spot_quant

Quantification of fluorescent signals within 3D or Z-projected masks.

Installation

Installation instructions are given for pixi, but you can also use conda/mamba.

1. If you don't have pixi yet, install it from:https://pixi.prefix.dev/latest/installation/.
2. Navigate to where you want to place the repository.
3. Clone the repository: git clone https://github.com/CellFateNucOrg/spot_quant/.
4. Install the required packages: pixi install.

Use

To do the quantification, first configure the parameters in spot_quant.sh:

• src_dirs=(): List of one or more directories (separated by space or line) with raw images you want to quantify.
• filter_out=(): Expression(s) (separated by space or line) to filter out specific files. If your input directory contains other files that should not be quantified, make sure to use this option to filter them out.
• props_c: Which channel (starting from zero) to quantify.
• ref_c: Which channel was used for the segmentation (i.e., to make the masks).
• masks_folder: In which folder the masks are located, relative to the input directory containing the raw images.
• mask_str: Suffix that differentiates the mask from its corresponding image (e.g., _mask.tif).
• do_3d: Whether to quantify within the full 3D mask (true or false).
• do_mip: Whether to quantify within a Z-projection of each mask (true or false).
• out_folder: Name of a subfolder in which the output is save (optional).
• min_planes: Filter used masks based on a minimum number of planes. Set to zero if not specified.
• pixi_dir: Path to this repository (including the folder spot_quant itself).

Then run the script by typing: sbatch spot_quant.sh.

Output

Measured properties are stored in files named props.json and/or mip_props.json, which are structured as follows:

[   
    {
        "image": filename (str),
        "measured_channel": value passed as 'props_c' (int),
        "min_planes": value passed as 'min_planes' (int),
        "time_points": [
            {
                "time_point": time point (int),
                "props": [
                    {
                        "label": label of the region (int),
                        "n_components": 1,
                        "bbox": [
                            minimum Z coordinate (int),
                            minimum Y coordinate (int),
                            minimum X coordinate (int),
                            maximum Z coordinate (int),
                            maximum Y coordinate (int),
                            maximum X coordinate (int)
                        ],
                        "z_index": position of the region in the output .tif file (after sorting all bboxes in ascending order),
                        "area": regionprops area (float),
                        "diameter": regionprops equivalent_diameter (float),
                        "solidity": solidity, measured by regionprops for 3D images or calculated manually for 2D images (float),
                        "intensity_mean": mean value of all pixels in area (float),
                        "intensity_std": standard deviation of all pixel values in area (float)
                    },
                    {
                        next label...
                    },
                    ...
                ]
            },
            {
                next time point...
            },
            ...

        ]
    },
    {
        next image...
    },
    ...
]

Put into words: the output .json file contains a list of objects; each object specifies the filename, which channel was measured, the minimum number of planes, and a key named "time_points", which is assigned a list, containing one object per time point; each such object specifies the time point and a key named "props", which is assigned a list of objects; each such object corresponds to a label (i.e., a a single segmentation) and its measurements.

In addition, for each image, a stack containing only the measured regions (3D and/or Z-stacked projections) is saved as a .tif file.