Apr 13, 2026
Perineuronal Net Mapping and Quantification with Quint
- Joseph Martin1,
- Matthew Calhoun1,
- Jacob Elrod1,
- Madison Gossard1,
- Devin Naredo1,
- Logan Dunn1,
- Keerthi Krishnan1
- 1University of Tennessee, Knoxville
Protocol Citation: Joseph Martin, Matthew Calhoun, Jacob Elrod, Madison Gossard, Devin Naredo, Logan Dunn, Keerthi Krishnan 2026. Perineuronal Net Mapping and Quantification with Quint. protocols.io https://dx.doi.org/10.17504/protocols.io.261ge1ddyv47/v1
License: This is an open access protocol distributed under the terms of the Creative Commons Attribution License, which permits unrestricted use, distribution, and reproduction in any medium, provided the original author and source are credited
Protocol status: Working
We use this protocol and it's working
Created: January 14, 2026
Last Modified: April 13, 2026
Protocol Integer ID: 238630
Keywords: Quint, Whole-brain analysis, Perineuronal nets, neuroscience, PNNs, QuickNII, VisuAlign, Nutil, MeshView, rodent brain section image, perineuronal net mapping, comprehensive characterization of perineuronal net expression, distribution of perineuronal net, perineuronal net expression, perineuronal net, standardized brain atlas, brain region, segmentation in ilastik, quantification of histological feature, brain in both mice, segmentation, histological feature, quint pipeline, mapping in quicknii, supporting segmentation, quint, quantification with quint, brain, sagittal section image, mapping
Funders Acknowledgements:
NSF (KK)
Grant ID: 2336907
UTK URF AURA (JM, DN)
Grant ID: N/A
UTK URF FRAF (JM)
Grant ID: N/A
Disclaimer
This protocol was written in the Krishnan lab by Joseph Martin, but all software and original protocols were developed and are available through the NeuroImaging Tools and Resources Collaboratory (NITRC).
Abstract
The Quint pipeline enables efficient, semi-automated alignment of rodent brain section images with standardized brain atlases, while also supporting segmentation and quantification of histological features. This protocol outlines our approach to analyzing the distribution of perineuronal nets across the brain in both mice and rats, using coronal and sagittal section images. The workflow begins with mapping in QuickNII and VisuAlign, followed by segmentation in Ilastik and quantification in Nutil. The resulting data enable comprehensive characterization of perineuronal net expression across brain regions.
Original Documentation and Other Protocols
Original Quint documentation: https://quint-workflow.readthedocs.io/en/latest/QUINTintro.html
Other protocols using Quint include: https://www.protocols.io/view/regional-mouse-brain-analysis-modified-quint-kqdg3xbkzg25/v1
Videos of selected steps (also linked individually throughout the protocol): https://youtube.com/playlist?list=PLQCli8HoaKipZD4ZB95tL_6uU-TQ4X3_G&si=UMTu8uUnU0Kypumb
Overview of Process
- Initial image prep (PowerRename, Nutil, FileBuilder)
- Mapping (QuickNII)
- Non-linear transformation (VisuAlign)
- Pixel Classification (Ilastik)
- Object Classification (Ilastik, Fiji)
- Quantification (Nutil, Scripts)
- 3D Visualization (MeshView, optional)
Software Installation
1. Create a "Quint" folder on your C drive for all software installation
2. Download all software from the following links:
- https://learn.microsoft.com/en-us/windows/powertoys/install (We used PowerToysUserSetup-0.85.1-x64)
- https://www.nitrc.org/projects/nutil/ (We used nutil_win_v.0.8.0)
- https://www.nitrc.org/projects/quicknii (See below)
- https://www.nitrc.org/projects/visualign (We used VisuAlign-v0_9)
- https://www.ilastik.org/documentation/basics/installation (We used version 1.2.0.post1-gpu)
Make sure to install the appropriate QuickNII atlas version for your images
Mice: We used ABAMouse-v3-2017
Rats: We used WHSRat-v4
3. Unzip all downloads within the "Quint" folder that you created
4. If desired, create desktop shortcuts for each software program or pin them to the task bar
5. Open all software programs to ensure that they are functional
Note: The preceding steps only need to be completed the first time you use Quint. Each subsequent time, you can begin with Section 1 below.
