" ] }, { "cell_type": "code", "execution_count": 50, "metadata": {}, "outputs": [ { "name": "stdout", "output_type": "stream", "text": [ " mean variance\n", "pert_dose 7.671828 19.973740\n", "Plate 25543.999718 728358.583856\n", "Cells_AreaShape_Compactness 0.089974 0.050854\n", "Cells_AreaShape_Extent -0.034355 0.028292\n", "Cells_AreaShape_FormFactor 0.020220 0.081711\n", "... ... ...\n", "Nuclei_Texture_SumVariance_Mito_5_0 -0.063953 0.090173\n", "Nuclei_Texture_SumVariance_RNA_10_0 -0.027541 0.044317\n", "Nuclei_Texture_Variance_AGP_5_0 -0.030047 0.067041\n", "Nuclei_Texture_Variance_ER_5_0 -0.008726 0.035092\n", "Nuclei_Texture_Variance_RNA_5_0 -0.010389 0.082160\n", "\n", "[777 rows x 2 columns]\n" ] }, { "name": "stderr", "output_type": "stream", "text": [ "/tmp/ipykernel_3838551/2502789472.py:3: FutureWarning: The default value of numeric_only in DataFrame.var is deprecated. In a future version, it will default to False. In addition, specifying 'numeric_only=None' is deprecated. Select only valid columns or specify the value of numeric_only to silence this warning.\n", " stats['variance'] = df_filled.var() # 添加方差列\n" ] } ], "source": [ "# 查看每列的基本统计信息\n", "stats = df_filled.describe().T # 转置方便查看\n", "stats['variance'] = df_filled.var() # 添加方差列\n", "print(stats[['mean', 'variance']])" ] }, { "cell_type": "markdown", "metadata": {}, "source": [ "# 数据清洗" ] }, { "cell_type": "code", "execution_count": 54, "metadata": {}, "outputs": [ { "name": "stdout", "output_type": "stream", "text": [ "最小值: -inf, 最大值: inf\n", "(NumPy) 最小值: -inf, 最大值: inf\n" ] } ], "source": [ "import numpy as np\n", "\n", "# 假设 df_filled 的前 3 列是 index 0,1,2\n", "data_subset = df_filled.iloc[:, 3:] # 取第4列开始到最后\n", "\n", "# 计算极值\n", "min_val = data_subset.min().min()\n", "max_val = data_subset.max().max()\n", "print(f\"最小值: {min_val}, 最大值: {max_val}\")\n", "\n", "# 或用 NumPy\n", "min_val_np = np.min(data_subset.values)\n", "max_val_np = np.max(data_subset.values)\n", "print(f\"(NumPy) 最小值: {min_val_np}, 最大值: {max_val_np}\")\n" ] }, { "cell_type": "code", "execution_count": 57, "metadata": {}, "outputs": [ { "name": "stdout", "output_type": "stream", "text": [ "含有无穷大的列: ['Cells_Correlation_Costes_AGP_DNA', 'Cells_Correlation_Costes_DNA_AGP', 'Cells_Correlation_Costes_DNA_ER', 'Cells_Correlation_Costes_DNA_Mito', 'Cells_Correlation_Costes_ER_DNA', 'Cells_Correlation_Costes_Mito_DNA', 'Cells_Correlation_Costes_RNA_AGP', 'Cells_Correlation_Costes_RNA_DNA', 'Cells_Correlation_RWC_AGP_DNA', 'Cells_Correlation_RWC_DNA_AGP', 'Cells_Neighbors_NumberOfNeighbors_5', 'Cells_Neighbors_NumberOfNeighbors_Adjacent', 'Cells_Neighbors_PercentTouching_Adjacent', 'Cytoplasm_Correlation_Costes_AGP_ER', 'Cytoplasm_Correlation_Costes_AGP_RNA', 'Cytoplasm_Correlation_Costes_DNA_ER', 'Cytoplasm_Correlation_Costes_DNA_Mito', 'Cytoplasm_Correlation_Costes_DNA_RNA', 'Cytoplasm_Correlation_Costes_ER_AGP', 'Cytoplasm_Correlation_Costes_ER_DNA', 'Cytoplasm_Correlation_Costes_ER_RNA', 'Cytoplasm_Correlation_Costes_RNA_AGP', 'Cytoplasm_Correlation_Costes_RNA_DNA', 'Cytoplasm_Correlation_RWC_DNA_AGP', 'Cytoplasm_Correlation_RWC_Mito_DNA', 'Cytoplasm_Granularity_10_AGP', 'Cytoplasm_Granularity_10_ER', 'Cytoplasm_Granularity_10_Mito', 'Cytoplasm_Granularity_10_RNA', 'Cytoplasm_Granularity_11_AGP', 'Cytoplasm_Granularity_11_ER', 'Cytoplasm_Granularity_11_Mito', 'Cytoplasm_Granularity_11_RNA', 'Cytoplasm_Granularity_12_AGP', 'Cytoplasm_Granularity_12_ER', 'Cytoplasm_Granularity_12_Mito', 'Cytoplasm_Granularity_13_AGP', 'Cytoplasm_Granularity_13_ER', 'Cytoplasm_Granularity_14_AGP', 'Cytoplasm_Granularity_14_ER', 'Cytoplasm_Granularity_15_AGP', 'Cytoplasm_Granularity_15_ER', 'Cytoplasm_Granularity_16_AGP', 'Cytoplasm_Granularity_16_ER', 'Cytoplasm_Granularity_16_Mito', 'Cytoplasm_Granularity_16_RNA', 'Cytoplasm_Granularity_8_AGP', 'Cytoplasm_Granularity_9_AGP', 'Cytoplasm_Granularity_9_Mito', 'Cytoplasm_Granularity_9_RNA', 'Cytoplasm_RadialDistribution_RadialCV_AGP_1of4', 'Cytoplasm_RadialDistribution_RadialCV_DNA_1of4', 'Cytoplasm_RadialDistribution_RadialCV_ER_1of4', 'Nuclei_Correlation_Costes_DNA_AGP', 'Nuclei_Correlation_Costes_DNA_ER', 'Nuclei_Correlation_Costes_DNA_Mito', 'Nuclei_Correlation_Costes_RNA_AGP', 'Nuclei_Correlation_RWC_DNA_AGP', 'Nuclei_Granularity_10_DNA', 'Nuclei_Granularity_13_AGP', 'Nuclei_Granularity_13_DNA', 'Nuclei_Granularity_14_AGP', 'Nuclei_Granularity_14_DNA', 'Nuclei_Granularity_15_AGP', 'Nuclei_Granularity_15_DNA', 'Nuclei_Granularity_16_AGP', 'Nuclei_Granularity_16_DNA', 'Nuclei_Granularity_5_AGP', 'Nuclei_Granularity_7_RNA', 'Nuclei_Granularity_8_ER', 'Nuclei_Granularity_8_Mito', 'Nuclei_Granularity_8_RNA', 'Nuclei_Granularity_9_DNA']\n" ] }, { "data": { "text/plain": [ "73" ] }, "execution_count": 57, "metadata": {}, "output_type": "execute_result" } ], "source": [ "# 查看哪些列含有 inf/-inf\n", "cols_with_inf = data_subset.columns.to_list()\n", "inf_cols = data_subset.loc[:, np.isinf(data_subset).any()].columns.tolist()\n", "print(\"含有无穷大的列:\", inf_cols)\n", "len(inf_cols)\n", "# 查看具体行数\n", "# rows_with_inf = data_subset[np.isinf(data_subset).any(axis=1)]\n", "# print(\"含有无穷大的行数:\", len(rows_with_inf))\n" ] }, { "cell_type": "code", "execution_count": 58, "metadata": {}, "outputs": [ { "name": "stderr", "output_type": "stream", "text": [ "/tmp/ipykernel_3838551/3153003127.py:3: FutureWarning: The default value of numeric_only in DataFrame.median is deprecated. In a future version, it will default to False. In addition, specifying 'numeric_only=None' is deprecated. Select only valid columns or specify the value of numeric_only to silence this warning.\n", " data_subset.fillna(df_filled.median(), inplace=True)\n" ] }, { "name": "stdout", "output_type": "stream", "text": [ "处理后最小值: -3029.4872422556355, 最大值: 141697211.2165867\n" ] } ], "source": [ "# 将无穷大值替换为 NaN\n", "data_subset.replace([np.inf, -np.inf], np.nan, inplace=True)\n", "data_subset.fillna(df_filled.median(), inplace=True)\n", "\n", "# data_subset = df_filled.iloc[:, 3:] # 忽略前3列\n", "min_val = data_subset.min().min()\n", "max_val = data_subset.max().max()\n", "print(f\"处理后最小值: {min_val}, 最大值: {max_val}\")\n" ] }, { "cell_type": "code", "execution_count": 60, "metadata": {}, "outputs": [ { "name": "stdout", "output_type": "stream", "text": [ "处理后的统计信息: {'min': -108.14135045725546, 'max': 28.784896487400754, 'mean': -0.002332358880140924, 'std': 0.9206103389156143}\n" ] } ], "source": [ "import pandas as pd\n", "import numpy as np\n", "from sklearn.preprocessing import RobustScaler\n", "\n", "def preprocess_dl_auto_log(df: pd.DataFrame, skip_cols: int = 3,\n", " log_keywords: list = None, lower_percentile=0.01, upper_percentile=0.99):\n", " \"\"\"\n", " 深度学习训练前数据预处理（自动识别 log 列）：\n", " - skip_cols: 保留前几列不处理（如 ID、SMILES、剂量等）\n", " - log_keywords: 用于识别需要 log1p 的列的关键字（如 'Granularity', 'Neighbors', 'RadialDistribution'）\n", " - lower_percentile, upper_percentile: 百分位剪切极值\n", " \n", " 返回：\n", " df_clean: 处理后的 DataFrame\n", " stats: 最终统计信息（min, max, mean, std）\n", " \"\"\"\n", " df_clean = df.copy()\n", " \n", " # --- 1. 