ageas.Launch

class ageas.Launch(model_config_path: Optional[str] = None, mute_unit: bool = True, protocol: str = 'solo', unit_num: int = 1, warning_filter: str = 'ignore', correlation_thread: float = 0.2, database_path: Optional[str] = None, database_type: str = 'gem_files', class1_path: Optional[str] = None, class2_path: Optional[str] = None, interaction_database: str = 'gtrd', log2fc_thread: Optional[float] = None, meta_load_path: Optional[str] = None, mww_p_val_thread: float = 0.05, normalize: Optional[str] = None, prediction_thread='auto', psgrn_load_path: Optional[str] = None, specie: str = 'mouse', sliding_window_size: int = 100, sliding_window_stride: Optional[int] = None, std_value_thread: float = 1.0, std_ratio_thread: Optional[float] = None, clf_keep_ratio: float = 0.5, cpu_mode: bool = False, feature_dropout_ratio: float = 0.1, feature_select_iteration: int = 3, top_grp_amount: int = 100, grp_changing_thread: float = 0.05, model_select_iteration: int = 3, outlier_thread: float = 3.0, regulatory_trace_depth: int = 1, stabilize_patient: int = 3, stabilize_iteration: int = 10, max_train_size: float = 0.95, z_score_extract_thread: float = 0.0)

Object containing basic pipeline to launch AGEAS.

Results are stored in attributes and can be saved as files.

__init__(model_config_path: Optional[str] = None, mute_unit: bool = True, protocol: str = 'solo', unit_num: int = 1, warning_filter: str = 'ignore', correlation_thread: float = 0.2, database_path: Optional[str] = None, database_type: str = 'gem_files', class1_path: Optional[str] = None, class2_path: Optional[str] = None, interaction_database: str = 'gtrd', log2fc_thread: Optional[float] = None, meta_load_path: Optional[str] = None, mww_p_val_thread: float = 0.05, normalize: Optional[str] = None, prediction_thread='auto', psgrn_load_path: Optional[str] = None, specie: str = 'mouse', sliding_window_size: int = 100, sliding_window_stride: Optional[int] = None, std_value_thread: float = 1.0, std_ratio_thread: Optional[float] = None, clf_keep_ratio: float = 0.5, cpu_mode: bool = False, feature_dropout_ratio: float = 0.1, feature_select_iteration: int = 3, top_grp_amount: int = 100, grp_changing_thread: float = 0.05, model_select_iteration: int = 3, outlier_thread: float = 3.0, regulatory_trace_depth: int = 1, stabilize_patient: int = 3, stabilize_iteration: int = 10, max_train_size: float = 0.95, z_score_extract_thread: float = 0.0)

Pipeline to launch AGEAS.

Parameters

• model_config_path –

  <str Default = None> Path to load model config file which will be used to initialize classifiers.

  By default, AGEAS will use internalized model config file which contians following model types:

  Transformer

  Random Forest(RFC)

  Support Vector Machine(SVM)

  Gradient Boosting Machine(GBM)

  Convolutional Neural Network(CNN_1D, CNN_Hybrid)

  Recurrent Neural Network(RNN)

  Long Short Term Memory(LSTM)

  Gated Recurrent Unit(GRU)

• mute_unit – <bool Default = True> Whether AGEAS unit print out log while running. It is not mandatory but encouraged to remain True especially when using ‘multi’ protocol.

• protocol –

  <str Default = ‘solo’> AGEAS unit launching protocol.

  Supporting:

  ’solo’: All units will run separately.

  ’multi’: All units will run parallelly by multithreading.

• unit_num – <int Default = 1> Amount of AGEAS extractor units to launch.

• warning_filter – <str Default = ‘ignore’> How warnings should be filtered. For other options, please check ‘The Warnings Filter’ section in: https://docs.python.org/3/library/warnings.html#warning-filter

Additional Parameters:

All args in ageas.Data_Preprocess()

All args in ageas.Unit() excluding database_info, meta, model_config, and pseudo_grns,

Attributes:

self.atlas

self.meta

self.pseudo_grns

Methods

__init__([model_config_path, mute_unit, ...])	Pipeline to launch AGEAS.
save_reports([folder_path, network_header, ...])	Save meta processed GRN, pseudo-sample GRNs, meta-GRN based analysis report, AGEAS based analysis report, and key atlas extracted by AGEAS.