3D cancer tissues stored in a virtual biobank
Researchers from the University of Newcastle and the Hunter Medical Research Institute (HMRI) have developed what is claimed to be the world’s first virtual platform to host 3D copies of human cancer tissues, changing the way scientists access critical information to advance cancer treatment.
With cancer a leading cause of death in Australia, Dr Jamie Flynn, Dr Antony Martin and Dr William Palmer developed The Virtual Biobank as an open resource to help speed up and enhance medical research activities. The biobank will digitise tissue samples donated by patients, which up until now could only be requested by researchers through physical biobanks as part of a months-long process.
“We’ve taken a tiny sample from tumour biopsies stored at the Hunter Cancer Biobank in the HMRI building and converted them into a virtual copy, enabling anyone around the world with an internet connection to carry out research from their computers or easily request access to the physical sample they need,” said Dr Flynn.
“Each digital cancer sample in The Virtual Biobank is made up of high-resolution microscopy images in both 2D and 3D, plus important clinical and molecular information that provides the foundation for virtual research into cancer.
“This process ensures the physical sample remains intact, but a 3D digital copy with clinical and experimental information is kept online for future use. This is particularly critical for rare cancers, which are hard to study due to a limited number of samples.”
The addition of 3D images to the biobank was made possible thanks to HMRI’s 3D Tissue Clearing and Lightsheet Microscope Facility, which specialises in making tissue samples ‘see through’ for cutting-edge 3D microscopy. The team has also embedded specialised software tools into the online platform that would otherwise be too expensive for researchers to access, with the help of colleagues across the University of Newcastle’s IT department and Library Research Services.
“We really wanted to make The Virtual Biobank interactive and dynamic, so with the help of lead developer Dr Bill Pascoe and the Academic Research Computing Support team (ARCS), we developed a software toolkit for other people to contribute additional information such as findings from their own investigations,” Dr Martin said.
The ‘Tissue Tracker’ component of the site allows users to automatically log the experimental procedures for all new tissue samples coming into a lab, removing the need for handwritten notes and allowing other researchers to easily replicate experiments.
“Users can also create full 3D images for free by simply uploading their own microscopy data to our ‘Terastitcher Pipeline’, eliminating the need for expensive commercial software that can ordinarily cost upwards of $50,000,” Dr Martin said.
Working with the Hunter Cancer Biobank, Dr Flynn and his colleagues have uploaded the first 20 breast cancer samples to The Virtual Biobank with accompanying clinical descriptions. In future, The Virtual Biobank aims to work with biobanks around the country to continue uploading a variety of cancer tissues, with potential for use in areas of research, education and even virtual reality.
“The technology and tools that we’ve put together make this data available to anyone in the world,” said Dr Palmer. “People in the fields of engineering or computer science can now easily apply their knowledge to questions surrounding cancer.
“Educators can also access the site and use real-world examples to teach pathology to the next generation of cancer clinicians and researchers.
“We’d also like to convert the 3D data into virtual reality for education and general awareness. Hopefully soon, anyone with a smartphone and Google Cardboard could experience the internal environment of cancer tissue and bring about new insights.”
The Virtual Biobank can be accessed at https://virtualbiobank.newcastle.edu.au/.
Known as Pattern to Knowledge (P2K), the technology uses artificial intelligence to leverage deep...
Stanford scientists have developed an algorithm, called Phrank, that automates the most...
By analysing the brain activity of tetraplegia patients, a new deep-learning algorithm can help...