分离元数据和特征 ---\n", " meta_cols = df_clean.iloc[:, :skip_cols]\n", " data_cols = df_clean.iloc[:, skip_cols:]\n", " \n", " # --- 2. 替换 inf/-inf 为 NaN ---\n", " data_cols.replace([np.inf, -np.inf], np.nan, inplace=True)\n", " \n", " # --- 3. 填充 NaN（用中位数） ---\n", " data_cols.fillna(data_cols.median(), inplace=True)\n", " \n", " # --- 4. 百分位剪切极值 ---\n", " lower = data_cols.quantile(lower_percentile)\n", " upper = data_cols.quantile(upper_percentile)\n", " data_cols = data_cols.clip(lower=lower, upper=upper, axis=1)\n", " \n", " # --- 5. 自动识别 log1p 列 ---\n", " log_cols = []\n", " if log_keywords:\n", " log_cols = [col for col in data_cols.columns if any(k in col for k in log_keywords)]\n", " \n", " # --- 6. 对计数/强度型特征做 log1p ---\n", " for col in log_cols:\n", " # 保证非负\n", " data_cols[col] = np.log1p(data_cols[col] - data_cols[col].min())\n", " \n", " # --- 7. RobustScaler 缩放 ---\n", " scaler = RobustScaler()\n", " data_scaled = scaler.fit_transform(data_cols.values)\n", " data_cols.iloc[:, :] = data_scaled\n", " \n", " # --- 8. 合并回原 DataFrame ---\n", " df_clean = pd.concat([meta_cols.reset_index(drop=True), pd.DataFrame(data_cols, columns=data_cols.columns)], axis=1)\n", " \n", " # --- 9. 统计信息 ---\n", " stats = {\n", " 'min': data_cols.min().min(),\n", " 'max': data_cols.max().max(),\n", " 'mean': data_cols.mean().mean(),\n", " 'std': data_cols.std().mean()\n", " }\n", " \n", " return df_clean, stats\n", "\n", "# ------------------------\n", "# 使用示例\n", "# ------------------------\n", "keywords = ['Granularity', 'Neighbors', 'RadialDistribution'] # 自动识别 log1p 特征\n", "df_processed, stats = preprocess_dl_auto_log(df_filled, skip_cols=3, log_keywords=keywords)\n", "print(\"处理后的统计信息:\", stats)" ] }, { "cell_type": "code", "execution_count": 61, "metadata": {}, "outputs": [ { "data": { "text/html": [ "
\n", "\n", "\n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", "
SMILESpert_dosePlateCells_AreaShape_CompactnessCells_AreaShape_ExtentCells_AreaShape_FormFactorCells_AreaShape_MaxFeretDiameterCells_AreaShape_MaximumRadiusCells_AreaShape_SolidityCells_AreaShape_Zernike_0_0...Nuclei_Texture_SumEntropy_DNA_5_0Nuclei_Texture_SumEntropy_ER_3_0Nuclei_Texture_SumEntropy_Mito_3_0Nuclei_Texture_SumEntropy_RNA_5_0Nuclei_Texture_SumVariance_ER_3_0Nuclei_Texture_SumVariance_Mito_5_0Nuclei_Texture_SumVariance_RNA_10_0Nuclei_Texture_Variance_AGP_5_0Nuclei_Texture_Variance_ER_5_0Nuclei_Texture_Variance_RNA_5_0
0CCCOc1cc(N)ccc1C(=O)OCCN(CC)CC6.05242770.833037-0.542840-0.9602581.3054791.3523710.395747-0.755727...0.4237981.5856730.2018600.556881-0.1998640.2274950.111083-1.374047-0.939027-0.163633
1C=C(C)[C@@H]1CCC(C)=C[C@H]1c1c(OC)cc(C)cc1OC10.0024277-0.0712110.577948-0.1655950.4480060.3367850.7903700.112848...0.1626521.6405650.0708640.396778-0.0584210.0081920.025142-0.811356-0.5250500.249372
2COc1cc(O)cc(/C=C/c2ccccc2)c110.0024277-0.7267020.5317690.609702-0.2006820.2141360.5141190.705779...0.4518421.2307730.3115910.4387960.2916720.530425-0.2213580.406866-0.0098310.399941
3COc1c(O)cc2c(c1O)[C@H]1O[C@H](CO)[C@@H](O)[C@H...10.00242771.509908-1.361061-0.2490281.5027401.2619000.835140-1.446454...1.2493690.9943360.2721731.2195750.2046900.1371270.766081-0.355158-0.0540761.263733
4CCOc1ccc2ccccc2c1C(=O)N[C@@H]1C(=O)N2[C@@H]1SC...4.39242770.7972120.0979490.3013971.4630501.6789541.160718-0.808732...1.6860781.7636370.5537621.4499731.1709510.8732051.4551860.0860020.6318801.640317
\n", "

5 rows × 778 columns

\n", "
" ], "text/plain": [ " SMILES pert_dose Plate \\\n", "0 CCCOc1cc(N)ccc1C(=O)OCCN(CC)CC 6.05 24277 \n", "1 C=C(C)[C@@H]1CCC(C)=C[C@H]1c1c(OC)cc(C)cc1OC 10.00 24277 \n", "2 COc1cc(O)cc(/C=C/c2ccccc2)c1 10.00 24277 \n", "3 COc1c(O)cc2c(c1O)[C@H]1O[C@H](CO)[C@@H](O)[C@H... 10.00 24277 \n", "4 CCOc1ccc2ccccc2c1C(=O)N[C@@H]1C(=O)N2[C@@H]1SC... 4.39 24277 \n", "\n", " Cells_AreaShape_Compactness Cells_AreaShape_Extent \\\n", "0 0.833037 -0.542840 \n", "1 -0.071211 0.577948 \n", "2 -0.726702 0.531769 \n", "3 1.509908 -1.361061 \n", "4 0.797212 0.097949 \n", "\n", " Cells_AreaShape_FormFactor Cells_AreaShape_MaxFeretDiameter \\\n", "0 -0.960258 1.305479 \n", "1 -0.165595 0.448006 \n", "2 0.609702 -0.200682 \n", "3 -0.249028 1.502740 \n", "4 0.301397 1.463050 \n", "\n", " Cells_AreaShape_MaximumRadius Cells_AreaShape_Solidity \\\n", "0 1.352371 0.395747 \n", "1 0.336785 0.790370 \n", "2 0.214136 0.514119 \n", "3 1.261900 0.835140 \n", "4 1.678954 1.160718 \n", "\n", " Cells_AreaShape_Zernike_0_0 ... Nuclei_Texture_SumEntropy_DNA_5_0 \\\n", "0 -0.755727 ... 0.423798 \n", "1 0.112848 ... 0.162652 \n", "2 0.705779 ... 0.451842 \n", "3 -1.446454 ... 1.249369 \n", "4 -0.808732 ... 1.686078 \n", "\n", " Nuclei_Texture_SumEntropy_ER_3_0 Nuclei_Texture_SumEntropy_Mito_3_0 \\\n", "0 1.585673 0.201860 \n", "1 1.640565 0.070864 \n", "2 1.230773 0.311591 \n", "3 0.994336 0.272173 \n", "4 1.763637 0.553762 \n", "\n", " Nuclei_Texture_SumEntropy_RNA_5_0 Nuclei_Texture_SumVariance_ER_3_0 \\\n", "0 0.556881 -0.199864 \n", "1 0.396778 -0.058421 \n", "2 0.438796 0.291672 \n", "3 1.219575 0.204690 \n", "4 1.449973 1.170951 \n", "\n", " Nuclei_Texture_SumVariance_Mito_5_0 Nuclei_Texture_SumVariance_RNA_10_0 \\\n", "0 0.227495 0.111083 \n", "1 0.008192 0.025142 \n", "2 0.530425 -0.221358 \n", "3 0.137127 0.766081 \n", "4 0.873205 1.455186 \n", "\n", " Nuclei_Texture_Variance_AGP_5_0 Nuclei_Texture_Variance_ER_5_0 \\\n", "0 -1.374047 -0.939027 \n", "1 -0.811356 -0.525050 \n", "2 0.406866 -0.009831 \n", "3 -0.355158 -0.054076 \n", "4 0.086002 0.631880 \n", "\n", " Nuclei_Texture_Variance_RNA_5_0 \n", "0 -0.163633 \n", "1 0.249372 \n", "2 0.399941 \n", "3 1.263733 \n", "4 1.640317 \n", "\n", "[5 rows x 778 columns]" ] }, "execution_count": 61, "metadata": {}, "output_type": "execute_result" } ], "source": [ "df_processed.head()" ] }, { "cell_type": "code", "execution_count": null, "metadata": {}, "outputs": [], "source": [] }, { "cell_type": "code", "execution_count": null, "metadata": {}, "outputs": [], "source": [ "# save to parquet\n", "output_path = 'Step2_CP_cleaned.parquet'\n", "df_processed.to_parquet(output_path, index=False)" ] }, { "cell_type": "code", "execution_count": 1, "metadata": {}, "outputs": [], "source": [ "import pandas as pd\n", "output_path = 'Step2_CP_cleaned.parquet'\n", "df_processed = pd.read_parquet(output_path)" ] }, { "cell_type": "markdown", "metadata": {}, "source": [ "# 数据处理到这就可以了，对齐数据先不做，想好怎么处理直接对 df 做反而方便。" ] }, { "cell_type": "code", "execution_count": 2, "metadata": {}, "outputs": [ { "data": { "text/html": [ "
\n", "\n", "\n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", "
SMILESpert_dosePlateCells_AreaShape_CompactnessCells_AreaShape_ExtentCells_AreaShape_FormFactorCells_AreaShape_MaxFeretDiameterCells_AreaShape_MaximumRadiusCells_AreaShape_SolidityCells_AreaShape_Zernike_0_0...Nuclei_Texture_SumEntropy_DNA_5_0Nuclei_Texture_SumEntropy_ER_3_0Nuclei_Texture_SumEntropy_Mito_3_0Nuclei_Texture_SumEntropy_RNA_5_0Nuclei_Texture_SumVariance_ER_3_0Nuclei_Texture_SumVariance_Mito_5_0Nuclei_Texture_SumVariance_RNA_10_0Nuclei_Texture_Variance_AGP_5_0Nuclei_Texture_Variance_ER_5_0Nuclei_Texture_Variance_RNA_5_0
