Spaces:
Sleeping
Sleeping
File size: 11,084 Bytes
adecc9b | 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60 61 62 63 64 65 66 67 68 69 70 71 72 73 74 75 76 77 78 79 80 81 82 83 84 85 86 87 88 89 90 91 92 93 94 95 96 97 98 99 100 101 102 103 104 105 106 107 108 109 110 111 112 113 114 115 116 117 118 119 120 121 122 123 124 125 126 127 128 129 130 131 132 133 134 135 136 137 138 139 140 141 142 143 144 145 146 147 148 149 150 151 152 153 154 155 156 157 158 159 160 161 162 163 164 165 166 167 168 169 170 171 172 173 174 175 176 177 178 179 180 181 182 183 184 185 186 187 188 189 190 191 192 193 194 195 196 197 198 199 200 201 202 203 204 205 206 207 208 209 210 211 212 213 214 215 216 217 218 219 220 221 222 | {
"cells": [
{
"cell_type": "markdown",
"metadata": {},
"source": [
"### Zero-shot cell type annotation\n",
"Given the gene expression profiles of the cells, as well as textual descriptions of alternative cell types, LangCell can automatically perform cell type annotation."
]
},
{
"cell_type": "code",
"execution_count": 1,
"metadata": {},
"outputs": [
{
"name": "stderr",
"output_type": "stream",
"text": [
"/AIRvePFS/dair/conda_envs/biomed/lib/python3.9/site-packages/tqdm/auto.py:21: TqdmWarning: IProgress not found. Please update jupyter and ipywidgets. See https://ipywidgets.readthedocs.io/en/stable/user_install.html\n",
" from .autonotebook import tqdm as notebook_tqdm\n",
"/AIRvePFS/dair/conda_envs/biomed/lib/python3.9/site-packages/transformers/deepspeed.py:23: FutureWarning: transformers.deepspeed module is deprecated and will be removed in a future version. Please import deepspeed modules directly from transformers.integrations\n",
" warnings.warn(\n"
]
}
],
"source": [
"import os\n",
"import sys\n",
"parent = os.path.dirname(os.path.abspath(''))\n",
"sys.path.append(parent)\n",
"os.chdir(parent)\n",
"from open_biomed.core.pipeline import InferencePipeline\n",
"from open_biomed.data import Cell, Text\n",
"from datasets import load_from_disk\n",
"import json\n",
"from open_biomed.data import Cell, Text\n",
"from sklearn.metrics import classification_report"
]
},
{
"cell_type": "code",
"execution_count": null,
"metadata": {},
"outputs": [
{
"name": "stderr",
"output_type": "stream",
"text": [
"05/23/2025 18:35:11 - INFO - root - The config of this process is:\n",
"{\n",
" \"model\": {\n",
" \"name\": \"langcell\",\n",
" \"cell_model\": \"./checkpoints/langcell/cell_bert\",\n",
" \"cell_proj\": \"./checkpoints/langcell/cell_proj.bin\",\n",
" \"text_tokenizer\": \"./checkpoints/langcell/pubmedbert-base\",\n",
" \"text_model\": \"./checkpoints/langcell/text_bert\",\n",
" \"text_proj\": \"./checkpoints/langcell/text_proj.bin\",\n",
" \"ctm_head\": \"./checkpoints/langcell/ctm_head.bin\"\n",
" },\n",
" \"task\": \"cell_annotation\",\n",
" \"model_ckpt\": \"\",\n",
" \"device\": \"cuda:2\",\n",
" \"logging_level\": \"info\"\n",
"}\n",
"Some weights of BertModel were not initialized from the model checkpoint at ./checkpoints/langcell/cell_bert and are newly initialized: ['pooler.dense.bias', 'pooler.dense.weight']\n",
"You should probably TRAIN this model on a down-stream task to be able to use it for predictions and inference.\n"
]
}
],
"source": [
"# Load the model\n",
"cfg_path = \"./configs/model/langcell.yaml\"\n",
"pipeline = InferencePipeline(model='langcell', task='cell_annotation', device='cuda:2')"
]
},
{
"cell_type": "code",
"execution_count": 7,
"metadata": {},
"outputs": [
{
"name": "stdout",
"output_type": "stream",
"text": [
"B cells ---- cell type: b cell. a lymphocyte of b lineage that is capable of b cell mediated immunity.; \n",
"\n",
"CD8 T cells ---- cell type: cd8-positive, alpha-beta t cell. a t cell expressing an alpha-beta t cell receptor and the cd8 coreceptor.; \n",
"\n",
"CD14+ Monocytes ---- cell type: cd14-positive monocyte. a monocyte that expresses cd14 and is negative for the lineage markers cd3, cd19, and cd20.; \n",
"\n",
"Dendritic Cells ---- cell type: dendritic cell. a cell of hematopoietic origin, typically resident in particular tissues, specialized in the uptake, processing, and transport of antigens to lymph nodes for the purpose of stimulating an immune response via t cell activation. these cells are lineage negative (cd3-negative, cd19-negative, cd34-negative, and cd56-negative).; \n",