1. Initial Image Prep
1. Download and duplicate a copy of the "QUINT Template Folders found here:
QUINT Template Folders-20260205T120135Z-1-001.zip9.1KB
QUINT Template Folders-20260205T120135Z-1-001.zip9.1KB Note: for sagittal sections, you will need both "Left" and "Right" folders that each contain the Quint template folders. They will essentially be treated as separate brains for the purpose of Quint.
2. Obtain images (Likely TIFFs)
Example of file nomenclature at this point: #111722_NW_PNN_1.1.1.tif
3. Put images in “A. Original TIFFs”
Note: If images were excluded for quality control purposes, the images that you should map will be in a
folder called “AA. Original TIFFs, Images Excluded”
4. Ensure that images are ordered and in the correct orientation
Coronal: ordered rostral to caudal, dorsal on the top, left on the left
Sagittal: ordered lateral to medial, dorsal on the top, rostral on the left
Example of file nomenclature at this point: 01#111722_NW_PNN_1.1.1.tif
5. Copy images from A. (or AA.) and paste them into "B. Renamed TIFFs"
Rename images to ensure they are compatible with Quint software
- Select all images and right-click
- Select "Rename with PowerRename"
- Use PowerRename to replace all periods, hyphens, and hashtags in the file name with an underscore (uses a "find and replace"-based interface"
Note: ensure that "apply to" is set to "Filename only" to avoid corrupting files
4. Add a serialized number in the form "_sXXX" to the end of each file name (example: _s001 for the first image) – possible with PowerRename
- Change "apply to" to "Filename + extension"
- Type ".tif" in the top box and "_s${padding=3,start=1}.tif in the bottom box
- Click apply once
Example of file nomenclature at this point: 01_111722_NW_PNN_1_1_1_s001.png
Resize images (downsizing so they will be compatible with Quint requirements)
- Open Nutil
- Select "Operation" – "New" – "Resize"
- Name the project according to the cohort/condition you are working on
- Use "B. Renamed TIFFs" as the Input Folder
- Use "C. PNGs (with section numbers)" as the Output Folder
- Make sure "Percent" is selected as the Resize Type
- Set the Resize Factor to 85 (coronal) or 60 (sagittal)
- Select "Save As," name the resize file, then save it to "C. PNGs (with section numbers)"
- Click "Start"
- Open FileBuilder (located in the QuickNII download folder)
- Navigate to "C. PNGs (with section numbers)" within your copy of the Quint template folders, select all images (CTRL + A), and click "Open"
- Ensure that all remarks in the rightmost column are either green or orange (NOT red)
- If any remarks are red, repeat steps 1-12, decreasing the Resize Factor by 10 (75 for coronal or 50 for sagittal)
- Repeat steps 1-13 as needed until all remarks are either green or orange
13. Select "Save XML" name the file according to the cohort/condition, and save it to "D. QuickNII"
14. Copy and paste all PNGs from "C." to "D." – if your images don't show up in QuickNII, you probably forgot to do this
2. Mapping (QuickNII)
Best Practices for Successful QuickNII Mapping
1. Use the following sections for general reference throughout the mapping process:
- Mapping Landmarks (Coronal)
- Mapping Landmarks (Sagittal)
- Overall Mapping Strategy
2. Use the following sections for specific steps to follow while mapping:
- Opening and Configuring QuickNII
- Mapping Mechanics (Coronal)
- Mapping Mechanics (Sagittal)
- Finishing Up Mapping
3. Refer back to general reference sections as needed for clarification
4. Check your work (with another person if at all possible) before moving on to VisuAlign. After QuickNII, images should be mostly mapped, with all structures being roughly the same size when you compare images to the atlas. Slight deviation in alignment of regions between images and the atlas is acceptable in QuickNII and correctable later with VisuAlign.
Mapping Landmarks (Coronal)
- Anterior Commissure
- Lateral Ventricles
- Hippocampus
Mapping Landmarks (Sagittal)
- Hippocampus
- Piriform cortex
Overall Mapping Strategy
1. Start mapping by choosing approximately 3 images with clear landmarks to map initially (ideally equally distributed across the rostrocaudal/mediolateral axis)
2. Map the image (see subsequent steps for Mapping Mechanics)
3. Select "Store" (top left) to save your mapped image and apply changes to remaining unmapped images.
4. Continue until all images are mapped
Opening and Configuring QuickNII
- Open QuickNII (use appropriate version for species (mouse/rat))
- Select "Manage Data" (top right)
- Open the mapping window (first window that pops up when you open QuickNII) on your left monitor, and move the data management window (which pops up when you click "Manage Data") to your right monitor
- In the window that pops up, select "Load" (bottom left)
- Select and open the XML file from your "D." folder
- In the data management window, double click on an image's filename to prepare to map it. This should make it appear in the mapping window. If an image does not appear, make sure that images are present in the "D. QuickNII" folder. This is a common mistake.
