Cancer is an advanced class of illnesses that are not only inadequately understood but also make up the second-top cause of death across the globe.
Cancer development comprises changes in many biochemical and chemical molecules as well as pathways and reactions.
Recently, cancer researchers and computer scientists based at the University of Cambridge created an artificial intelligence-powered literature-based system that helps scientists involved in search of cancer-based discoveries.
The system, LION LBD or Literature-Based Discovery, is built to save researchers a considerable amount of time, which they would have spent manually sifting through vast piles of published research.
LION LBD marks the first-ever literature-related discovery system for backing cancer research.
The article containing all the details regarding the system’s results can be found in the Bioinformatics journal.
Massive volumes of scientific literature have made it difficult for researchers to identify relevant information to advance their studies.
Professor Anna Korhonen, the co-director of Cambridge Language Technology Lab and among the leaders of the LION LBD development, said: “As a cancer researcher, even if you knew what you were looking for, there are literally thousands of papers appearing every day.
LION LBD uses AI to help scientists keep up-to-date with published discoveries in their field, but could also help them make new discoveries by combining what is already known in the literature by making connections between sources that may appear to be unrelated.”
The LBD initials found in LION LBD represent Literature-Based Discovery, which is a concept created in the 1980s with the aim of making new discoveries through integrating pieces of information drawn from detached sources.
The main idea behind the initial version of LBD is that all the concepts, which are never explicitly connected in the literature, might be indirectly connected through intermediate concepts.
LION LBD’s design enables real-time search to identify indirect links between entities contained in a database with millions of publications while preserving users’ ability to explore every mention in its initial context.
“For example, you may know that a cancer drug affects the behavior of a certain pathway, but with LION LBD, you may find that a drug developed for a totally different disease affects the same pathway,” Korhonen said.
As the first-of-its-kind system created to meet the cancer research needs, LION LBD boasts a unique focus, specifically on the molecular biology of cancer as well as using natural language processing (NLP) and machine learning techniques in detecting various references to cancer traits available in the text.
LION LBD has shown a unique potential to spot undiscovered connections and rank essential concepts highly amongst possible links.
The system is also created using open standards, open-source, and open data, and it is accessible as a programmable API or interactive web-based interface.
Currently, researchers are striving to extend LION LBD’s scope to include more relations and concepts.
They are also collaborating with cancer researchers in a bid to assist and boost the technology, specifically for end users.
LION LBD was created in conjunction with Cancer Research UK Cambridge Institute, Karolinska Institutet in Sweden and University of Cambridge Language Technology Lab, and was financed by the Medical Research Council.