Titulo Estágio
Developing automated screening tools to detect common problems or beneficial practices in research publications
Local do Estágio
UC Polo 1 and DEI/FCTUC, CISUC
Enquadramento
Rationale: Research publications are often missing essential details needed to critically evaluate the quality of a study, reproduce results, or reuse methods and data. Automated screening tools can rapidly identify common problems or beneficial practices in preprints, draft manuscripts and research papers, and provide authors with individualized feedback on how to improve reporting. This can make papers more transparent and reproducible.
The ScreenIT pipeline includes a combination of tools that detect features like open data and code, limitations sections, blinding, randomization and ethics statements, and common data visualization problems. During the pandemic, ScreenIT was used to screen almost 24,000 COVID-19 preprints on bioRxiv and medRxiv, and post public reports using a web annotation tool (preliminary report here: https://www.nature.com/articles/s41591-020-01203-7).
Tools can be used to provide authors with feedback on draft papers, monitor reporting for research institutions or journals, or assess the impact of other interventions (e.g., journal policy changes requiring open data).
Objetivo
Goal: To create an automated screening tool to detect a common problem or beneficial practice in research publications. Examples might include tools for detecting open and reproducible research practices (e.g., protocol sharing, use of ORCIDs or author contributions statements), adherence to reporting guidelines (e.g., the PRISMA guidelines for systematic reviews and meta-analyses, the SAGER guidelines for reporting of sex and gender in biomedical research), or common data visualization problems.
Plano de Trabalhos - Semestre 1
Project: The student will use the ScreenIT pipeline to screen biomedical papers from researchers at the University of Coimbra; then adapt code for a dashboard from another institution to create a dashboard for biomedical research at the University of Coimbra.
Plano de Trabalhos - Semestre 2
Project: The student will use the ScreenIT pipeline to screen biomedical papers from researchers at the University of Coimbra; then adapt code for a dashboard from another institution to create a dashboard for biomedical research at the University of Coimbra.
Condições
Project: We are seeking students to help us expand our tool pipeline, by developing new tools that will allow ScreenIT to provide better feedback on different types of research studies. New tools will be integrated into the ScreenIT pipeline and used for ongoing projects and intervention studies.
Required skills: Students need to have intermediate level programming skills in R and/or Python.
Observações
Required skills: Students need to have intermediate level programming skills in R and/or Python.
Orientador
Supervision: Tracey Weissgerber, Teresa Cunha-Oliveira, Joel P. Arrais (DEI)
teresa.oliveira@cnc.uc.pt 📩