--- language: - en pipeline_tag: tabular-classification tags: - computational neuroscience --- To reduce the effort in manual curation, we developed a machine learning approach using Neuropixels probes, incorporating quality metrics to automatically identify noise clusters and isolate single-cell activity. Compatible with the Spikeinterface API, our method generalizes across various probes and speices. The we generated a machine learning model that is trained on 11 mice in V1, SC and ALM using Neuropixels on mice. Each recording was labelled by at least two people, and in different combinations. The agreement amongst labellers is 80%. You can use the models "noise_meuron_model.skops" that is used to identify noise and "sua_mua_model.skops" that is used to isolate SUA. There are two tutorial notebooks: 1. Model_based_curation.ipynb This notebook helps you apply pretrained models to new recordings. Simply load the models and use them to label your spike-sorted data. ``` python from spikeinterface.curation import auto_label_units labels = auto_label_units( sorting_analyzer = sorting_analyzer, model_folder = “SpikeInterface/a_folder_for_a_model”, trusted = [‘numpy.dtype’]) ``` 2. Train_new_model.ipynb If you have your own manually curated data (e.g., from other species), this notebook allows you to train a new model using your specific data. ``` python from spikeinterface.curation.train_manual_curation import train_model trainer = train_model(mode = "analyzers", labels = labels, analyzers = [labelled_analyzer, labelled_analyzer], output_folder = str(output_folder), imputation_strategies = None, scaling_techniques = None, classifiers = None, # Default to Random Forest only. Other classifiers you can try [ "AdaBoostClassifier","GradientBoostingClassifier", # "LogisticRegression","MLPClassifier","XGBoost","LightGBM", "CatBoost"] ) ```