CaImAn implements a set of essential methods required in the analysis pipeline of large scale calcium imaging data. Fast and scalable algorithms are implemented for motion correction, source extraction, spike deconvolution, and component registration across multiple days. It is suitable for both two-photon and one-photon fluorescence microscopy data. The implementation of CaImAn given here was created by NeuroCAAS developers to benchmark performance of our platform. To try out CaImAn, look at the template job provided. Data (N.01.01.zip, images_YST.zip) is provided from the CaImAn paper.
The NeuroCAAS implementation of CaImAn runs an integrated pipeline equivalent to the 'fit_file' method provided in the repo, including motion correction. We offer several workflow options, including component evaluation to eliminate false positives after fitting, and model refitting using the output of an initial run as initialization. Users can write CaImAn parameters into the config file by hand, or provide a path to a CaImAn parameter dictionary in pickled format.
Args:
-Input: (zip file) Either A) a (.zip) archive containing .tif files representing the dataset to be analyzed or B) a (.tar.gz) file containing the directory and tif files you want to analyze. IMPORTANT: If working with (.zip) archives, zip ONLY the tif files you want to analyze, NOT the folder (i.e. select files from inside the folder). If the archive contains the folder itself, or invisible mac os files, analysis will fail.
-Config: (json) A json file containing two important sets of parameters.
1) Workflow Parameters: Determine how many times to fit the data, whether or not to evaluate for false positives at the end of analysis.
2) Compute Parameters: CaImAn parameters used in fitting models to provided data. See provided template file for an example, and 'https://caiman.readthedocs.io/en/master/Getting_Started.html#parameters' for more info.
Outputs:
-(hdf5) Full model fit from the data. Can be loaded into a local implementation of CNMF and queried for estimates of various model parameters with CaImAn native model loading functions. See CaImAn documentation for details on how to use this model. (https://caiman.readthedocs.io/en/master/Getting_Started.html#result-interpretation).
-(npz) Individual files that provide the estimates of spatial components (A), temporal components (C), and residuals (R) from processing, saved for convenience as numpy archives. Can be inspected with standard numpy loading functions .
You must login to use an analysis.