0CCCOc1cc(N)ccc1C(=O)OCCN(CC)CC6.05242770.833037-0.542840-0.9602581.3054791.3523710.395747-0.755727...0.4237981.5856730.2018600.556881-0.1998640.2274950.111083-1.374047-0.939027-0.163633
1C=C(C)[C@@H]1CCC(C)=C[C@H]1c1c(OC)cc(C)cc1OC10.0024277-0.0712110.577948-0.1655950.4480060.3367850.7903700.112848...0.1626521.6405650.0708640.396778-0.0584210.0081920.025142-0.811356-0.5250500.249372
2COc1cc(O)cc(/C=C/c2ccccc2)c110.0024277-0.7267020.5317690.609702-0.2006820.2141360.5141190.705779...0.4518421.2307730.3115910.4387960.2916720.530425-0.2213580.406866-0.0098310.399941
3COc1c(O)cc2c(c1O)[C@H]1O[C@H](CO)[C@@H](O)[C@H...10.00242771.509908-1.361061-0.2490281.5027401.2619000.835140-1.446454...1.2493690.9943360.2721731.2195750.2046900.1371270.766081-0.355158-0.0540761.263733
4CCOc1ccc2ccccc2c1C(=O)N[C@@H]1C(=O)N2[C@@H]1SC...4.39242770.7972120.0979490.3013971.4630501.6789541.160718-0.808732...1.6860781.7636370.5537621.4499731.1709510.8732051.4551860.0860020.6318801.640317
\n", "

5 rows × 778 columns

\n", "
" ], "text/plain": [ " SMILES pert_dose Plate \\\n", "0 CCCOc1cc(N)ccc1C(=O)OCCN(CC)CC 6.05 24277 \n", "1 C=C(C)[C@@H]1CCC(C)=C[C@H]1c1c(OC)cc(C)cc1OC 10.00 24277 \n", "2 COc1cc(O)cc(/C=C/c2ccccc2)c1 10.00 24277 \n", "3 COc1c(O)cc2c(c1O)[C@H]1O[C@H](CO)[C@@H](O)[C@H... 10.00 24277 \n", "4 CCOc1ccc2ccccc2c1C(=O)N[C@@H]1C(=O)N2[C@@H]1SC... 4.39 24277 \n", "\n", " Cells_AreaShape_Compactness Cells_AreaShape_Extent \\\n", "0 0.833037 -0.542840 \n", "1 -0.071211 0.577948 \n", "2 -0.726702 0.531769 \n", "3 1.509908 -1.361061 \n", "4 0.797212 0.097949 \n", "\n", " Cells_AreaShape_FormFactor Cells_AreaShape_MaxFeretDiameter \\\n", "0 -0.960258 1.305479 \n", "1 -0.165595 0.448006 \n", "2 0.609702 -0.200682 \n", "3 -0.249028 1.502740 \n", "4 0.301397 1.463050 \n", "\n", " Cells_AreaShape_MaximumRadius Cells_AreaShape_Solidity \\\n", "0 1.352371 0.395747 \n", "1 0.336785 0.790370 \n", "2 0.214136 0.514119 \n", "3 1.261900 0.835140 \n", "4 1.678954 1.160718 \n", "\n", " Cells_AreaShape_Zernike_0_0 ... Nuclei_Texture_SumEntropy_DNA_5_0 \\\n", "0 -0.755727 ... 0.423798 \n", "1 0.112848 ... 0.162652 \n", "2 0.705779 ... 0.451842 \n", "3 -1.446454 ... 1.249369 \n", "4 -0.808732 ... 1.686078 \n", "\n", " Nuclei_Texture_SumEntropy_ER_3_0 Nuclei_Texture_SumEntropy_Mito_3_0 \\\n", "0 1.585673 0.201860 \n", "1 1.640565 0.070864 \n", "2 1.230773 0.311591 \n", "3 0.994336 0.272173 \n", "4 1.763637 0.553762 \n", "\n", " Nuclei_Texture_SumEntropy_RNA_5_0 Nuclei_Texture_SumVariance_ER_3_0 \\\n", "0 0.556881 -0.199864 \n", "1 0.396778 -0.058421 \n", "2 0.438796 0.291672 \n", "3 1.219575 0.204690 \n", "4 1.449973 1.170951 \n", "\n", " Nuclei_Texture_SumVariance_Mito_5_0 Nuclei_Texture_SumVariance_RNA_10_0 \\\n", "0 0.227495 0.111083 \n", "1 0.008192 0.025142 \n", "2 0.530425 -0.221358 \n", "3 0.137127 0.766081 \n", "4 0.873205 1.455186 \n", "\n", " Nuclei_Texture_Variance_AGP_5_0 Nuclei_Texture_Variance_ER_5_0 \\\n", "0 -1.374047 -0.939027 \n", "1 -0.811356 -0.525050 \n", "2 0.406866 -0.009831 \n", "3 -0.355158 -0.054076 \n", "4 0.086002 0.631880 \n", "\n", " Nuclei_Texture_Variance_RNA_5_0 \n", "0 -0.163633 \n", "1 0.249372 \n", "2 0.399941 \n", "3 1.263733 \n", "4 1.640317 \n", "\n", "[5 rows x 778 columns]" ] }, "execution_count": 2, "metadata": {}, "output_type": "execute_result" } ], "source": [ "df_processed.head()" ] }, { "cell_type": "code", "execution_count": 3, "metadata": {}, "outputs": [ { "data": { "text/plain": [ "DMSO 26572\n", "CC(C)(C)NC[C@@H](O)COc1cccc2c1CCCC2=O 16\n", "CC(C)NCC(O)COc1cccc2[nH]ccc12 16\n", "COCC(=O)O[C@]1(CCN(C)CCCc2nc3ccccc3[nH]2)CCc2cc(F)ccc2[C@@H]1C(C)C 12\n", "COc1c(C(C)C)oc2cc3oc(=O)ccc3cc12 12\n", " ... \n", "O=C(Nc1ccc2c(c1)[C@@H]1C[C@H](CC(=O)N3CCCCC3)O[C@H](CO)[C@@H]1O2)NC1CCCC1 1\n", "O=C(Nc1ccc2c(c1)[C@H]1C[C@@H](CC(=O)N3CCCCC3)O[C@@H](CO)[C@H]1O2)NC1CCCC1 1\n", "O=C(Nc1ccc2c(c1)[C@@H]1C[C@@H](CC(=O)N3CCCCC3)O[C@H](CO)[C@@H]1O2)NC1CCCC1 1\n", "O=C(Nc1ccc2c(c1)[C@H]1C[C@H](CC(=O)N3CCCCC3)O[C@@H](CO)[C@H]1O2)NC1CCCC1 1\n", "CS(=O)(=O)Nc1ccc2c(c1)[C@H]1C[C@H](CC(=O)NCc3cc(F)ccc3F)O[C@H](CO)[C@H]1O2 1\n", "Name: SMILES, Length: 20342, dtype: int64" ] }, "execution_count": 3, "metadata": {}, "output_type": "execute_result" } ], "source": [ "df_processed['SMILES'].value_counts()" ] }, { "cell_type": "code", "execution_count": null, "metadata": {}, "outputs": [ { "name": "stdout", "output_type": "stream", "text": [ "原始特征数量: 775\n", "Phase 1: 清洗异常值 (Inf/NaN)...\n", "清洗后特征数量: 775\n", "Phase 2: Log-transform...\n", "Phase 3: Robust Z-score Normalization...\n", "Phase 4: Clamping outliers to [-10, 10]...\n", "\n", "【清洗后统计量检查】\n", "Min: -10.0\n", "Max: 10.0\n", "Mean: -0.0166932575210808\n", "成功：数据已清洗完毕，无 NaN/Inf。\n", "文件已保存至: BBBC047_CP_data_normalized.csv\n" ] } ], "source": [ "import pandas as pd\n", "import numpy as np\n", "\n", "def clean_and_normalize(df, output_file):\n", " # print(f\"正在读取文件: {input_file} ...\")\n", " # df = pd.read_csv(input_file)\n", " \n", " # 分离 Metadata 和 Feature\n", " metadata_cols = df.columns[:3] \n", " feature_cols = df.columns[3:]\n", " \n", " df_meta = df[metadata_cols]\n", " df_features = df[feature_cols]\n", " \n", " print(f\"原始特征数量: {len(feature_cols)}\")\n", "\n", " # ==========================================\n", " # Phase 1: 数据清洗 (Sanitization)\n", " # ==========================================\n", " print(\"Phase 1: 清洗异常值 (Inf/NaN)...\")\n", " \n", " # 1.1 将 inf/-inf 替换为 NaN\n", " df_features = df_features.replace([np.inf, -np.inf], np.nan)\n", " \n", " # 1.2 检查缺失值比例\n", " nan_counts = df_features.isna().sum()\n", " cols_to_drop = nan_counts[nan_counts > 0.1 * len(df_features)].index\n", " if len(cols_to_drop) > 0:\n", " print(f\"警告: 丢弃 {len(cols_to_drop)} 个缺失值超过 10% 的列。\")\n", " df_features = df_features.drop(columns=cols_to_drop)\n", " \n", " # 1.3 剩余 NaN 用该列的中位数填充 (Imputation)\n", " # 这里的 fillna 使用全数据的中位数，也可以改用 DMSO 的中位数，但在清洗阶段全数据通常足够\n", " df_features = df_features.fillna(df_features.median())\n", "\n", " # 1.4 移除方差为 0 的列 (Constant Features)\n", " # 这些列没有任何信息量，且会导致除以 0 错误\n", " from sklearn.feature_selection import VarianceThreshold\n", " selector = VarianceThreshold(threshold=0)\n", " try:\n", " selector.fit(df_features)\n", " # 获取保留的列名\n", " valid_cols = df_features.columns[selector.get_support(indices=True)]\n", " dropped_cols_count = len(df_features.columns) - len(valid_cols)\n", " if dropped_cols_count > 0:\n", " print(f\"剔除 {dropped_cols_count} 个方差为 0 (常数) 的特征。\")\n", " df_features = df_features[valid_cols]\n", " except ValueError:\n", " print(\"警告: 所有特征方差均为0? 请检查数据。\")\n", "\n", " print(f\"清洗后特征数量: {len(df_features.columns)}\")\n", "\n", " # ==========================================\n", " # Phase 2: 对数变换\n", " # ==========================================\n", " print(\"Phase 2: Log-transform...\")\n", " # 再次确保没有负数 (Correlation 特征可能为负，不进行 log; Area/Intensity 为正)\n", " # 策略：只对 Max > 50 的列做 Log (通常是 Area/Intensity)，且最小值 >= 0\n", " # 小数值特征 (Correlation -1~1) 保持原样\n", " cols_to_log = []\n", " for col in df_features.columns:\n", " col_min = df_features[col].min()\n", " col_max = df_features[col].max()\n", " # 启发式规则：如果是非负数且跨度很大，则做 log\n", " if col_min >= 0 and col_max > 50:\n", " cols_to_log.append(col)\n", " \n", " if cols_to_log:\n", " # 使用 log1p 避免 log(0)\n", " df_features[cols_to_log] = np.log1p(df_features[cols_to_log])\n", " print(f\"已对 {len(cols_to_log)} 个大数值特征执行 Log 变换。