Standards, collaborative research for reproducibility and FAIRness in life sciences Event as iCalendar

(Science Event Tags, Computer Science, Seminars)

11 January 2019

12 - 1pm

Venue: Room 303.561

Location: City Campus

Host: Department of Computer Science

MassimilianoIzzo

Speaker: Massimiliano Izzo

Abstract 

In recent years, the increase in the volume and heterogeneity of data produced in life science and clinical research has posed growing problems of data management and of proper metadata annotation in scientific studies and experiments. For this reason, since the advent of high throughput technologies, researchers have developed a huge number of standards — data formats, controlled vocabularies, reporting guidelines — to improve documentation,

understanding, and ultimately reproducibility of experiments. However, as evidenced in a social science study by Paul N. Edwards et al. (Social Studies of Science, 2011), well-refined standards can act as an element of friction in the scientific process of research collaborations, especially in multi-disciplinary projects on a global scale, where groups with different subject expertise (e.g. biology/medicine, computer science, physics...) and nationalities are involved.

In this talk, I will describe a different set of tools I have developed and worked on during my academic career in Genoa and Oxford dealing with these issues: XTENS (https://github.com/xtens-suite/xtens-app), ISA (https://isa-tools.org/) and FAIRsharing (https://fairsharing.org/). To some extent, they all aim to provide a framework to describe studies and experiments in an extensive yet flexible way, giving the users control over the choice of standards and the extent of their adoption, so reducing the friction they generate and the resistance to their adoption. Furthermore, I will show how these tools can help improving FAIRness (Findability, Accessibility, Interoperability and Reusability) of data, a concept that has gained increasing traction in the data science community as a mean to improve reproducibility in research.

About our speaker

Massimiliano is a Research Software Engineer with an educational background in Biomedical Engineering. His main interests involve design and development of innovative data models for life sciences, structured/unstructured data management and full-stack software development. Massi works at the Oxford e-Research Centre, University of Oxford, in Susanna-Assunta Sansone's team. Before joining the Centre he was a Research Collaborator at the Giannini Gaslini Institute, in Genoa (Italy) where he designed distributed data management systems for integrated biobanking management. In his spare time he enjoys reading (mostly speculative fiction novels), rowing, and wandering aimlessly in bookshops and cafes.