- Click on "Values and Control" on the bottom left of the mapping window. If this isn't visible, try hovering your mouse over the bottom left of the mapping window, and an option to drag a menu up from the bottom should appear.
7. Sagittal only: change "Coronal" to "Sagittal"
8. If your images are generally dark, you can use the "Image Darkness" slider to brighten the image
9. Mouse only: To see the outlines of regions, select "Outline" and change your atlas type to either "Rainbow 2017" or "Original 2017"
10. Sagittal only: Use the bottommost window on the right side of the screen to change lateral/medial positioning. Ensure that left hemisphere tissues are mapped on the left side of the atlas and right hemisphere tissues are mapped on the right side of the atlas.
11. Sagittal only: Change the angles in the red (coronal), green (horizontal), and blue (sagittal) boxes to either 180 or 0 in order to orient your tissues correctly (dorsal on the top, rostral on the left)
Mapping Mechanics (Coronal) – videos may be especially helpful for these steps
1. Choose an viewing mode in the top left corner of the mapping window. For mice, STPt avg 2015 works well for initial alignment, and Rainbow 2017 (with "Outline" checked in the Values and Controls settings) works well for refinement. For rats, MRI works well for initial alignment, and Segmentation (with "Outline" checked in the Values and Controls settings) works well for refinement.
2. Adjust the opacity of the atlas using the slider on the left side of the mapping window. Pressing "H" (button on screen or with keyboard) displays only your image, while pressing "V" (button on screen or with keyboard) displays only the atlas
3. To adjust rostrocaudal position, drag the red dot in the Horizontal View (bottom right panel) from left to right
4. To carry out mediolateral linear translation, click and drag the "image" containing the atlas left and right across the main mapping window
5. To carry out mediolateral scaling (stretching), press the spacebar while holding the mouse pointer over the place you want the reference point for scaling. A small cross will appear. Usually it is easier to choose a side and not place the cross in the middle of the section. Then, click on the scaling button (top left, double-headed left-right arrow) and a drouble arrow will appear. Place your mouse pointer at the opposite side of the double cross and press the left button of your mouse. While keeping the left button of the mouse pressed, you can now gently drag the atlas in the direction indicated by the double arrow.
6. To carry out dorsoventral linear translation, click and drag the "image" containing the atlas up and down in the main mapping window.
7. To carry out dorsoventral scaling, press the spacebar while holding the mouse pointer over the place you want the reference point for scaling. A small cross will appear. Usually it is easier to choose a side and not place the cross in the middle of the section. Then, click on the scaling button (top left, double-headed up-down arrow): a double arrow will appear. Place your mouse pointer at the opposite side of the double cross, and press the left button of your mouse. While keeping the left button of the mouse pressed, you can now gently drag the atlas in the direction indicated by the double arrow.
8. To carry out rotation, use the small, curved arrows (upper left) to rotate the atlas in relation to your image.
9. To carry out rostrocaudal angle change (addressing sectioning asymmetry), click and drag the yellow line on the Coronal View (right panel) up and down
10. To carry out dorsoventral angle change (addressing sectioning asymmetry), click and drag the yellow line on the Horizontal View (right panel) left and right
Mapping Mechanics (Sagittal) – videos may be especially helpful for these steps
1. Choose an viewing mode in the top left corner of the mapping window. For mice, STPt avg 2015 works well for initial alignment, and Rainbow 2017 (with "Outline" checked in the Values and Controls settings) works well for refinement. For rats, MRI works well for initial alignment, and Segmentation (with "Outline" checked in the Values and Controls settings) works well for refinement.
2. Adjust the opacity of the atlas using the slider on the left side of the mapping window. Pressing "H" (button on screen or with keyboard) displays only your image, while pressing "V" (button on screen or with keyboard) displays only the atlas
3. To adjust mediolateral position, drag the red dot in the Horizontal View (bottom right panel) from left to right
4. To carry out rostrocaudal linear translation, click and drag the "image" containing the atlas left and right across the main mapping window
5. To carry out rostrocaudal scaling (stretching), press the spacebar while holding the mouse pointer over the place you want the reference point for scaling. A small cross will appear. Usually it is easier to choose a side and not place the cross in the middle of the section. Then, click on the scaling button (top left, double-headed left-right arrow) and a drouble arrow will appear. Place your mouse pointer at the opposite side of the double cross and press the left button of your mouse. While keeping the left button of the mouse pressed, you can now gently drag the atlas in the direction indicated by the double arrow.