\")\n", "\n", " # ==========================================\n", " # Phase 3: Robust Normalization\n", " # ==========================================\n", " print(\"Phase 3: Robust Z-score Normalization...\")\n", " \n", " # 重新定位 DMSO\n", " dmso_mask = df_meta['SMILES'].astype(str).str.contains('DMSO', case=False, na=False)\n", " if not dmso_mask.any():\n", " dmso_mask = df_meta['dose'] == 0\n", " \n", " dmso_data = df_features.loc[dmso_mask]\n", " \n", " medians = dmso_data.median()\n", " mads = (dmso_data - medians).abs().median()\n", " \n", " # 安全处理 MAD=0 的情况\n", " # 如果 MAD=0 (DMSO中该特征是常数)，除法会产生 inf。\n", " # 策略：如果 MAD < 1e-4，我们将 MAD 设为 1 (即不缩放，只减去中位数)，或者直接丢弃该列。\n", " # 这里我们选择设为 1，并在后续步骤 Clip。\n", " mads_safe = mads.copy()\n", " mads_safe[mads_safe < 1e-5] = 1.0 \n", "\n", " df_scaled = (df_features - medians) / (mads_safe * 1.4826)\n", "\n", " # ==========================================\n", " # Phase 4: 极值截断 (Clamping/Winsorizing)\n", " # ==========================================\n", " # 这一步对于深度学习至关重要。\n", " # 我们不希望因为一个奇怪的气泡导致输入值为 2000000。\n", " print(\"Phase 4: Clamping outliers to [-10, 10]...\")\n", " df_final_features = df_scaled.clip(lower=-10, upper=10)\n", "\n", " # 4. 再次检查结果\n", " print(\"\\n【清洗后统计量检查】\")\n", " print(f\"Min: {df_final_features.min().min()}\")\n", " print(f\"Max: {df_final_features.max().max()}\")\n", " print(f\"Mean: {df_final_features.mean().mean()}\")\n", " \n", " if np.isinf(df_final_features.values).any():\n", " print(\"错误：数据中仍存在 inf！\")\n", " elif df_final_features.isna().any().any():\n", " print(\"错误：数据中仍存在 NaN！\")\n", " else:\n", " print(\"成功：数据已清洗完毕，无 NaN/Inf。\")\n", "\n", " # 保存\n", " df_final = pd.concat([df_meta, df_final_features], axis=1)\n", " return df_final\n", " df_final.to_csv(output_file, index=False)\n", " print(f\"文件已保存至: {output_file}\")\n", "\n", "# --- 执行代码 ---\n", "# 确保你的目录下有 BBBC036.csv 文件\n", "if __name__ == \"__main__\":\n", " df_processed = clean_and_normalize(df_processed, 'BBBC047_CP_data_normalized.csv')" ] }, { "cell_type": "code", "execution_count": 5, "metadata": {}, "outputs": [ { "data": { "text/html": [ "
\n", "\n", "\n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", "
SMILESpert_dosePlateCells_AreaShape_CompactnessCells_AreaShape_ExtentCells_AreaShape_FormFactorCells_AreaShape_MaxFeretDiameterCells_AreaShape_MaximumRadiusCells_AreaShape_SolidityCells_AreaShape_Zernike_0_0...Nuclei_Texture_SumEntropy_DNA_5_0Nuclei_Texture_SumEntropy_ER_3_0Nuclei_Texture_SumEntropy_Mito_3_0Nuclei_Texture_SumEntropy_RNA_5_0Nuclei_Texture_SumVariance_ER_3_0Nuclei_Texture_SumVariance_Mito_5_0Nuclei_Texture_SumVariance_RNA_10_0Nuclei_Texture_Variance_AGP_5_0Nuclei_Texture_Variance_ER_5_0Nuclei_Texture_Variance_RNA_5_0
0CCCOc1cc(N)ccc1C(=O)OCCN(CC)CC6.05242771.798728-1.067827-1.3908692.3426742.1911470.603093-1.579134...0.5540602.3499770.0607780.611735-0.5273890.040127-0.110624-2.284495-1.438048-0.403544
1C=C(C)[C@@H]1CCC(C)=C[C@H]1c1c(OC)cc(C)cc1OC10.00242770.3030870.574654-0.1846270.9679080.6067411.175119-0.233026...0.1581812.434973-0.1336890.383268-0.312258-0.275593-0.232614-1.437120-0.8580000.206142
2COc1cc(O)cc(/C=C/c2ccccc2)c110.0024277-0.7811050.5069800.992217-0.0721190.4153980.7746800.685892...0.5965721.8004340.2236770.4432260.2202240.476239-0.5825150.397441-0.1360950.428414
3COc1c(O)cc2c(c1O)[C@H]1O[C@H](CO)[C@@H](O)[C@H...10.00242772.918283-2.266905-0.3112732.6589382.0500041.240017-2.649614...1.8055711.4343230.1651601.5574040.087928-0.0899720.819127-0.750117-0.1980891.703560
4CCOc1ccc2ccccc2c1C(=O)N[C@@H]1C(=O)N2[C@@H]1SC...4.39242771.739474-0.1287710.5242322.5953052.7006461.711957-1.661281...2.4675932.6255430.5831851.8861841.5575830.9697221.797292-0.0857590.7630472.259479
\n", "

5 rows × 778 columns

\n", "
" ], "text/plain": [ " SMILES pert_dose Plate \\\n", "0 CCCOc1cc(N)ccc1C(=O)OCCN(CC)CC 6.05 24277 \n", "1 C=C(C)[C@@H]1CCC(C)=C[C@H]1c1c(OC)cc(C)cc1OC 10.00 24277 \n", "2 COc1cc(O)cc(/C=C/c2ccccc2)c1 10.00 24277 \n", "3 COc1c(O)cc2c(c1O)[C@H]1O[C@H](CO)[C@@H](O)[C@H... 10.00 24277 \n", "4 CCOc1ccc2ccccc2c1C(=O)N[C@@H]1C(=O)N2[C@@H]1SC... 4.39 24277 \n", "\n", " Cells_AreaShape_Compactness Cells_AreaShape_Extent \\\n", "0 1.798728 -1.067827 \n", "1 0.303087 0.574654 \n", "2 -0.781105 0.506980 \n", "3 2.918283 -2.266905 \n", "4 1.739474 -0.128771 \n", "\n", " Cells_AreaShape_FormFactor Cells_AreaShape_MaxFeretDiameter \\\n", "0 -1.390869 2.342674 \n", "1 -0.184627 0.967908 \n", "2 0.992217 -0.072119 \n", "3 -0.311273 2.658938 \n", "4 0.524232 2.595305 \n", "\n", " Cells_AreaShape_MaximumRadius Cells_AreaShape_Solidity \\\n", "0 2.191147 0.603093 \n", "1 0.606741 1.175119 \n", "2 0.415398 0.774680 \n", "3 2.050004 1.240017 \n", "4 2.700646 1.711957 \n", "\n", " Cells_AreaShape_Zernike_0_0 ... Nuclei_Texture_SumEntropy_DNA_5_0 \\\n", "0 -1.579134 ... 0.554060 \n", "1 -0.233026 ... 0.158181 \n", "2 0.685892 ... 0.596572 \n", "3 -2.649614 ... 1.805571 \n", "4 -1.661281 ... 2.467593 \n", "\n", " Nuclei_Texture_SumEntropy_ER_3_0 Nuclei_Texture_SumEntropy_Mito_3_0 \\\n", "0 2.349977 0.060778 \n", "1 2.434973 -0.133689 \n", "2 1.800434 0.223677 \n", "3 1.434323 0.165160 \n", "4 2.625543 0.583185 \n", "\n", " Nuclei_Texture_SumEntropy_RNA_5_0 Nuclei_Texture_SumVariance_ER_3_0 \\\n", "0 0.611735 -0.527389 \n", "1 0.383268 -0.312258 \n", "2 0.443226 0.220224 \n", "3 1.557404 0.087928 \n", "4 1.886184 1.557583 \n", "\n", " Nuclei_Texture_SumVariance_Mito_5_0 Nuclei_Texture_SumVariance_RNA_10_0 \\\n", "0 0.040127 -0.110624 \n", "1 -0.275593 -0.232614 \n", "2 0.476239 -0.582515 \n", "3 -0.089972 0.819127 \n", "4 0.969722 1.797292 \n", "\n", " Nuclei_Texture_Variance_AGP_5_0 Nuclei_Texture_Variance_ER_5_0 \\\n", "0 -2.284495 -1.438048 \n", "1 -1.437120 -0.858000 \n", "2 0.397441 -0.136095 \n", "3 -0.750117 -0.198089 \n", "4 -0.085759 0.763047 \n", "\n", " Nuclei_Texture_Variance_RNA_5_0 \n", "0 -0.403544 \n", "1 0.206142 \n", "2 0.428414 \n", "3 1.703560 \n", "4 2.259479 \n", "\n", "[5 rows x 778 columns]" ] }, "execution_count": 5, "metadata": {}, "output_type": "execute_result" } ], "source": [ "df_processed = pd.read_csv('BBBC047_CP_data_normalized.csv')\n", "df_processed.head()" ] }, { "cell_type": "code", "execution_count": 6, "metadata": {}, "outputs": [ { "name": "stdout", "output_type": "stream", "text": [ "\n", "========================================\n", "【全局统计量对比】 (所有特征聚合)\n", "========================================\n", "Metric | Raw Data | Normalized Data \n", "------------------------------------------------------------\n", "Min Value | -10.0000 | -10.0000 \n", "Max Value | 10.0000 | 10.0000 \n", "Mean | -0.0167 | -0.0167 \n", "------------------------------------------------------------\n", "解读: 处理后数据的 Min/Max 应该主要集中在 -5 到 +5 之间（个别离群值除外）。\n", "\n" ] } ], "source": [ "# 1. 