## Summary I led a session today at the LSHTM [Modern Techniques in Modelling](https://www.lshtm.ac.uk/study/courses/short-courses/modern-techniques-disease-modelling) course on best computational practice and along with everyone else running the session generated a list of resources in the chat after the session. These resources are examples of projects implemented with different styles, some recommendations for blogs to read, recommendations for packages, and finally the number one piece of advice from all R users to all R users read [R for Data Science](https://r4ds.had.co.nz/) and [Advanced R](https://adv-r.hadley.nz/) if you are feeling fancy. I'll likely circle back to this post and update as more examples occur and potentially move it onto my main site if it gains enough novel content. ## Project examples Note this represents only examples drawn from projects I have recently been involved with. What should be clear is that all these examples share the same structure in that they are documented, they make use of folders (each storing a single type of content), they have some kind of overall control method for reproducing the analysis, and each idea is defined in a function. In my view if you do these things (in what ever way you want) you can't go far wrong. The key is to spend some time at each stage of a project thinking about how you are implementing it. - [A simple script using functions and comments to demonstrate an implementation of a deterministic deconvolution](https://gist.github.com/seabbs/fb1bc9c79c3dd7117f9314cb97e71615) - [A slightly more involved example using folders, and `{Rmarkdown}`](https://github.com/epiforecasts/omicron-sgtf-forecast) to produce real-time reports characterising the Omicron COVID-19 variant. - [A project to evaluate a novel nowcasting method using Germany COVID-19 hospitalisations](https://github.com/epiforecasts/eval-germany-sp-nowcasting) making use of the `{targets}` package. - [Early work estimating the potential transmissibility of Omicron in South Africa](https://github.com/SACEMA/omicronSA) by Pearson et al. making use of `make`. ## Blogs - [Rweeky](https://rweekly.org/) - [rOpenSci](https://ropensci.org/blog/) - [Project-oriented workflow by Jenny Bryan](https://www.tidyverse.org/blog/2017/12/workflow-vs-script/) - [Project as an R package: An okay idea by Miles Mcbain](https://milesmcbain.xyz/posts/an-okay-idea/) - [How to survive a laptop fire by Miles Mcbain](https://www.milesmcbain.com/posts/how-to-survive-a-laptop-fire/) ## Books - [R for Data Science](https://r4ds.had.co.nz/) - [Advanced R](https://adv-r.hadley.nz/) ## Packages - [`{targets}`](https://here.r-lib.org/) - [`{here}`](https://here.r-lib.org/) ## Tags #blog/2022/ #software-engineering #rstats #stream-of-consciousness