6. To carry out dorsoventral linear translation, click and drag the "image" containing the atlas up and down in the main mapping window.
7. To carry out dorsoventral scaling, press the spacebar while holding the mouse pointer over the place you want the reference point for scaling. A small cross will appear. Usually it is easier to choose a side and not place the cross in the middle of the section. Then, click on the scaling button (top left, double-headed up-down arrow): a double arrow will appear. Place your mouse pointer at the opposite side of the double cross, and press the left button of your mouse. While keeping the left button of the mouse pressed, you can now gently drag the atlas in the direction indicated by the double arrow.
8. To carry out rotation, use the small, curved arrows (upper left) to rotate the atlas in relation to your image.
9. To carry out mediolateral angle change (addressing sectioning asymmetry), click and drag the yellow line on the Coronal View (right panel) up and down
10. To carry out dorsoventral angle change (addressing sectioning asymmetry), click and drag the yellow line on the Horizontal View (right panel) left and right
Finishing Up Mapping
1. Check the graph on the right side of the data management window. If your images are ordered and mapped correctly, the graph should be approximately linear and monotonic (always increasing).
2. Click "Save JSON" (data management window, bottom left), name the file, and save it to "D. QuickNII"
Note: If you need to save your progress in QuickNII for any reason, use this step and name the file logically according to your progress
3. Exit out of QuickNII
3. Non-Linear Transformation (VisuAlign)
Preparation
- Paste all images from "C. PNGs with Section Numbers" to your "E. VisuAlign" folder
- Open VisuAlign
- Click "File" (top left) – "Open" and open your saved QuickNII file from "D."
- Drag the opacity slider in the upper left corner all the way to the right so that the atlas appears as an outline
- Change the marker/outline color(s) as desired so that they are more visible
Adding Markers and Making Adjustments
- Hover the mouse over a specific part of an image
- Press the spacebar to add a marker (looks like a plus sign)
- Click and drag the marker in the direction that adjustment needs to be made
Note: Avoid making more than ~50 annotations per image
Deleting Markers
- Hover the mouse over the marker to be deleted
- Press the delete key
Steps for Each Image:
- Add markers to areas of the tissue that are already mapped and looked good (usually toward the middle of the image or at landmarks). Do not make any adjustments to these markers. (Adding markers here just "locks" things into place and keeps you from "over-adjusting).
- Begin adding markers to areas of the tissue that need adjustment. Generally begin with the tissue outline before working your way inward.
- Ensure that key landmarks (hippocampus, ventricles, differentiation of cortex and corpus callosum, cortical layers)
- Use the single arrows (top right) to navigate between images. Other arrow options allow you to move forward to backward 10 images or to navigate to the beginning or end images
Saving Progress:
- After carrying out nonlinear transformation for all images, select "File" – "Save As", name the file, and save it to "E. VisuAlign"
Note: If you need to save your progress in VisuAlign for any reason, use this step and name the file logically.
- Select "File" – "Export". Set the save location to "E. VisuAlign"
- Copy and paste all files within "E. VisuAlign" with names ending in ".flat" to the "Atlas_Maps" folder within "I. Nutil"
Note: If you are trying to open and reload a saved VisuAlign state, make sure that your images (PNGs) are in the same folder that your VisuAlign file is saved to. (Otherwise, nothing will show up when you open the file).
Note:
Sections 1 through 3 should be completed for each brain (coronal sections) or hemisphere (sagittal sections) being analyzed together before moving on to Sections 4 through 6.