提取特征列（排除前4列 Metadata）\n", "feat_cols = df_processed.columns[3:]\n", "\n", "# ---------------------------------------------------------\n", "# 分析 A: 全局极值对比 (Global Extremes)\n", "# ---------------------------------------------------------\n", "print(\"\\n\" + \"=\"*40)\n", "print(\"【全局统计量对比】 (所有特征聚合)\")\n", "print(\"=\"*40)\n", "\n", "raw_feats = df_processed[feat_cols]\n", "norm_feats = df_processed[feat_cols]\n", "\n", "print(f\"{'Metric':<15} | {'Raw Data':<20} | {'Normalized Data':<20}\")\n", "print(\"-\" * 60)\n", "print(f\"{'Min Value':<15} | {raw_feats.min().min():<20.4f} | {norm_feats.min().min():<20.4f}\")\n", "print(f\"{'Max Value':<15} | {raw_feats.max().max():<20.4f} | {norm_feats.max().max():<20.4f}\")\n", "print(f\"{'Mean':<15} | {raw_feats.mean().mean():<20.4f} | {norm_feats.mean().mean():<20.4f}\")\n", "print(\"-\" * 60)\n", "print(\"解读: 处理后数据的 Min/Max 应该主要集中在 -5 到 +5 之间（个别离群值除外）。\\n\")" ] }, { "cell_type": "markdown", "metadata": {}, "source": [ "# 做对齐数据" ] }, { "cell_type": "code", "execution_count": null, "metadata": {}, "outputs": [], "source": [ "import pandas as pd\n", "import numpy as np\n", "import h5py\n", "\n", "# 1. 确保数值列为 float32 类型，这能减少内存使用并提高计算效率\n", "data_cols = [col for col in df_processed.columns if col not in ['Plate', 'SMILES']]\n", "df_processed[data_cols] = df_processed[data_cols].astype(np.float32)\n", "\n", "# 2. 对每个 Plate 中的相同 SMILES 进行求均值，一步到位\n", "# 这是一个高效的 Pandas 操作，能将数据处理流程大大简化\n", "processed_df = df_processed.groupby(['Plate', 'SMILES']).mean().reset_index()\n", "\n", "# 3. 初始化列表来存储数据。这种方式比在循环中频繁创建和合并 NumPy 数组更高效\n", "all_smiles = []\n", "all_control = []\n", "all_target = []\n", "\n", "# 4. 按 Plate 分组，遍历每个 Plate\n", "# `grouped` 是一个迭代器，可以逐个处理每个 Plate\n", "grouped = processed_df.groupby('Plate')\n", "for plate_name, plate_data in grouped:\n", " # 5. 分离 DMSO 和 Target 数据\n", " dmso_data = plate_data[plate_data['SMILES'] == 'DMSO']\n", " target_data = plate_data[plate_data['SMILES'] != 'DMSO']\n", "\n", " # 6. 处理 DMSO 行（Control）\n", " if not dmso_data.empty:\n", " # 提取数值列并转换为 NumPy 数组\n", " control = dmso_data[data_cols].values.flatten().astype(np.float32)\n", " else:\n", " print(f\"警告：Plate '{plate_name}' 没有 DMSO 数据。将用全零填充。\")\n", " control = np.zeros(len(data_cols), dtype=np.float32)\n", "\n", " # 7. 处理非 DMSO 行（Target）\n", " if not target_data.empty:\n", " # 提取 SMILES 和数值数据\n", " smiles = target_data['SMILES'].tolist()\n", " target = target_data[data_cols].values.astype(np.float32)\n", "\n", " # 重复 control，使其与 target 的行数匹配\n", " control_aligned = np.tile(control, (target.shape[0], 1))\n", "\n", " # 将处理后的数据追加到列表中\n", " all_smiles.extend(smiles)\n", " all_control.append(control_aligned)\n", " all_target.append(target)\n", "\n", "# 8. 一次性将列表合并为 NumPy 数组\n", "# 使用 astype('S') 转换为字节串，这是 HDF5 存储字符串的最佳方式\n", "final_smiles = np.array(all_smiles, dtype='S')\n", "# 使用三元表达式防止列表为空时报错\n", "final_control = np.vstack(all_control) if all_control else np.array([], dtype=np.float32).reshape(0, len(data_cols))\n", "final_target = np.vstack(all_target) if all_target else np.array([], dtype=np.float32).reshape(0, len(data_cols))\n", "\n", "# 9. 保存到 HDF5 文件\n", "output_path = './Processed_Paired_CP.h5'\n", "with h5py.File(output_path, 'w') as f:\n", " f.create_dataset('canonical_smiles', data=final_smiles)\n", " f.create_dataset('control', data=final_control, dtype=np.float32)\n", " f.create_dataset('target', data=final_target, dtype=np.float32)\n", "\n", "print(f\"数据已成功处理并保存到 {output_path} 文件中。\")\n", "print(f\"共有 {len(final_smiles)} 个样本。\")\n" ] }, { "cell_type": "code", "execution_count": null, "metadata": {}, "outputs": [], "source": [] }, { "cell_type": "code", "execution_count": null, "metadata": {}, "outputs": [ { "name": "stdout", "output_type": "stream", "text": [ "数据已保存到 Processed_Paired_CP.h5 文件中。\n" ] } ], "source": [ "# # 保存到 HDF5 文件\n", "# with h5py.File('./Processed_Paired_CP.h5', 'w') as f:\n", "# f.create_dataset('canonical_smiles', data=final_smiles)\n", "# f.create_dataset('control', data=final_control, dtype=np.float32)\n", "# f.create_dataset('target', data=final_target, dtype=np.float32)\n", "\n", "# print(\"数据已保存到 Processed_Paired_CP.h5 文件中。\")" ] }, { "cell_type": "code", "execution_count": 25, "metadata": {}, "outputs": [ { "data": { "text/plain": [ "array(['C#CCOC(=O)C1=CCCN(C)C1',\n", " 'C#C[C@]1(O)CC[C@H]2[C@@H]3CCC4=CC(=O)CC[C@@H]4[C@H]3CC[C@@]21CC',\n", " 'C#C[C@]1(O)CC[C@H]2[C@@H]3CCc4cc(OC)ccc4[C@H]3CC[C@@]21C', ...,\n", " 'Cc1noc(C)c1NC(=O)N(C)C[C@@H]1OCc2cnnn2CCCC(=O)N([C@@H](C)CO)C[C@H]1C',\n", " 'Cc1noc(C)c1NC(=O)N(C)C[C@H]1OCc2cn(nn2)CCCC(=O)N([C@@H](C)CO)C[C@@H]1C',\n", " 'Cc1noc(C)c1NC(=O)N(C)C[C@H]1OCc2cnnn2CCCC(=O)N([C@H](C)CO)C[C@@H]1C'],\n", " dtype='\n", "\n", "\n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", "
pert_dosedet_plateCPD_SMILES221227_x_at212345_s_at218597_s_at217140_s_at209253_at214404_x_at219888_at...218397_at202996_at204608_at211071_s_at203341_at202801_at206414_s_at204978_at205379_at203897_at
050.0PAC001_U2OS_6H_X1_B1_UNI4445LCC(C)[C@H](CO)Nc1nc(Nc2cc(N)cc(Cl)c2)c2ncn(C(C...-0.118505-0.293445-0.2940880.292745-0.1458960.973580.247915...-0.5299050.8049920.430269-0.97559-1.7406050.3403820.4987780.3521520.535910.52484
112.7PAC001_U2OS_6H_X1_B1_UNI4445LIc1cccc(CSc2nnc(o2)-c2ccncc2)c10.043264-0.264945-0.0006330.0009350.177604-0.504720.214315...0.228895-0.3813080.5569690.25051-0.025625-0.024428-0.2251220.533052-0.19329-0.25436
250.0PAC001_U2OS_6H_X1_B1_UNI4445LClc1cc(Cl)c(NC(=O)Nc2ccncc2)c(Cl)c1-0.0708050.1957550.004606-0.040855-0.067346-0.19872-0.053195...-0.031795-0.051728-0.0177410.056810.1306950.005919-0.408722-0.057177-0.256990.31004
316.2PAC001_U2OS_6H_X1_B1_UNI4445LCOc1ccc(CNC(=O)Nc2ncc(s2)[N+]([O-])=O)cc10.027165-0.1495450.173113-0.100695-0.309296-0.20782-0.323985...-0.013787-0.1776080.219869-0.161290.365195-0.040848-0.2511220.0407320.17091-0.05144
412.5PAC001_U2OS_6H_X1_B1_UNI4445LOC(=O)c1ccc2c3nc4ccccc4n3c(=O)c3cccc1c230.3032940.254455-0.055418-0.0536350.000455-0.109080.031425...-0.052585-0.1961080.2454690.007840.093595-0.3034180.474878-0.022947-0.285890.06423
\n", "

5 rows × 980 columns

\n", "" ], "text/plain": [ " pert_dose det_plate \\\n", "0 50.0 PAC001_U2OS_6H_X1_B1_UNI4445L \n", "1 12.7 PAC001_U2OS_6H_X1_B1_UNI4445L \n", "2 50.0 PAC001_U2OS_6H_X1_B1_UNI4445L \n", "3 16.2 PAC001_U2OS_6H_X1_B1_UNI4445L \n", "4 12.5 PAC001_U2OS_6H_X1_B1_UNI4445L \n", "\n", " CPD_SMILES 221227_x_at \\\n", "0 CC(C)[C@H](CO)Nc1nc(Nc2cc(N)cc(Cl)c2)c2ncn(C(C... -0.118505 \n", "1 Ic1cccc(CSc2nnc(o2)-c2ccncc2)c1 0.043264 \n", "2 Clc1cc(Cl)c(NC(=O)Nc2ccncc2)c(Cl)c1 -0.070805 \n", "3 COc1ccc(CNC(=O)Nc2ncc(s2)[N+]([O-])=O)cc1 0.027165 \n", "4 OC(=O)c1ccc2c3nc4ccccc4n3c(=O)c3cccc1c23 0.303294 \n", "\n", " 212345_s_at 218597_s_at 217140_s_at 209253_at 214404_x_at 219888_at \\\n", "0 -0.293445 -0.294088 0.292745 -0.145896 0.97358 0.247915 \n", "1 -0.264945 -0.000633 0.000935 0.177604 -0.50472 0.214315 \n", "2 0.195755 0.004606 -0.040855 -0.067346 -0.19872 -0.053195 \n", "3 -0.149545 0.173113 -0.100695 -0.309296 -0.20782 -0.323985 \n", "4 0.254455 -0.055418 -0.053635 0.000455 -0.10908 0.031425 \n", "\n", " ... 218397_at 202996_at 204608_at 211071_s_at 203341_at 202801_at \\\n", "0 ... -0.529905 0.804992 0.430269 -0.97559 -1.740605 0.340382 \n", "1 ... 