Publicação
Design and implementation of an R-Based Infrastructure for an improved flow of data across the VICT3R Project
| dc.contributor.advisor | Vaas, Lea | |
| dc.contributor.advisor | Faria, Brígida Mónica | |
| dc.contributor.author | Ribeiro, Mariana Louro | |
| dc.date.accessioned | 2026-02-24T09:36:16Z | |
| dc.date.available | 2026-02-24T09:36:16Z | |
| dc.date.issued | 2025-09-29 | |
| dc.description.abstract | Toxicology studies in pharmaceutical drug discovery are a critical component of evaluating the safety profile of a new drug candidate. These studies aim to identify potential adverse effects of a compound on biological systems, ensuring that the drug is safe for clinical trials. When conducting animal experiments, ethical aspects are of special importance. The concept of Virtual Control Groups (VCGs) refers to the use of historical data as a substitute or complement to the conventional control groups, reducing the number of animals and saving time, costs and other resources. The VICT3R project is a public-private partnership funded by the European Innovative Health Initiative that provides methodology and infrastructure to enable the reduction of the number of animals used in safety testing by creating VCGs by creating a database containing historical control data from toxicology studies. This dissertation aims to support the VICT3R project by developing a Pipeline and an R package that facilitate the quality checking, validation and retrieval of standardized toxicology data. Once collected from the donor companies, the quality of the data can be assessed by the SEND Data Quality Control Pipeline wrapped into a graphical user interface of a shiny app, and then cleaned and stored in the VICT3R database. Retrieval from the database for creation of the VCGs is facilitated by the here developed R package. The evaluation of the tools was performed by the development of unit tests, run time analysis and user satisfaction forms. The run time evaluation indicates that these tools execution times were mainly influenced by their inputs, while still revealing good scalability of the SEND Data Quality Control Pipeline and their overall low run times for the tested real-world case scenarios. Ultimately, the user feedback highlights the good usability of both the Pipeline and the R package. The user interface of the Pipeline was rated as “Excellent” by one third of the participants and as “Good” by 50%, while considering the R package, 84% found its functions easy to use. These results emphasize the tools valued contribution to the VICT3R project. | eng |
| dc.description.sponsorship | No 101172693 | |
| dc.identifier.tid | 204178746 | |
| dc.identifier.uri | http://hdl.handle.net/10400.22/31889 | |
| dc.language.iso | eng | |
| dc.rights.uri | N/A | |
| dc.subject | Virtual control groups | |
| dc.subject | Toxicology studies | |
| dc.subject | Data pipeline | |
| dc.subject | R Package | |
| dc.title | Design and implementation of an R-Based Infrastructure for an improved flow of data across the VICT3R Project | eng |
| dc.type | master thesis | |
| dspace.entity.type | Publication | |
| thesis.degree.name | Master in Biostatistics and Bioinformatics Applied to Health |