4. Pixel Classification (Ilastik)
Preparation
1. Copy and paste all images (in PNG format) from "C. PNGs with Section Numbers" to the folder "A. Images", which is located within "G. Ilastik"
2. Open Ilastik
3. Under "Create New Project", choose "Pixel Classification". Name and save the project file within "G. Ilastik"
4. Select "1. Input Data" (left panel) – "Add New".
5. Upload several training images that are representative of the dataset you are analyzing (Usually at least 1 image per brain/condition)
6. For the first training image, double-click the box in the column "Location" for each image
7. In the popup window, next to the heading "Storage", choose the option to "Copy into project file"
8. Click "Ok"
9. Repeat for all training images
10. Select "2. Feature selection" – "Select Features"
11. Click and drag over all white boxes. They should all turn green with check marks.
12. Select "3. Training"
13. Click "Add Label" twice (should be 4 labels now: Label 1, Label 2, Label 3 Label4)
14. Double-click on each label and change both the "Color for drawing" and "Color for probability display" to the following for each of the respective labels:
- Label 1 (high-intensity PNNs): bright green
- Label 2 (moderate-intensity PNNs) : dark green
- Label 3 (diffuse PNN signal): bright blue
- Label 4 (background): dark blue
Determining which PNNS are high-intensity
Open your training images in ImageJ
- Select "Image" – "Adjust" – "Brightness/Contrast"
- Turn contrast all the way up
- Only label PNNs that are clearly-defined in this view as high-intensity. All others are low intensity
Annotating Training Images
General strategy: make a couple annotations on one image before switching to another and making a couple annotations there. (This helps ensure that all training images are used somewhat equally to train the model).
Label definitions:
- Label 1: high intensity, well-defined PNNs (brightest green, visible on high-contrast ImageJ image)
- Label 2: low intensity, moderately-defined PNNs (moderate green)
- Label 3: diffuse PNN staining (dim green signal)
- Label 4: background (dark green/black)
(Only Labels 1 and 2 are visible below)
To annotate: select a label in the left panel and draw on the image, marking a few of that label's pixels
To erase an annotation: select eraser (below "Add Label" button on left panel); draw to erase
To navigate the image: hold "Ctrl" key and hover your mouse over the area you want to zoom in on; scroll to zoom in and out
Viewing segmentation results:
- Select "Live Update" (left panel)
- Toggle "Prediction" and "Segmentation" eyeball icons for each label as needed to check segmentation
Note: "Segmentation" views are generally most useful
3. Toggle "Uncertainty" eyeball icon to see areas where the model is unsure of segmentation. Focus on these areas (highlighted in blue) to add annotations and refine segmentation
Saving segmentation: When satisfied with your segmentation (and at any point), save your Ilastik project by clicking "Project" (top left) – "Save Project"
Batch Processing
- Select "4. Prediction Export"
- Set "Source:" to "Probabilities"
- Click "Choose Export Image Settings"
- In the window the pops up, ensure that within the "Output File Info" section, the format is "hdf5"
- Click "Ok"
- Select "5. Batch Processing"
- Click "Select Raw Data Files"
- Navigate to and select all images in your "A. Images" folder (within "G. Ilastik")
- Click "Open"
- Click "Process All Files". This can take several minutes
- Once processing is finished, cut and paste all simple segmentation files (labeled with "Probabilities.h5", solid white icons) from your "A. Images" folder to your "B. Probabilities (Pixel Classification)" folder
5. Object Classification (Ilastik + Fiji)
Preparation
- Open Ilastik
- Under "Create New Project", choose "Object Classification [Inputs: Raw Data, Pixel Prediction Map]"
- Name and save the project file within "G. Ilastik"
- Select "1. Input Data" (left panel) – "Add New".
- Upload the same training images you used for Pixel Classification
- For the first training image, double-click the box in the column "Location" for each image
- In the popup window, next to the heading "Storage", choose the option to "Copy into project file"
- Click "Ok"
- Repeat for all training images
- Click "Prediction Maps"
- Select "1. Input Data"(left panel) – "Add" – "Add Separate Image(s)"
- Select all of the probability files corresponding to your training images (from "B. Probabilities (Pixel Classification"
- Change the storage locations to "Copy into project file" for each probability file you upload
- Select "2. Threshold and Size Filter" Note: The thresholds and parameters mentioned in the following steps can be changed based on your project and focus, but make sure to stay consistent
- Method – "Simple"
- Input – "0"
- Smooth – both boxes "0.6"
- Threshold – "0.5"
- Size filter – Min: "0" Max "400" – This is the main number you may want to manipulate. In later steps, you can always lower the cutoff for the maximum if desired, but once lowered here, you cannot lower it later, so err on the side of a higher cutoff if unsure.
