The “Climate Sensitive Infectious Disease” (CSID) network wants to connect a global community of actors that contribute to impactful CSID software tools and localized CSID communities that can link existing issues and initiatives to the development and maintenance of CSID tools.
On July 21, during the 2025 Annual General Meeting of the CSIDNETYanina Bellini Sibbene and Samuel Schildhauer lead a session with the title entitled “Open science, R -packages and development software development”. The session was not only designed to introduce participants to tools and practices in Open Science with R, but also to encourage active reflection and discussion about their own workflows and the challenges with which they are confronted.
Things from kicking: the human barometer
To break the ice and get people moving and think Documented method of a ROPSCI -OUNF, called the human barometer. Participants physically positioned themselves along a line in response to a series of light -hearted and thinking instructions, such as:
- Cats or dogs?
- I practice open science in my activity.
- I use R in my research.
Participants then discussed their positions and tried to convince the other group to change your mind. This dynamic approach not only brought energy into the room, but also offered a quick look in the practices and perspectives of the community, giving the participants the opportunity to get acquainted interactively.
Participants share why they are in that group when he is asked: “I use R in my research.” Photo by Yanina Bellini Sibbene.
Mapping our tools: Data Science Workflows in practice
The first structured discussion was aimed at the tools that participants use in their Data Scienceworkflows. With the help of the well -known R For Data Science (R4DS) FrameworkThe participants were invited to post post-it notes under each step of the process import, tidy, transform, visualize, model and communicate-what indicates which tools or packages they trust.
Yanina and Sam who guide the conversation about the tools used by participants in their data science work flows and of the participants who talk and place sticky color notes in the diagram. Photos by: Natthaphon Sakulvanaporn (Bangkokvdo)
This exercise led to rich conversations about the diversity of tools that people use, the areas with strong support (such as data visualization) and the steps where participants often struggle to find the right tools. It was also noted that tools such as Arrowso that large amounts of data can be treated in R, and goMaking work with pipelines possible were not mentioned between the tools. The exercise opened the door to discuss gaps in tooling – advocate for future development or cooperation, such as simpler mapping – tools and shared code brokes between different notebooks in R.
The table summarizes the tools that the attendees shared for each of the steps in the process.
Discover barriers: a chalk junction about obstacles
The second discussion focused on the focus on obstacles that participants are confronted with trying to deal with open science and research software development. In a chalk jout format, participants wrote down their thoughts under three color -coded categories:
- Expertise Obstacles (e.g. Lack of training, steep learning curves)
- Language obstacles (eg lack of documentation in local languages)
- Infrastructure obstacles (eg limited access to hardware or internet)
Participants in the session and host during the Chalk Talk discussion. Photos by: Natthaphon Sakulvanaporn (Bangkokvdo)
This open and honest exchange revealed a wide range of barriers. It also underlined the importance of initiatives relating to fairness and access in research infrastructure.
Expertise Obstacles: Knowledge of expertise, bad/no update documentation, not knowing what is (avoid the wheel again), packages that are no longer supported, which packages/function to use, lack of a leather route map, problem shooting for less used packages, translations between different fields fields
Linguist: Human language (material availability), multilingual documentation, training and documents in native languages, documentation too complex, not clear examples of use, need for a rather understanding
Infrastructure obstacles: Equipment, processing sources, storage capacity, external HPC access, hosting for app, financing software development
Sharing solutions: Use-Case Presentations
To ground the discussion in real-world examples, Sam and Yani each presented a use case:
Serodynamics: Development of collaborative R-package
SAM is part of the Serepidemiology Research Group (SERG) at UC Davis, led by Dr. Kristen Aiemjoy and Dr. Ezra Morrison. Sam gave a presentation about the development of his team of an R package called Serodynamics.
The package implements an innovative technique that uses data from longitudinal serological cohorts to model antibody -kinetic serore series for a certain infection. This information can then be applied to Transversal Serosurveys to estimate the incidence and time of the disease since infection with another package that the team has developed with the name Serocalculator. The package uses a Bayesian (Runjags) framework to estimate the initial antibody concentration (Y0), peak concentration (Y1), time to peak (T1), explanatory speed (α) and feature form (R). The SERG team hopes to have serodynamics ready for use in the coming year.
Conceptual diagram for estimating serocon version speed using longitudinal antibodic lady dynamics
Ropsci: Support for open and sustainable research software
Yani introduced the ROPSCI-ECOSY SYSTEM and emphasized how it supports the development, peer review and sustainability of research software in R.. She emphasized the community -driven approach and the importance of good practices in software development for science.
Us Software Peer Review is a transparent, open, non-advisorar R-package assessment with the aim of improving the software that we build to make science.
R-Universe Is the ROPSCI platform for improving the publication and discovery of research software in R. R-Universe Build the Binaries for various OS, including WebAssembly! It also builds the documentation, distributes datasets and keeps up with many statistics about your package.
Us Multilingual publication project Focus on creating the process and the infrastructure to offer our content in other languages ​​than English, while the quality of that content is guaranteed. It also seems to offer the R -community with multilingual possibilities in different contexts.
Us Champions program Tryes to identify, recognize and reward leaders in a community. It is aimed at people from historically and systematically excluded groups that are interested in contributing to Ropenci and the wider open source and open science communities. Through mentoring and training, champions develop a project for a year.
Yani said how each of the ROPSCI projects can serve as aids to overcome some of the barriers mentioned in the discussions, such as Peer Review and the community of developers who not only maintain but also generate documentation, use Cases and answer questions, with research-oriented packages, even if they are niche. Also the champion program as a learning path For scientists who develop research software.
She also shared the existence of pyopsci For people who program in Python and The Timpentry’s For introductory workshops and open lessons.
Last thoughts
This session offered a mix of interaction, reflection and learning. It helped participants to connect their personal experiences with broader open science practices, gave up common challenges and introduced them in new tools and communities. Towards the end of the 90 minutes it was clear that the path to more open, accessible and sustainable research software went together best. If you are interested in finding out more about upcoming CSIDNET events and sources, you can sign up for the main list.
Related
#Research #Open #Science #Packages #research #software #development #CSIDNET #AGM #RBloggers