0.228895 -0.381308 0.556969 0.25051 -0.025625 -0.024428 \n", "2 ... -0.031795 -0.051728 -0.017741 0.05681 0.130695 0.005919 \n", "3 ... -0.013787 -0.177608 0.219869 -0.16129 0.365195 -0.040848 \n", "4 ... -0.052585 -0.196108 0.245469 0.00784 0.093595 -0.303418 \n", "\n", " 206414_s_at 204978_at 205379_at 203897_at \n", "0 0.498778 0.352152 0.53591 0.52484 \n", "1 -0.225122 0.533052 -0.19329 -0.25436 \n", "2 -0.408722 -0.057177 -0.25699 0.31004 \n", "3 -0.251122 0.040732 0.17091 -0.05144 \n", "4 0.474878 -0.022947 -0.28589 0.06423 \n", "\n", "[5 rows x 980 columns]" ] }, "execution_count": 30, "metadata": {}, "output_type": "execute_result" } ], "source": [ "GE_CSV_matched.head()" ] }, { "cell_type": "code", "execution_count": 31, "metadata": {}, "outputs": [ { "name": "stdout", "output_type": "stream", "text": [ "太棒了！整个 DataFrame 都不存在 NaN 值。\n" ] } ], "source": [ "# 检查整个 DataFrame 那存在NaN的列\n", "nan_columns = GE_CSV_matched.columns[GE_CSV_matched.isnull().any()].tolist()\n", "\n", "if nan_columns:\n", " print(\"以下列仍包含 NaN 值：\")\n", " for col in nan_columns:\n", " # 打印出具体列及其 NaN 值的数量\n", " print(f\"- {col}: {GE_CSV_matched[col].isnull().sum()} 个 NaN\")\n", "else:\n", " print(\"太棒了！整个 DataFrame 都不存在 NaN 值。\")" ] }, { "cell_type": "code", "execution_count": 32, "metadata": {}, "outputs": [ { "data": { "text/html": [ "
\n", "\n", "\n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", "
pert_dosedet_plateCPD_SMILES221227_x_at212345_s_at218597_s_at217140_s_at209253_at214404_x_at219888_at...218397_at202996_at204608_at211071_s_at203341_at202801_at206414_s_at204978_at205379_at203897_at
050.0PAC001_U2OS_6H_X1_B1_UNI4445LCC(C)[C@H](CO)Nc1nc(Nc2cc(N)cc(Cl)c2)c2ncn(C(C...-0.118505-0.293445-0.2940880.292745-0.1458960.973580.247915...-0.5299050.8049920.430269-0.97559-1.7406050.3403820.4987780.3521520.535910.52484
112.7PAC001_U2OS_6H_X1_B1_UNI4445LIc1cccc(CSc2nnc(o2)-c2ccncc2)c10.043264-0.264945-0.0006330.0009350.177604-0.504720.214315...0.228895-0.3813080.5569690.25051-0.025625-0.024428-0.2251220.533052-0.19329-0.25436
250.0PAC001_U2OS_6H_X1_B1_UNI4445LClc1cc(Cl)c(NC(=O)Nc2ccncc2)c(Cl)c1-0.0708050.1957550.004606-0.040855-0.067346-0.19872-0.053195...-0.031795-0.051728-0.0177410.056810.1306950.005919-0.408722-0.057177-0.256990.31004
316.2PAC001_U2OS_6H_X1_B1_UNI4445LCOc1ccc(CNC(=O)Nc2ncc(s2)[N+]([O-])=O)cc10.027165-0.1495450.173113-0.100695-0.309296-0.20782-0.323985...-0.013787-0.1776080.219869-0.161290.365195-0.040848-0.2511220.0407320.17091-0.05144
412.5PAC001_U2OS_6H_X1_B1_UNI4445LOC(=O)c1ccc2c3nc4ccccc4n3c(=O)c3cccc1c230.3032940.254455-0.055418-0.0536350.000455-0.109080.031425...-0.052585-0.1961080.2454690.007840.093595-0.3034180.474878-0.022947-0.285890.06423
\n", "

5 rows × 980 columns

\n", "
" ], "text/plain": [ " pert_dose det_plate \\\n", "0 50.0 PAC001_U2OS_6H_X1_B1_UNI4445L \n", "1 12.7 PAC001_U2OS_6H_X1_B1_UNI4445L \n", "2 50.0 PAC001_U2OS_6H_X1_B1_UNI4445L \n", "3 16.2 PAC001_U2OS_6H_X1_B1_UNI4445L \n", "4 12.5 PAC001_U2OS_6H_X1_B1_UNI4445L \n", "\n", " CPD_SMILES 221227_x_at \\\n", "0 CC(C)[C@H](CO)Nc1nc(Nc2cc(N)cc(Cl)c2)c2ncn(C(C... -0.118505 \n", "1 Ic1cccc(CSc2nnc(o2)-c2ccncc2)c1 0.043264 \n", "2 Clc1cc(Cl)c(NC(=O)Nc2ccncc2)c(Cl)c1 -0.070805 \n", "3 COc1ccc(CNC(=O)Nc2ncc(s2)[N+]([O-])=O)cc1 0.027165 \n", "4 OC(=O)c1ccc2c3nc4ccccc4n3c(=O)c3cccc1c23 0.303294 \n", "\n", " 212345_s_at 218597_s_at 217140_s_at 209253_at 214404_x_at 219888_at \\\n", "0 -0.293445 -0.294088 0.292745 -0.145896 0.97358 0.247915 \n", "1 -0.264945 -0.000633 0.000935 0.177604 -0.50472 0.214315 \n", "2 0.195755 0.004606 -0.040855 -0.067346 -0.19872 -0.053195 \n", "3 -0.149545 0.173113 -0.100695 -0.309296 -0.20782 -0.323985 \n", "4 0.254455 -0.055418 -0.053635 0.000455 -0.10908 0.031425 \n", "\n", " ... 218397_at 202996_at 204608_at 211071_s_at 203341_at 202801_at \\\n", "0 ... -0.529905 0.804992 0.430269 -0.97559 -1.740605 0.340382 \n", "1 ... 0.228895 -0.381308 0.556969 0.25051 -0.025625 -0.024428 \n", "2 ... -0.031795 -0.051728 -0.017741 0.05681 0.130695 0.005919 \n", "3 ... -0.013787 -0.177608 0.219869 -0.16129 0.365195 -0.040848 \n", "4 ... -0.052585 -0.196108 0.245469 0.00784 0.093595 -0.303418 \n", "\n", " 206414_s_at 204978_at 205379_at 203897_at \n", "0 0.498778 0.352152 0.53591 0.52484 \n", "1 -0.225122 0.533052 -0.19329 -0.25436 \n", "2 -0.408722 -0.057177 -0.25699 0.31004 \n", "3 -0.251122 0.040732 0.17091 -0.05144 \n", "4 0.474878 -0.022947 -0.28589 0.06423 \n", "\n", "[5 rows x 980 columns]" ] }, "execution_count": 32, "metadata": {}, "output_type": "execute_result" } ], "source": [ "GE_CSV_matched.head()" ] }, { "cell_type": "code", "execution_count": 33, "metadata": {}, "outputs": [ { "data": { "text/html": [ "
\n", "\n", "\n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", "
pert_dosePlateSMILES221227_x_at212345_s_at218597_s_at217140_s_at209253_at214404_x_at219888_at...218397_at202996_at204608_at211071_s_at203341_at202801_at206414_s_at204978_at205379_at203897_at
050.0PAC001_U2OS_6H_X1_B1_UNI4445LCC(C)[C@H](CO)Nc1nc(Nc2cc(N)cc(Cl)c2)c2ncn(C(C...-0.118505-0.293445-0.2940880.292745-0.1458960.973580.247915...-0.5299050.8049920.430269-0.97559-1.7406050.3403820.4987780.3521520.535910.52484
112.7PAC001_U2OS_6H_X1_B1_UNI4445LIc1cccc(CSc2nnc(o2)-c2ccncc2)c10.043264-0.264945-0.0006330.0009350.177604-0.504720.214315...0.228895-0.3813080.5569690.25051-0.025625-0.024428-0.2251220.533052-0.19329-0.25436
250.0PAC001_U2OS_6H_X1_B1_UNI4445LClc1cc(Cl)c(NC(=O)Nc2ccncc2)c(Cl)c1-0.0708050.1957550.004606-0.040855-0.067346-0.19872-0.053195...-0.031795-0.051728-0.0177410.056810.1306950.005919-0.408722-0.057177-0.256990.31004
316.2PAC001_U2OS_6H_X1_B1_UNI4445LCOc1ccc(CNC(=O)Nc2ncc(s2)[N+]([O-])=O)cc10.027165-0.1495450.173113-0.100695-0.309296-0.20782-0.323985...-0.013787-0.1776080.219869-0.161290.365195-0.040848-0.2511220.0407320.17091-0.05144
412.5PAC001_U2OS_6H_X1_B1_UNI4445LOC(=O)c1ccc2c3nc4ccccc4n3c(=O)c3cccc1c230.3032940.254455-0.055418-0.0536350.000455-0.109080.031425...-0.052585-0.1961080.2454690.007840.093595-0.3034180.474878-0.022947-0.285890.06423
\n", "

5 rows × 980 columns

\n", "
