Panel maker

Installation

git clone https://github.com/hatch-lab/panel-maker
python3 -m venv .venv
source .venv/bin/activate
pip install --upgrade pip
pip install -r requirements.txt

Usage

The basic usage pattern is to have a folder of individual TIFF files. This tool will combine those images into a single panel and, by default, generate a merged image. Images will be added left-to-right, top-to-bottom in alphabetical order.

Basic usage

python make-panel.py /folder/of/tiff

Output are 3 files: panel.tif, panel.jpg, and meta.json. The first two are TIFF and JPEG versions of your panel, and the last is a text file containing information about how the panel was generated.

You can also stack multiple rows by providing multiple input folders and an output folder:

python make-panel.py /folder/of/tiff /folder/of/tiff2 --out=/combined

Options

Levels

To set minimum and maximum intensity levels, use --min and --max. You may specify any integer 0–255. Each image in the panel can have its own levels: the 2nd image's levels are controlled by the 2nd instances of --min and --max.

python make-panel.py /folder/of/tiff --min=10 --max=120 --max=250

This will rescale the first image’s levels to be between 10 and 120 and the second image to be 0–250. No other images will be adjusted.

Channel labels

By default, channels will be unlabeled. To specify channel labels, use --channel:

python make-panel.py /folder/of/tiff --channel="TagRFP" --channel="GFP" --channel="DAPI"

This will set the first image’s label to DAPI, the second to GFP, and third to TagRFP.

Row labels

By default, each row of inputs will be unlabelled. But, if you have multiple conditions, it can be helpful to label them, using the --label option:

python make-panel.py /folder/of/tiff /folder/of/tiff2 --out=/combined --channel="TagRFP" --channel="GFP" --channel="DAPI" --label="Condition 1" --label="Condition 2"

Colors

If your panel has a merged image (the default), you can specify the colors with --color. This specifies which LUT to use, among those LUTs that are bundled with this program. By default, the last channel will be gray, the second from last will be cyan, the third from last will be magenta, and the fourth from last will be yellow. These were developed by Amanda. Alternatively, there are LUTs I took from: https://github.com/nelas/color-blind-luts, with the names:

• blue
• magenta
• orange
• red
• turq
• yellow

For example:

python make-panel.py /folder/of/tiff --channel="TagRFP" --channel="GFP" --channel="DAPI" --color=magenta --color=red --color=blue

Merged image

By default, all images are combined into a color merge, using the colors specified by --color. If you don’t want to make a merged image, use --skip-merge.

python make-panel.py /folder/of/tiff --skip-merge

No merged image will be created.

Scale bar

By default, a 20 µm scale bar is added to the bottom right of an image. You can specify the width of the scale bar with bar-microns.

python make-panel.py /folder/of/tiff --bar-microns=10

This will add a 10 µm scale bar.

Padding

By default, 10 px of white space are added between each image. You can change this with --padding.

python make-panel.py /folder/of/tiff --padding=0

Images will have no white space between them.

Zoom

If you would like to zoom in rather than showing the entire field, you can use the --zoom option. By default, it will zoom into the center of the field. You can change this with the --zoom-anchor option.

python make-panel.py /folder/of/tiff --channel="TagRFP" --channel="GFP" --channel="DAPI" --zoom=1.5 --zoom-anchor=rt`

zoom-anchor takes the following values:

• lt: The left-top of the field
• mt: Middle-top
• rt: Right-top
• lm: Left-middle
• mm: Center (the default)
• rm: Right-middle
• lb: Left-bottom
• mb: Middle-bottom
• rb: Right-bottom

If you have multiple inputs, you can specify multiple zoom-anchor values:

python make-panel.py /folder/of/tiff /folder/of/tiff2 --out=/combined --channel="TagRFP" --channel="GFP" --channel="DAPI" --zoom=1.5 --zoom-anchor=rt --zoom-anchor=lb

Pixels per micron

By default, the pixel to micron conversion ratio is extracted from the first TIFF’s metadata. However, this may not always be accurate. To specify the pixels/micron directly, use --pixels-per-um.

python make-panel.py /folder/of/tiff --pixels-per-micron=0.577

The scale bar will be adjusted assuming that 1 µm = 0.577 px.

Help

If you need help, you can run:

python make-panel.py --help