-
Notifications
You must be signed in to change notification settings - Fork 0
Usage
Directory Path - Click "Browse" and select the directory containing all of the exported .csv files you want to process. They must all be placed into one directory. You must export the data for each channel as a .csv, CLAMS Wrangler can not process the raw data formats (.ADTA, .BDTA, .CDTA, .MDTA, .PDTA).
Trim Hours - This is the number of hours you want to trim from the beginning of your data. For most CLAMS studies, somewhere in the range of 1 to 6 hours should be sufficient to allow the animals to acclimate to the system. If you have a very short CLAMS run, then you may want to skip trimming any.
Trim hours has an additional option that allows you to define what light cycle to start in following the trim, which is either the light (LED LIGHTNESS = 100) or dark (LED LIGHTNESS = 100) cycle. Select "Start Light" or "Start Dark", respectively. If you do not select an option, it will just start at whatever the next light cycle is.
Keep Hours - This is the number of hours you want to retain following the trimming. For most CLAMS studies, at least 72 hours is a good starting point.
Bin Hours - This is the size of the data bins in hours. It must be a factor of 12 (1, 2, 3, 4, 6, 12). This option lets you quickly explore different bin sizes and potential impacts on data interpretation.
Config File - Click "Browse" to select a config file. If there is an existing one, you can select that. If there is no existing config file, the program will create a config directory in the location of your .csv files with a .xlsx file that has an ID and GROUP LABEL column. Fill out the ID and corresponding group label (e.g. 'Control', 'Treatment') for each subject and save.
Alternatively, you can manually enter the ID and group label using the input boxes in the CLAMS Wrangler GUI.
After all setting are configured, simply click Start Processing to run the program. If there are errors they will output to the text box above the button.
CLAMS Wrangler will create several folders containing various states of processed data. Most states are of little interest but are retained for quality checking purposes.
Below is sample directory tree starting in the folder with all of your .csv files to process.
.
├── Subject_1.csv
├── Subject_2.csv
├── Subject_3.csv
├── Subject_4.csv
├── Subject_5.csv
├── Subject_6.csv
└── timestamp_2023-08-10_20-19-56
├── Binned_CLAMS_data
│ ├── Subject_1_ID101_trimmed_binned.csv
│ ├── Subject_2_ID102_trimmed_binned.csv
│ ├── Subject_3_ID103_trimmed_binned.csv
│ ├── Subject_4_ID104_trimmed_binned.csv
│ ├── Subject_5_ID105_trimmed_binned.csv
│ ├── Subject_6_ID106_trimmed_binned.csv
├── Cleaned_CLAMS_data
│ ├── Subject_1_ID101.csv
│ ├── Subject_2_ID102.csv
│ ├── Subject_3_ID103.csv
│ ├── Subject_4_ID104.csv
│ ├── Subject_5_ID105.csv
│ ├── Subject_6_ID106.csv
├── Combined_CLAMS_data
│ ├── Reformatted_CSVs
│ │ ├── reformatted_ACCCO2.csv
│ │ ├── reformatted_ACCO2.csv
│ │ ├── reformatted_AMB ACC.csv
│ │ ├── reformatted_AMB.csv
│ │ ├── reformatted_FEED1 ACC.csv
│ │ ├── reformatted_FEED1.csv
│ │ ├── reformatted_RER.csv
│ │ ├── reformatted_VCO2.csv
│ │ ├── reformatted_VO2.csv
│ │ ├── reformatted_WHEEL ACC.csv
│ │ └── reformatted_WHEEL.csv
│ ├── ACCCO2.csv
│ ├── ACCO2.csv
│ ├── FEED1 ACC.csv
│ ├── FEED1.csv
│ ├── RER.csv
│ ├── TOT AMB.csv
│ ├── VCO2.csv
│ ├── VO2.csv
│ ├── WHEEL ACC.csv
│ └── WHEEL.csv
├── config
│ ├── experiment_config.csv
│ └── log_2023-08-10_20-19-56.txt
└── Trimmed_CLAMS_data
├── Subject_1_ID101_trimmed.csv
├── Subject_2_ID102_trimmed.csv
├── Subject_3_ID103_trimmed.csv
├── Subject_4_ID104_trimmed.csv
├── Subject_5_ID105_trimmed.csv
└── Subject_6_ID106_trimmed.csv
The contents of folders Cleaned_CLAMS_data, Trimmed_CLAMS_data, and Binned_CLAMS_data are relatively straight forward and not directly used in any analysis.
The Combined_CLAMS_data folder contains the combined data from all of your subjects, separated by CLAMS variables of interest.
But the main output is located in the ./Combined_CLAMS_data/Reformatted_CSVs directory. This is where all of the variables of interest (e.g. RER, VO2, VCO2) are reformatted using the experimental configuration file.
The format of these .csvs are intended to be input directly into statistics programs for rapid analysis. At the moment, this formatted data structure has only been tested on JASP, which is a free and open-source statistics tool that resembles SPSS. Direct support of output formats relevant to other statistics programs may be added in the future.
TODO: provide technical details on how various settings affect the data processing.