"\n",
"NK cells ---- cell type: natural killer cell. a lymphocyte that can spontaneously kill a variety of target cells without prior antigenic activation via germline encoded activation receptors and also regulate immune responses via cytokine release and direct contact with other cells.; \n",
"\n",
"Megakaryocytes ---- cell type: megakaryocyte. a large hematopoietic cell (50 to 100 micron) with a lobated nucleus. once mature, this cell undergoes multiple rounds of endomitosis and cytoplasmic restructuring to allow platelet formation and release.; \n",
"\n",
"FCGR3A+ Monocytes ---- cell type: cd14-low, cd16-positive monocyte. a patrolling monocyte that is cd14-low and cd16-positive.; \n",
"\n",
"CD4 T cells ---- cell type: cd4-positive, alpha-beta t cell. a mature alpha-beta t cell that expresses an alpha-beta t cell receptor and the cd4 coreceptor.; \n",
"\n"
]
}
],
"source": [
"# Load the dataset\n",
"# Download data: https://drive.google.com/drive/folders/1cuhVG9v0YoAnjW-t_WMpQQguajumCBTp\n",
"dataset = load_from_disk('/path/to/pbmc10k.dataset')\n",
"type2text = json.load(open('/path/to/type2text.json'))\n",
"for type in type2text:\n",
" print(type, '----', type2text[type], '\\n')"
]
},
{
"cell_type": "code",
"execution_count": 8,
"metadata": {},
"outputs": [],
"source": [
"# random sample\n",
"dataset = dataset.shuffle(seed=42).select(range(2000))"
]
},
{
"cell_type": "code",
"execution_count": null,
"metadata": {},
"outputs": [],
"source": [
"# Organize data into specific formats as model inputs\n",
"texts = []\n",
"type2label = {}\n",
"labels = []\n",
"for type in type2text:\n",
" texts.append(Text.from_str(type2text[type]))\n",
" type2label[type] = len(texts) - 1\n",
"input = {'cell': [], 'class_texts': [], 'label': []}\n",
"for data in dataset:\n",
" input['cell'].append(Cell.from_sequence(data['input_ids']))\n",
" input['class_texts'].append(texts)\n",
" input['label'].append(type2label[data['str_labels']])\n",
" labels.append(type2label[data['str_labels']])"
]
},
{
"cell_type": "code",
"execution_count": null,
"metadata": {},
"outputs": [
{
"name": "stderr",
"output_type": "stream",
"text": [
"Inference Steps: 0%| | 0/2000 [00:00<?, ?it/s]/AIRvePFS/dair/luoyz-data/projects/OpenBioMed/OpenBioMed_arch/open_biomed/models/cell/langcell/langcell_utils.py:1000: UserWarning: To copy construct from a tensor, it is recommended to use sourceTensor.clone().detach() or sourceTensor.clone().detach().requires_grad_(True), rather than torch.tensor(sourceTensor).\n",
" batch = {'cell': torch.tensor(batch['input_ids'], dtype=torch.int64),\n",
"/AIRvePFS/dair/luoyz-data/projects/OpenBioMed/OpenBioMed_arch/open_biomed/models/cell/langcell/langcell_utils.py:1001: UserWarning: To copy construct from a tensor, it is recommended to use sourceTensor.clone().detach() or sourceTensor.clone().detach().requires_grad_(True), rather than torch.tensor(sourceTensor).\n",
" 'attention_mask': torch.tensor(batch['attention_mask'], dtype=torch.int64),\n",
"Inference Steps: 100%|██████████| 2000/2000 [03:56<00:00, 8.44it/s]\n"
]
}
],
"source": [
"# Predict the cell type of each cell using the model\n",
"preds, _ = pipeline.run(batch_size=1, **input)\n",
"preds = [p.item() for p in preds]"
]
},
{
"cell_type": "code",
"execution_count": null,
"metadata": {},
"outputs": [
{
"name": "stdout",
"output_type": "stream",
"text": [
" precision recall f1-score support\n",
"\n",
" B cells 1.00 0.98 0.99 279\n",
" CD8 T cells 0.59 0.95 0.73 260\n",
" CD14+ Monocytes 0.96 0.99 0.98 387\n",
" Dendritic Cells 1.00 0.82 0.90 67\n",
" NK cells 0.82 0.98 0.90 57\n",
" Megakaryocytes 0.82 0.90 0.86 20\n",
"FCGR3A+ Monocytes 1.00 0.82 0.90 66\n",
" CD4 T cells 0.98 0.81 0.89 864\n",
"\n",
" accuracy 0.89 2000\n",
" macro avg 0.90 0.91 0.89 2000\n",
" weighted avg 0.93 0.89 0.90 2000\n",
"\n"
]
}
],
"source": [
"# Analyze the results\n",
"print(classification_report(labels, preds, labels=range(len(type2text)), target_names=type2text.keys()))"
]
}
],
"metadata": {
"kernelspec": {
"display_name": "biomed",
"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": "3.9.7"
}
},
"nbformat": 4,
"nbformat_minor": 2
} |