drcme.bin.run_spca_fit¶

Script to run sparse principal component analysis on electrophysiology feature vectors.

The electrophysiology feature vectors used as inputs are in the form processed by the IPFX package. An optional metadata file can be given as an input to filter the cells in the data set.

The script produces several outputs in the specified output_dir, named with a specified output_code.

sparse_principal_components_[output_code].csv: sPCA values for each cell.
spca_components_used_[output_code].json: List of kept components for each data subset.
spca_loadings_[output_code].pkl: sPCA loadings, adjust explained variance, and transformed values for each data subset. Uses the joblib library for saving/loading.

class drcme.bin.run_spca_fit.DatasetParameters(extra=None, only=None, exclude=(), prefix='', strict=None, many=False, context=None, load_only=(), dump_only=(), partial=False)[source]¶

Parameter schema for input datasets

DatasetParameters¶
key	description	default	field_type	json_type
fv_h5_file	HDF5 file with feature vectors	NA	`InputFile`	str
metadata_file	Metadata file in CSV format	None	`InputFile`	str
dendrite_type	Filter for dendrite type using information in metadata (all, spiny, aspiny)	all	`String`	str
allow_missing_structure	Whether or not structure value for cell in metadata can be missing	False	`Boolean`	bool
allow_missing_dendrite	Whether or not dendrite type value for cell in metadata can be missing	False	`Boolean`	bool
need_ramp_spike	Whether or not to exclude cells that did not fire an action potential from the ramp stimulus	True	`Boolean`	bool
limit_to_cortical_layers	List of cortical layers to limit the data set (using the metadata file)	[]	`List`	str
id_file	Text file with specimen IDs to use. Cells with IDs not in the file will be excluded.	None	`InputFile`	str

class drcme.bin.run_spca_fit.AnalysisParameters(extra=None, only=None, exclude=(), prefix='', strict=None, many=False, context=None, load_only=(), dump_only=(), partial=False)[source]¶

Parameter schema for sPCA analysis

This schema is designed to be a schema_type for an ArgSchemaParser object

AnalysisParameters¶
key	description	default	field_type	json_type
input_json	file path of input json file	NA	`InputFile`	str
output_json	file path to output json file	NA	`OutputFile`	str
log_level	set the logging level of the module	ERROR	`LogLevel`	str
params_file	JSON file with sPCA parameters	NA	`InputFile`	str
output_dir	Directory for output files	NA	`OutputDir`	str
output_code	Code for naming output files	NA	`String`	str
datasets	Schema for loading one or more specific datasets for the analysis	(REQUIRED)	`DatasetParameters`	list

Functions

main(params_file, output_dir, output_code, …)

Main runner function for script.

Classes

`AnalysisParameters`([extra, only, exclude, …])	Parameter schema for sPCA analysis
`DatasetParameters`([extra, only, exclude, …])	Parameter schema for input datasets