New CANARIE program encourages adaptation of research tools across disciplines

CANARIE, Canada’s high-performance computing network, is using $2 million in funding to encourage greater sharing of visualization software and other research platforms in an effort to reduce duplication and accelerate discoveries. Under the pilot program, five teams will receive funding to improve software tools they have already developed and adapt them for use by researchers in other disciplines. CANARIE says the pilot Research Software Program aims to demonstrate the viability of a new approach for making more efficient use of research funds.

In the past, CANARIE would fund individual teams to develop their own variations of the same software. Under this new model, support is given to build and adapt platforms that can be shared with other research teams, regardless of their field of scientific study. Research platforms, also known as virtual research environments or science gateways, are designed to support the entire research flow of a research project, and include capabilities such as data collection, processing, visualization and storage.

“We noticed that regardless of the discipline, a lot of these software packages we funded – we call them research platforms – look the same, which makes sense. Rather than fund each of the research teams to basically duplicate the same software over and over, we adjusted the program to include a service model,” says Scott Henwood, CANARIE’s director of research software. “Instead of developing everything from scratch, (researchers can) use what’s been contributed by others.

Researchers can access these new platforms for free at science.canarie.ca.

“In the future, hopefully, we’ll be able to get more researchers the software they need by getting them to adapt existing platforms,” Henwood adds.

Henwood says the pilot is intended to test how the research community responds to the reuse paradigm. While CANARIE will continue to fund software platforms for single use, he expects there will be fewer developed as more focus is put on platforms that can be shared across disciplines.

The five projects selected include mostly open-source platforms developed by university-based researchers:

- The 3D Slicer open-source platform supports image analysis and visualization, primarily in biomedicine. Developed by a Queen’s Univ research team led by Dr. Gabor Fichtinger, 3D Slicer has been funded to support image-guided therapy (IGT) through surgery, radiation, needle-based medical interventions, and more. The evolution of 3D Slicer is expected to speed up image translation from the laboratory to the operating room and commercial products.

- The Canadian Writing Research Collaboratory (CWRC) Extension at the Univ of Guelph will allow other researchers in epigraphy, history, ethnography, linguistics, legal history, and library and information science to access the project’s repository and software. Led by Dr. Susan Brown, the project combines computing hardware, software, and personnel to maintain a repository of over 250,000 digital texts, bibliographic records, and multimedia objects, together with a software toolkit that enables digital humanities research.

- Univ of British Columbia’s Montage software analyzes the genome sequences of thousands of individual cancer cells within a tumour. Led by Dr. Sohrab Shah, the the team will evolve the research platform into a unified cloud-based web service for use by other research groups studying aggressive breast cancer tumours.

- iEnvironment is software used by environmental science and engineering researchers who work with rich environmental data to deal with complex questions about surface water. The team, led by Dr. Donald D. Cowan of Univ of Waterloo, will adapt the software for users in geomorphology, hydraulic engineering, biology and environmental science who deal with multiple sources of data. The software can be used in other research projects in climate change, natural storm management, biodiversity, water pollution prevention, and water level management.

- Univ of Waterloo will adapt its OpenPNM (Pore Network Modeling) open-source platform for use in research on sustainable energy and greening the power grid. The platform is currently being used by a research team led by Dr. Jeff Gostick “to model how fluids move through a wide range of porous materials, including plants and bone tissues, paper, polymers, ice formations, and oil reservoirs.”

Each project has one year to adapt their platforms for new users, but will have access to additional maintenance funding to September 2020 to support the requirements of new users.

R$