Supplementary files for Machine learning for histological annotation and quantification of cortical layers
Authors/Creators
Description
Description
This dataset contains 7 QuPath projects. The raw data images linked to these projects and located in other Zenodo datasets need to be downloaded as well.
The raw data contains images of 14 hemispheres from height animals.
- Nissl_1 :
- animal 1413827 Right Hemisphere
- Nissl_2 :
- animal 1413829 Right Hemisphere
- animal 1413828 Right Hemisphere
- animal 1413827 Left Hemisphere
- Nissl_3 :
- animal 1413828 Left Hemisphere
- Nissl_4 :
- animal 1443459 Right Hemisphere
- animal 1443460 Right Hemisphere
- Nissl_5 :
- animal 1443459 Left Hemisphere
- animal 1443460 Left Hemisphere
- Nissl_6 :
- animal 1449920 Left Hemisphere
- animal 1449921 Left Hemisphere
- animal 1449921 Right Hemisphere
- animal 1449922 Left Hemisphere
- animal 1449922 Right Hemisphere
- QuPath_LayerBoundaries_GroundTruth_20220927:
- This is the QuPath project that contains S1HL layers annotations done by the experts and which have been used to trained the Random forest Machine Learning method for the S1HL brain classification. It contains some images from all the eight animals.
Animals
All animal procedures were approved by the Veterinary Authorities and the Cantonal Commission for Animal Experimentation of the Canton of Vaud, according to the Swiss animal protection laws, under license number VD3516.
Outbred Wistar Han rats (Janvier Laboratories, France) were ordered with their litter aged eight postnatal days (P8). Dams were housed individually and allowed to raise their own litters until experimentation on male offspring aged fourteen days (P14; N=8 animals; N=3 litters). Animals were housed in standard plastic laboratory cages, with bedding, nesting material and paper tube and ad libitum access to food (SAFE 150 SP-25) and water, cleaned once per week, and kept on a twelve-hour light-dark schedule with lights turned on at 06:30 AM, in rooms under controlled humidity and temperature. The sample size here is greater than those reported in other open source atlases (“Allen Reference Atlas - Mouse,” n.d.; “The Rat Brain in Stereotaxic Coordinates - 7th Edition,” n.d.).
Sample preparation
On postnatal day fourteen, rats were transferred to the experimental room in the morning to acclimate. The described procedure was conducted within a consistent 3-hour window of the day (09:00-12:00). Initially, the rats were deeply anesthetized using pentobarbital (intraperitoneal dose of 150 mg/kg; concentration of 150 mg/ml). This was succeeded by transcardial perfusion with ice cold 0.1 M phosphate buffer (PB; pH 7.4), followed by cold 4% paraformaldehyde (PFA) in 0.1 M PB. Subsequently, the brain was carefully removed from the skull, postfixed at 4°C in 4% PFA overnight, and then rinsed in 0.1 M PB. The brains underwent a sequential storage process: first in a 15% sucrose solution (in 0.1 M PB) at 4°C for approximately 24 hours, followed by a 30% sucrose solution at 4°C for an additional 24 hours. The hemispheres were carefully divided along the midline, after which both right and left hemispheres were precisely sliced sagittally using a cryostat (Leica, VT-1200S) at 50 µm employing an approximate angle rotation of 4 ± 1 degrees along the anterior-posterior axis to optimize alignment with apical dendrites. These brain slices were stored in a cryoprotectant solution (30% v/v ethylene glycol; 30% m/v sucrose in 0.1 M PB) at -20°C, preserving them until immunohistochemistry assays were executed (within a maximum of two weeks from extraction to immunohistochemistry).
In order to determine the cell densities in P14 rat, brain slices were immunostained using cresyl violet, a stain specifically targeting cell bodies, including the endoplasmic reticulum, also known as Nissl substance or Nissl bodies. Free-floating sections of 50 µm thickness were transferred from cryoprotectant into 0.1 M PB to thaw and eliminate any cryoprotectant remnants. Subsequently, they were transferred into 0.01 M PB to minimize salt residues before being meticulously mounted onto SuperFrost© glass slides (Thermo Fisher Scientific Inc., Gerhard Menzel B.V. & Co. KG, GE). This mounting was carried out while considering the brain’s orientation relative to the midline, from its external to internal regions. Slide-mounted sections were processed using an automated slide stainer Tissue-Tek® Prisma Plus (Sakura Finetek-Europe, NL). These sections were incubated for 6 minutes at room temperature (RT = 20°C) in a 0.5% cresyl violet solution in water (with pH adjusted to 2.85 using acetic acid), followed by a brief wash in tap water. The sections underwent dehydration through a series of ethanol concentrations (70%, 70%, 96%, 100%, 100%) with each step lasting one minute at RT. Subsequently cleared with two steps of xylene for one minute each at RT, and the sections were mounted using Pertex (Sakura Finetek-Europe, NL) before being cover-slipped using the automated glass coverslipper Tissue-Tek® Glas™ g2 (Sakura Finetek-Europe, NL). A meticulous assessment of the coloration was conducted and if the staining appeared faint, a repeat staining procedure was carried out.
Immunostained slides were scanned using an automated slide scanner (Olympus, VS120-L100, GER) equipped with a UPLSAPO 20x/0.75 air objective (Olympus, GER) and a Pike F505 Color camera leading to a pixel size of 0.346 μm/pixel. Each brain slice was entirely scanned. Subsequently, the digital images obtained were meticulously organized and subjected to analysis using the open-source software QuPath v0.3.2 (Bankhead et al., 2017).
Intructions
The projects contained in this dataset have been created with QuPath v0.3.2 but could be opened with new QuPath version.
- Download the dataset
- untar the tar balls included in this dataset
- install QuPath
- Open QuPath
- Open a project within QuPath (Files->Project...->Open Project...)
Files
README.md
Files
(4.0 GB)
| Name | Size | Download all |
|---|---|---|
|
md5:17db7c41bbbb50f6284c62dad92c3c24
|
222.5 MB | Download |
|
md5:9f560a67658eae0a47b8aa01ad870fbe
|
794.6 MB | Download |
|
md5:4238a8a1dd7ad735b918524446513abf
|
261.1 MB | Download |
|
md5:17fb7f4f3edff0753324b469d533d8f1
|
516.5 MB | Download |
|
md5:eed543a6529491fa86887ff2940cb893
|
407.3 MB | Download |
|
md5:8cb8334b0f4253ae616649044c24b684
|
1.3 GB | Download |
|
md5:d38c0f6d12b92cf2e07ed9c2c977c8d5
|
508.5 MB | Download |
|
md5:5a608e232affaded7398adac7438cfb3
|
6.8 kB | Preview Download |
Additional details
Funding
- École Polytechnique Fédérale de Lausanne
Dates
- Created
-
2022-01-01
Software
- Repository URL
- https://github.com/BlueBrain/layer-recognition
- Programming language
- Python