- Click "Apply" and make sure everything looks reasonable
- Select "3. Object Feature Selection"
- Click "Select Features:
- Click the arrow next to "Standard Object Features"
- Click the checkbox next to "Standard Object Features:
- Deselect the following three checkboxes under the heading "Location": "Bounding Box Maximum", "Bounding Box Minimum", and "Center of the Object"
- Click "Ok" (bottom right of object feature selection window)
Note: it is not necessary to do anything in the "4. Object Classification" window, but you can experiment with turning on and off the various prediction and segmentation options if you would like. Using the segmentation views, each individual PNN in the "Label 1" category should appear as differently colored blobs
Batch Processing
- Select "5. Object Information Export"
- Source – "Object Identities"
- Click "Choose Export Image Settings"
- Under "Output File Info", set Format – "tiff"
- Click "Ok"
- Select "6. Batch Processing"
- Click "Select Raw Data Files"
- Navigate to and select all your images in your "A. Images" folder (within "G. Ilastik")
- Click "Open"
- Click "Prediction Maps"
- Navigate to and select the probability files corresponding to all the images that you just selected (from the appropriate "B. Probabilities (Pixel Classification)" folder)
- Click "Open"
- Click "Process All Files". This should just take a couple minutes
- Once processing is finished, cut and paste all object identity files (labeled with "Object Identities" in the file name) from your "A. Images" folder to your "C. Object Identities (Object Classification)" folder (within "G. Ilastik)
Converting Object Identity Files for Quantification
- Open ImageJ/Fiji
- Click "Input"
- Change the input folder to the appropriate "C. Object Identities (Object Classification)" folder. Using the "Copy File Path" shortcut makes this quite easy.
Note: for this step and the following steps (including those using Nutil), be really sure that you have the correct folder for the correct brain selected – it is extremely easy to get confused when you are working on multiple brains at once
4. Click "Output"
5. Change the output folder to your "D. Object Identities Post-Color Change" folder. Using the "Copy File Path" shortcut makes this quite easy.
Change "Output format:" to PNG
6. Click "Open" (bottom left corner)
7. Navigate to "G. Ilastik"
8. Select "object color change macro"
9. Click "Open"
10. Change the file path included in the macro to correspond with where the "ALL GREEN LUT.lut" file is located on your specific computer. Note: make sure to replace back slashes with forward slashes when you do this.
11. In the Batch Process window, click "Process". Images (with some color now) should begin to show up in your "D. Object Identities Post Color-Change" folder. This usually takes around a minute
12. Copy and paste all newly created files in "D. Object Identities Post-Color Change" to the "Segmentation" folder within "I. Nutil"
6. Quantification (Nutil + Scripts)
Nutil
- Open Nutil
- Select "Operation" – "New" – "Quantifier"
- Name the project according to the cohort, animal, and hemisphere (sagittal) you are working on
- Analysis type – "Quint"
- Select the folder with your segmentations by selecting "Browse" then navigating to the "Segmentation" folder within your "I. Nutil" folder
- Select the folder with your atlas maps by selecting "Browse" then navigating to the "Atlas_Maps" folder within your "I. Nutil" folder
- Set the "Reference Atlas" to "Allen Mouse Brain 2017" for mice or "WHS Atlas Rat v4" for rats
- Navigate to and select your final VisuAlign saved state (JSON file) within your "E. VisuAlign" folder
- Select "Browse" then navigate to and select the "Output" folder within your "I. Nutil" folder
- Output Reports – "All"
- Open one of the images within your "Segmentation" folder in Fiji
- In Nutil click the colored box next to "Object colour"
- Select "Pick screen color"
- Zoom into the image you just opened and hover the mouse over a part of the segmentation depicting Label 1 (for high intensity PNNs, bright green if you used the default color change macro)
- Make sure that the color you are hovering over shows up in the "Select Color" window then click "Ok" Note: if you used the default color-change macro, the HTML code for the color of Label 1 should be "#00ff00". You can also just type this into the HTML box instead of doing steps 13 and 14
- Object splitting – "No"
- Click "Show advanced settings"
- Minimum object size – "1" Note: this can be changed to filter out small, non-PNN objects. Make sure to experiment with this before making any changes, and ensure that you stay consistent from brain to brain.
- Pixel scale – "1"
- Pixel scale unit – "pixels"
- Custom masks – "No"
- Output report type – "HTML"
- Customized reports – "Custom"
- Select "Browse" then navigate to and select the "updated selected-brain-regions.txt" file. This dictates the regions you will be analyzing. Note: This can be changed if you want to look at different regions. Refer to the original Quint documentation for details.
