Towards Machine-Readable (Meta) Data and the FAIR Value for Artificial Intelligence Exploration of COVID-19 and Cancer Research Data

Abstract

Even before COVID-19, the bioinformatics labs and life science industry were investing extensively in ecosystems of technological and analytical applications/appliances to store, curate, share, integrate, and analyze large amounts of data. With the pandemic coming at an accelerating pace, a series of global research actions are being implemented to strive against the virus and its effects and to create data-driven investigations to support more agile responses to future events 1 . Innovative solutions in COVID-19 research require more efficient and effective data management strategies and practices. Cancer research is an excellent example of the adoption of the FAIR (Findable, Accessible, Interoperable, and Reusable) data principles (Wilkinson et al., 2016) on precision oncology (Deist et al., 2020;Delgado and Llorente, 2020;Vesteghem et al., 2020) and major cancer data repositories, such as the NIH Cancer Research Data Commons, are gradually adhering to these principles.In a broad sense, understanding the whole scenario of a worldwide virus outbreak such as the COVID-19 pandemic has been a great challenge. It involves a huge effort to put together facts and a large volume of data related to the dynamics of the real world, as well as past and current results of scientific research. At the same time, cancer researchers are pivoting portions of their investigations and contributing with their expertise and resources to scientific studies of the SARS-Cov-2. Their findings have been wide in scope, ranging from