Geropathology Imaging - Part of the The Jackson Laboratory Shock Center of Excellence in the Basic Biology of Aging
Method Overview
Sample Preparation
Any good experiment should start with planning your experiment. Computational geropathology is no exception to this and involves thinking about lots of details such as organizing slides and how sections will be arranged on the slides, fixation, stains, as well as batch effects of processing at east stage. See slide preparation tutorial for more details. Once you have your slides in hand, you need to Image them.
Slide prep Protocols
Link to Protocols
Image Acquisition and Storage protocols
Link to Our Protocols and Link to OMERO
Image Capture
There are many ways to Image Slides and our platform can utilize take many inputs, but we do strongly recommend using a slide scanner and Image storage and organization in OMERO. This is the fastest way to image entire tissues and provides contextual information. OMERO is free, searchable but also can be private, can handle large files, can capture metadata with Images, and works with other tools such as ImageJ, Qupath and allows for scripting directly.
After you have the images in hand, you will need to convert them to a zarr file format. More information on what a zarr file is and why we use them and how they fit into the larger picture of what we are trying to do can be found here. (Link)
Additionally, If you are concerned with a particular structure, segmentation can be used to isolate aging in the context of that feature. See our publication on glomerular segmentation.
Classifying Images by Age
After we have zarr files made, we use them in our classifier. A classifier is the result of our machine learning pipeline and conceptually is a mini computer program that takes new images applies the program to them and provides an output that in this case provides and age score per pixel for the input image. Our Classifiers, practical tutorials on how to use them and how to build your own can be found here.
Classifiers are the result of our machine learning pipeline, Following this link we provide a more indepth description of them, we provide our trained classifiers to use, and provide directions to use them on your own data as well as directions to make your own.
After we have run our data through the classifer, we can do statistics and visualize age in a spatial content of the tissue. Here we show young and old kidney samples where we can quantify the difference in age and look to see where the tissues appear older.
The goal of our workflow is to quantify age and to visualize age. Our results allow you to do statistics on the age of a tissue and also to visualize if a tissue is aging in a spatially organized manner.
Aging Classifiers
Link to Classifiers and how to use them
Glomerular Segmentation
Link to Glomerular Segmentation
Identifying Glomeruli
Quantifying Messangial Matrix Expansion
MME Quantification
Your content goes here. Edit or remove this text inline or in the module Content settings. You can also style every aspect of this content in the module Design settings and even apply custom CSS to this text in the module Advanced settings.