- Coordinate extraction – "All"
- Point cloud density – "1"
- Display object IDs in image file and reports – "No"
- Display region IDs in image files – "Yes"
- Unique ID format – "_sXXX"
- RegEx for user ID – leave blank
31. Click "Save As" (top left)
32. Save the Nutil analysis file within your "I. Nutil" folder, named according to the cohort, animal, and hemisphere (sagittal) you are working on
33. Click "Start" (bottom left). This should take a couple minutes. All results will appear in your "Output" folder (within "I. Nutil"
34. Run Nutil again for each hemisphere/condition you are analyzing, making sure to change relevant settings when switching between hemispheres/conditions. Note: ensure that you change all appropriate input/output folders. It is extremely easy to mix things up here
7. 3D Visualization (MeshView, optional)
- Click "Choose Files"
- Open your "I. Nutil" folder
- Navigate to and select "PROJECTNAME_3D_combined.json" within "Output" – "Coordinates" Note: there is no way to save a MeshView visualization. You will have to reload this file every time you want to visualize it.
- Change the opacity of the displayed atlas brain meshes using the sliders in the left panel. To obtain a translucent whole brain mesh that is good for visualization, drag the "All" slider all the way to the left and the "Root" slider about 20% of the way to the right
- Toggle the brain regions of PNNs displayed using the sliders in the right panel
Protocol references
Yates SC, Groeneboom NE, Coello C, Lichtenthaler SF, Kuhn PH, Demuth HU,Hartlage-Rübsamen M, Roßner S, Leergaard T, Kreshuk A, Puchades MA, Bjaalie JG. QUINT: Workflow for quantification and spatial analysis of features in histological images from rodent brain. Front Neuroinform. 2019 Dec 3;13:75. https://doi.org/10.3389/fninf.2019.00075.
Puchades MA, Csucs G, Lederberger D, Leergaard TB and Bjaalie JG. Spatial registration of serial microscopic brain images to three-dimensional reference atlases with the QuickNII tool. PLosONE, 2019, 14(5): e0216796. https://doi.org/10.1371/journal.pone.0216796
Berg S., Kutra D., Kroeger T., Straehle C.N., Kausler B.X., Haubold C., et al. (2019) ilastik:interactive machine learning for (bio) image analysis. Nat Methods. 16, 1226–1232. https://doi.org/10.1038/s41592-019-0582-9
Groeneboom NE, Yates SC, Puchades MA and Bjaalie JG. Nutil: A Pre- and Post-processing Toolbox for Histological Rodent Brain Section Images. Front. Neuroinform. 2020,14:37. https://doi.org/10.3389/fninf.2020.00037
Puchades MA, Csucs G, Lederberger D, Leergaard TB and Bjaalie JG. Spatial registration of serial microscopic brain images to three-dimensional reference atlases with the QuickNII tool. PLosONE, 2019, 14(5): e0216796. https://doi.org/10.1371/journal.pone.0216796
Gurdon B, Yates SC, Csusc G, Groeneboom N, Hadad N, Telpoukhovskaia M, Ouellette A, Ouellette T, O’Connell K, Singh S, Murdy T, Merchant E, Bjerke I, Kleven H, Schlegel U, Leergaard T, Puchades M, Bjaalie J, Kaczorowski C (2023). Detecting the effect of genetic diversity on brain composition in an Alzheimer’s disease mouse model. Commun Biol. 2024 May 20;7(1):605. https://doi.org/10.1038/s42003-024-06242-1.
Puchades MA, Csucs G, Ledergerber D, Leergaard TB, Bjaalie JG (2019). Spatial registration of serial microscopic brain images to three-dimensional reference atlases with the QuickNII tool. PLOS ONE 14(5): e0216796. https://doi.org/10.1371/journal.pone.0216796
Wang et al. (2020). The Allen Mouse Brain Common Coordinate Framework: A 3D Reference Atlas. Cell. https://doi.org/10.1016/j.cell.2020.04.007
Kleven et al (2023). Waxholm Space atlas of the rat brain: a 3D atlas supporting data analysis and integration. Nat Methods. DOI: 10.1038/s41592-023-02034-3
Acknowledgements
Special thanks to Devin Naredo, Madison Gossard, and Johnson Zhang for critical review of this protocol and to Payal Patel, Alexandra McBryar, Kiran Hussaini, Reagan Moore, Trevor Eisenbacher, and Jessica Belmont for their contributions to mapping efforts with Quint.