" ], "text/plain": [ " pert_dose Plate \\\n", "0 50.0 PAC001_U2OS_6H_X1_B1_UNI4445L \n", "1 12.7 PAC001_U2OS_6H_X1_B1_UNI4445L \n", "2 50.0 PAC001_U2OS_6H_X1_B1_UNI4445L \n", "3 16.2 PAC001_U2OS_6H_X1_B1_UNI4445L \n", "4 12.5 PAC001_U2OS_6H_X1_B1_UNI4445L \n", "\n", " SMILES 221227_x_at \\\n", "0 CC(C)[C@H](CO)Nc1nc(Nc2cc(N)cc(Cl)c2)c2ncn(C(C... -0.118505 \n", "1 Ic1cccc(CSc2nnc(o2)-c2ccncc2)c1 0.043264 \n", "2 Clc1cc(Cl)c(NC(=O)Nc2ccncc2)c(Cl)c1 -0.070805 \n", "3 COc1ccc(CNC(=O)Nc2ncc(s2)[N+]([O-])=O)cc1 0.027165 \n", "4 OC(=O)c1ccc2c3nc4ccccc4n3c(=O)c3cccc1c23 0.303294 \n", "\n", " 212345_s_at 218597_s_at 217140_s_at 209253_at 214404_x_at 219888_at \\\n", "0 -0.293445 -0.294088 0.292745 -0.145896 0.97358 0.247915 \n", "1 -0.264945 -0.000633 0.000935 0.177604 -0.50472 0.214315 \n", "2 0.195755 0.004606 -0.040855 -0.067346 -0.19872 -0.053195 \n", "3 -0.149545 0.173113 -0.100695 -0.309296 -0.20782 -0.323985 \n", "4 0.254455 -0.055418 -0.053635 0.000455 -0.10908 0.031425 \n", "\n", " ... 218397_at 202996_at 204608_at 211071_s_at 203341_at 202801_at \\\n", "0 ... -0.529905 0.804992 0.430269 -0.97559 -1.740605 0.340382 \n", "1 ... 0.228895 -0.381308 0.556969 0.25051 -0.025625 -0.024428 \n", "2 ... -0.031795 -0.051728 -0.017741 0.05681 0.130695 0.005919 \n", "3 ... -0.013787 -0.177608 0.219869 -0.16129 0.365195 -0.040848 \n", "4 ... -0.052585 -0.196108 0.245469 0.00784 0.093595 -0.303418 \n", "\n", " 206414_s_at 204978_at 205379_at 203897_at \n", "0 0.498778 0.352152 0.53591 0.52484 \n", "1 -0.225122 0.533052 -0.19329 -0.25436 \n", "2 -0.408722 -0.057177 -0.25699 0.31004 \n", "3 -0.251122 0.040732 0.17091 -0.05144 \n", "4 0.474878 -0.022947 -0.28589 0.06423 \n", "\n", "[5 rows x 980 columns]" ] }, "execution_count": 33, "metadata": {}, "output_type": "execute_result" } ], "source": [ "GE_CSV_matched = GE_CSV_matched.rename(columns={\n", " 'det_plate': 'Plate',\n", " 'CPD_SMILES': 'SMILES'\n", "})\n", "GE_CSV_matched.head()" ] }, { "cell_type": "code", "execution_count": 34, "metadata": {}, "outputs": [ { "name": "stdout", "output_type": "stream", "text": [ "最终 SMILES 唯一值数量: 20342\n" ] } ], "source": [ "import pandas as pd\n", "from rdkit import Chem\n", "\n", "def canonicalize_smiles(smiles):\n", " \"\"\"尝试将一个 SMILES 转换为规范化形式，忽略 DMSO\"\"\"\n", " if pd.isna(smiles): \n", " return smiles\n", " if str(smiles).upper() == \"DMSO\": # 特殊情况，直接跳过\n", " return \"DMSO\"\n", " try:\n", " mol = Chem.MolFromSmiles(smiles)\n", " if mol is not None:\n", " return Chem.MolToSmiles(mol)\n", " else:\n", " return smiles\n", " except:\n", " return smiles\n", "\n", "# 应用到整列\n", "GE_CSV_matched['SMILES'] = GE_CSV_matched['SMILES'].apply(canonicalize_smiles)\n", "\n", "# 查看结果\n", "final_smiles = GE_CSV_matched['SMILES'].unique()\n", "print(f\"最终 SMILES 唯一值数量: {len(final_smiles)}\")\n" ] }, { "cell_type": "code", "execution_count": 35, "metadata": {}, "outputs": [], "source": [ "# save to parquet\n", "output_path = './Step2_GE_cleaned.parquet'\n", "GE_CSV_matched.to_parquet(output_path, index=False)" ] }, { "cell_type": "code", "execution_count": null, "metadata": {}, "outputs": [], "source": [] }, { "cell_type": "code", "execution_count": 6, "metadata": {}, "outputs": [], "source": [ "import pandas as pd\n", "GE_CSV_matched = pd.read_parquet('./Step2_GE_cleaned.parquet')" ] }, { "cell_type": "code", "execution_count": 16, "metadata": {}, "outputs": [ { "name": "stdout", "output_type": "stream", "text": [ "(61313, 980)\n" ] }, { "data": { "text/html": [ "
\n", "\n", "\n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", "
pert_dosePlateSMILES221227_x_at212345_s_at218597_s_at217140_s_at209253_at214404_x_at219888_at...218397_at202996_at204608_at211071_s_at203341_at202801_at206414_s_at204978_at205379_at203897_at
050.0PAC001_U2OS_6H_X1_B1_UNI4445LCC(C)[C@H](CO)Nc1nc(Nc2cc(N)cc(Cl)c2)c2ncn(C(C...-0.118505-0.293445-0.2940880.292745-0.1458960.973580.247915...-0.5299050.8049920.430269-0.97559-1.7406050.3403820.4987780.3521520.535910.52484
112.7PAC001_U2OS_6H_X1_B1_UNI4445LIc1cccc(CSc2nnc(-c3ccncc3)o2)c10.043264-0.264945-0.0006330.0009350.177604-0.504720.214315...0.228895-0.3813080.5569690.25051-0.025625-0.024428-0.2251220.533052-0.19329-0.25436
250.0PAC001_U2OS_6H_X1_B1_UNI4445LO=C(Nc1ccncc1)Nc1c(Cl)cc(Cl)cc1Cl-0.0708050.1957550.004606-0.040855-0.067346-0.19872-0.053195...-0.031795-0.051728-0.0177410.056810.1306950.005919-0.408722-0.057177-0.256990.31004
316.2PAC001_U2OS_6H_X1_B1_UNI4445LCOc1ccc(CNC(=O)Nc2ncc([N+](=O)[O-])s2)cc10.027165-0.1495450.173113-0.100695-0.309296-0.20782-0.323985...-0.013787-0.1776080.219869-0.161290.365195-0.040848-0.2511220.0407320.17091-0.05144
412.5PAC001_U2OS_6H_X1_B1_UNI4445LO=C(O)c1ccc2c3c1cccc3c(=O)n1c3ccccc3nc210.3032940.254455-0.055418-0.0536350.000455-0.109080.031425...-0.052585-0.1961080.2454690.007840.093595-0.3034180.474878-0.022947-0.285890.06423
\n", "

5 rows × 980 columns

\n", "
" ], "text/plain": [ " pert_dose Plate \\\n", "0 50.0 PAC001_U2OS_6H_X1_B1_UNI4445L \n", "1 12.7 PAC001_U2OS_6H_X1_B1_UNI4445L \n", "2 50.0 PAC001_U2OS_6H_X1_B1_UNI4445L \n", "3 16.2 PAC001_U2OS_6H_X1_B1_UNI4445L \n", "4 12.5 PAC001_U2OS_6H_X1_B1_UNI4445L \n", "\n", " SMILES 221227_x_at \\\n", "0 CC(C)[C@H](CO)Nc1nc(Nc2cc(N)cc(Cl)c2)c2ncn(C(C... -0.118505 \n", "1 Ic1cccc(CSc2nnc(-c3ccncc3)o2)c1 0.043264 \n", "2 O=C(Nc1ccncc1)Nc1c(Cl)cc(Cl)cc1Cl -0.070805 \n", "3 COc1ccc(CNC(=O)Nc2ncc([N+](=O)[O-])s2)cc1 0.027165 \n", "4 O=C(O)c1ccc2c3c1cccc3c(=O)n1c3ccccc3nc21 0.303294 \n", "\n", " 212345_s_at 218597_s_at 217140_s_at 209253_at 214404_x_at 219888_at \\\n", "0 -0.293445 -0.294088 0.292745 -0.145896 0.97358 0.247915 \n", "1 -0.264945 -0.000633 0.000935 0.177604 -0.50472 0.214315 \n", "2 0.195755 0.004606 -0.040855 -0.067346 -0.19872 -0.053195 \n", "3 -0.149545 0.173113 -0.100695 -0.309296 -0.20782 -0.323985 \n", "4 0.254455 -0.055418 -0.053635 0.000455 -0.10908 0.031425 \n", "\n", " ... 218397_at 202996_at 204608_at 211071_s_at 203341_at 202801_at \\\n", "0 ... -0.529905 0.804992 0.430269 -0.97559 -1.740605 0.340382 \n", "1 ... 0.228895 -0.381308 0.556969 0.25051 -0.025625 -0.024428 \n", "2 ... -0.031795 -0.051728 -0.017741 0.05681 0.130695 0.005919 \n", "3 ... -0.013787 -0.177608 0.219869 -0.16129 0.365195 -0.040848 \n", "4 ... -0.052585 -0.196108 0.245469 0.00784 0.093595 -0.303418 \n", "\n", " 206414_s_at 204978_at 205379_at 203897_at \n", "0 0.498778 0.352152 0.53591 0.52484 \n", "1 -0.225122 0.533052 -0.19329 -0.25436 \n", "2 -0.408722 -0.057177 -0.25699 0.31004 \n", "3 -0.251122 0.040732 0.17091 -0.05144 \n", "4 0.474878 -0.022947 -0.28589 0.06423 \n", "\n", "[5 rows x 980 columns]" ] }, "execution_count": 16, "metadata": {}, "output_type": "execute_result" } ], "source": [ "print(GE_CSV_matched.shape)\n", "\n", "GE_CSV_matched.head()" ] }, { "cell_type": "code", "execution_count": 17, "metadata": {}, "outputs": [ { "name": "stdout", "output_type": "stream", "text": [ "警告: Plate PAC035_U2OS_6H_X1_B1_UNI4445L 中没有其他化合物数据。跳过。\n", "警告: Plate PAC035_U2OS_6H_X2_B1_UNI4445L 中没有其他化合物数据。跳过。\n", "警告: Plate PAC035_U2OS_6H_X3_B1_UNI4445L 中没有其他化合物数据。跳过。\n", "警告: Plate PAC062_U2OS_6H_X1_B1_UNI4445L 中没有 DMSO 行，使用其他 Plate 的 DMSO 均值填充。\n", "警告: Plate PAC062_U2OS_6H_X2_B1_UNI4445L 中没有 DMSO 行，使用其他 Plate 的 DMSO 均值填充。\n", "警告: Plate PAC062_U2OS_6H_X3_B1_UNI4445L 中没有 DMSO 行，使用其他 Plate 的 DMSO 均值填充。\n" ] } ], "source": [ "import pandas as pd\n", "import numpy as np\n", "import h5py\n", "\n", "# 初始化存储结果的字典\n", "results_CP = {\n", " 'smiles': [], # 存储非 DMSO 的 smiles\n", " 'control': [], # 存储 DMSO 行对应的 672 维数据\n", " 'target': [] # 存储非 DMSO 行对应的 672 维数据\n", "}\n", "\n", "def process_plate(df):\n", " # 对当前 Plate 中的相同 SMILES 进行求均值\n", " # 显式添加 numeric_only=True 来消除警告\n", " return df.groupby('SMILES', as_index=False).mean(numeric_only=True)\n", "\n", "df = GE_CSV_matched\n", "# 按 det_plate 分组，对每个组应用 process_plate 函数\n", "processed_data_list = []\n", "for plate_name, plate_df in df.groupby('Plate'):\n", " processed_plate_df = process_plate(plate_df)\n", " # 为处理后的数据添加分组ID，以便后续使用\n", " processed_plate_df['Plate'] = plate_name\n", " processed_data_list.append(processed_plate_df)\n", "\n", "if processed_data_list:\n", " processed_data = pd.concat(processed_data_list, ignore_index=True)\n", "else:\n", " print(\"没有数据可供处理。\")\n", " exit()\n", "\n", "# 获取数值列的列名列表\n", "# 这将排除 det_plate 和 CPD_SMILES\n", "numeric_cols = [col for col in processed_data.columns if col not in ['Plate', 'SMILES', 'pert_dose']]\n", "\n", "# 按 Plate 分组\n", "grouped = processed_data.groupby('Plate')\n", "\n", "# 遍历每个 Plate\n", "for plate_name, plate_data in grouped:\n", " # 分离 DMSO 和非 DMSO 行\n", " na_mask = plate_data['SMILES'] == 'DMSO'\n", " na_data = plate_data[na_mask]\n", " non_na_data = plate_data[~na_mask]\n", "\n", " # 处理 DMSO 行（control）\n", " if not na_data.empty:\n", " # 使用显式列名列表来提取数据\n", " control = na_data[numeric_cols].values\n", " control = control.flatten()\n", " else:\n", " # 这里的警告提示了你数据中可能存在的问题\n", " print(f\"警告: Plate {plate_name} 中没有 DMSO 行，使用其他 Plate 的 DMSO 均值填充。\")\n", " # 可以用所有 DMSO 的平均值作为代替\n", " all_dmso_data = processed_data[processed_data['SMILES'] == 'DMSO']\n", " if not all_dmso_data.empty:\n", " control = all_dmso_data[numeric_cols].mean().values.flatten()\n", " else:\n", " # 如果整个数据集都没有 DMSO，用全零填充\n", " print(f\"警告: 整个数据集中都没有 DMSO，用全零填充 control。\")\n", " control = np.zeros(len(numeric_cols))\n", "\n", " # 处理非 DMSO 行（smiles 和 target）\n", " if not non_na_data.empty:\n", " smiles = non_na_data['SMILES'].tolist()\n", " # 使用显式列名列表来提取数据\n", " target = non_na_data[numeric_cols].values\n", "\n", " # 对齐 control 和 target：通过重复 control 使其长度与 target 一致\n", " control_aligned = np.tile(control, (target.shape[0], 1))\n", "\n", " # 存储结果\n", " results_CP['smiles'].extend(smiles)\n", " results_CP['control'].append(control_aligned)\n", " results_CP['target'].append(target)\n", " else:\n", " print(f\"警告: Plate {plate_name} 中没有其他化合物数据。跳过。\")\n", "\n", "# 将结果转换为 NumPy 数组并保存\n", "if results_CP['smiles']:\n", " results_CP['smiles'] = np.array(results_CP['smiles'], dtype='S256') # 使用 S256 来存储字符串\n", " results_CP['control'] = np.vstack(results_CP['control'])\n", " results_CP['target'] = np.vstack(results_CP['target'])\n", "\n", " # 输出结果，用逗号分隔元素\n", " # print(\"smiles:\", np.array2string(results_CP['smiles'], separator=', '))\n", " # print(\"control:\", np.array2string(results_CP['control'], separator=', '))\n", " # print(\"target:\", np.array2string(results_CP['target'], separator=', '))\n", "\n", " # 保存到 HDF5 文件\n", " # with h5py.File('./Processed_Paired_GE.h5', 'w') as f:\n", " # f.create_dataset('smiles', data=results_CP['smiles'])\n", " # f.create_dataset('control', data=results_CP['control'])\n", " # f.create_dataset('target', data=results_CP['target'])\n", "\n", " # print(\"数据已保存到 Processed_Paired_GE.h5 文件中。\")\n", "else:\n", " print(\"没有找到任何可用的数据来保存。\")" ] }, { "cell_type": "code", "execution_count": 18, "metadata": {}, "outputs": [], "source": [ "# 使用 astype('S') 转换为字节串，这是 HDF5 存储字符串的最佳方式\n", "final_smiles = np.array(results_CP['smiles'], dtype='S')\n", "\n", "# 9. 保存到 HDF5 文件\n", "output_path = './Processed_Paired_GE.h5'\n", "with h5py.File(output_path, 'w') as f:\n", " f.create_dataset('canonical_smiles', data=final_smiles)\n", " f.create_dataset('control', data=results_CP['control'], dtype=np.float32)\n", " f.create_dataset('target', data=results_CP['target'], dtype=np.float32)\n" ] }, { "cell_type": "code", "execution_count": 23, "metadata": {}, "outputs": [ { "data": { "text/plain": [ "(57797, 977)" ] }, "execution_count": 23, "metadata": {}, "output_type": "execute_result" } ], "source": [ "results_CP['control'].shape" ] }, { "cell_type": "code", "execution_count": 15, "metadata": {}, "outputs": [ { "data": { "text/plain": [ "array(['Brc1c(NC2=NCCN2)ccc2nccnc12',\n", " 'C/C=C1\\\\[C@H]2C=C(C)C[C@]1(N)c1ccc(=O)[nH]c1C2',\n", " 'C=C(C)[C@@H]1CC[C@]2(CO)CC[C@]3(C)[C@H](CC[C@@H]4[C@@]5(C)CC[C@H](O)C(C)(C)[C@@H]5CC[C@]43C)[C@@H]12',\n", " ..., 'c1ccc(-c2cnc(-c3ccccn3)nn2)cc1',\n", " 'c1ccc(Cn2ncc3c(N4CCN(c5ccccc5)CC4)ncnc32)cc1',\n", " 'c1ccc2c(Nc3nnc(-c4ccncc4)s3)cccc2c1'], dtype=' 0:\n", " print(f\"警告: 发现 {missing_count} 个样本所在的 Plate 没有 DMSO。使用全局 DMSO 中位数填充。\")\n", " # 填充 NaN\n", " # 构建一个临时的 DataFrame 来填充，列名要对应\n", " global_fill_values = pd.DataFrame([global_control], columns=ctrl_cols, index=[0])\n", " # 这里用一种高效的方式填充：直接赋值\n", " # 注意：pandas 的 fillna 对多列操作有时较慢，这里我们直接定位赋值\n", " for i, col in enumerate(ctrl_cols):\n", " merged_df.loc[missing_mask, col] = global_control[i]\n", "\n", " # 6. 提取最终数组\n", " # Target 数据 (原始特征)\n", " target_data = merged_df[data_cols].values.astype(np.float32)\n", " # Control 数据 (对应的板内中位数)\n", " control_data = merged_df[ctrl_cols].values.astype(np.float32)\n", " # SMILES\n", " smiles_data = merged_df['SMILES'].values.astype('S') # 转为字节串\n", " \n", " print(f\"{modality_name} 处理完毕。最终样本对数量: {len(smiles_data)}\")\n", " return smiles_data, control_data, target_data\n", "\n", "# ==========================================\n", "# 1. 处理 Cell Painting (CP)\n", "# ==========================================\n", "# 读取清洗后的数据\n", "cp_path = 'Step2_CP_cleaned.parquet'\n", "df_cp = pd.read_parquet(cp_path)\n", "\n", "# 识别特征列 (排除元数据)\n", "cp_meta_cols = ['SMILES', 'Plate', 'pert_dose']\n", "cp_features = [c for c in df_cp.columns if c not in cp_meta_cols]\n", "\n", "# 执行处理\n", "cp_smiles, cp_control, cp_target = process_alignment_keep_replicates(df_cp, cp_features, \"Cell Painting\")\n", "\n", "# 保存 CP 结果\n", "with h5py.File('./Processed_Paired_CP_KeepRep.h5', 'w') as f:\n", " f.create_dataset('canonical_smiles', data=cp_smiles)\n", " f.create_dataset('control', data=cp_control, dtype=np.float32)\n", " f.create_dataset('target', data=cp_target, dtype=np.float32)\n", "print(\"CP 数据已保存到 ./Processed_Paired_CP_KeepRep.h5\\n\")\n", "\n", "# ==========================================\n", "# 2. 处理 Gene Expression (GE)\n", "# ==========================================\n", "# 读取清洗后的数据\n", "ge_path = './Step2_GE_cleaned.parquet'\n", "df_ge = pd.read_parquet(ge_path)\n", "\n", "# 识别特征列\n", "ge_meta_cols = ['SMILES', 'Plate', 'pert_dose'] # 注意：你的 GE 数据里可能叫 det_plate 或 Plate，请确保之前 rename 过了\n", "ge_features = [c for c in df_ge.columns if c not in ge_meta_cols]\n", "\n", "# 执行处理\n", "ge_smiles, ge_control, ge_target = process_alignment_keep_replicates(df_ge, ge_features, \"Gene Expression\")\n", "\n", "# 保存 GE 结果\n", "with h5py.File('./Processed_Paired_GE_KeepRep.h5', 'w') as f:\n", " f.create_dataset('canonical_smiles', data=ge_smiles)\n", " f.create_dataset('control', data=ge_control, dtype=np.float32)\n", " f.create_dataset('target', data=ge_target, dtype=np.float32)\n", "print(\"GE 数据已保存到 ./Processed_Paired_GE_KeepRep.h5\")\n", "\n", "# ==========================================\n", "# 3. 验证数据形状\n", "# ==========================================\n", "print(\"-\" * 30)\n", "print(\"最终数据形状检查:\")\n", "print(f\"CP Smiles: {cp_smiles.shape}, Control: {cp_control.shape}, Target: {cp_target.shape}\")\n", "print(f\"GE Smiles: {ge_smiles.shape}, Control: {ge_control.shape}, Target: {ge_target.shape}\")" ] }, { "cell_type": "code", "execution_count": null, "metadata": {}, "outputs": [], "source": [] } ], "metadata": { "kernelspec": { "display_name": "boom", "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": "3.11.10" } }, "nbformat": 4, "nbformat_minor": 2 }