Aha awkwardly this is actually for the MRC BioMed MRC DTP program: gw4biomed.ac.uk. Similar programs but different application process! 🙃
Oh I also worked out how to do pie chart points in ggplot2 about 5 years ago for this project. Bit of a niche use case maybe but makes lovely plots. Used it to show community composition of the resident community.
We think that tests of coexistence should be routine for model microbial communities - especially those explicitly testing theory where that is the assumption - and outline ideas (some of which we have started to do) on how this community can be harnessed for future research!
As ever, all code is available here: github.com/padpadpadpad..., the paper is open access with the excellent society owned journal Microbiology, and the data availability statement for this one is LONG reference genomes are on ncbi and isolates are in a culture collection!
Crucially for evolutionary studies, the isolates have different colony morphologies so can be re-isolated to look at evolutionary and phenotypic changes which is neat. They vary in their species interactions and we have high quality reference genomes for each species.
Here we characterise a stable 5-species community that we created by disassembling a start community of 46 isolates and seeing what could stick together at the end. We then experimentally checked that all species could coexist and found that the five species can be cultured together for >1 year!
This project is part of the GW4 SWBioscience Doctoral Training Partnership. And is so far my 5th attempt at getting a PhD student! 😆 Come join a supportive lab group, do exciting science, and learn wet lab techniques, programming, bioinformatics, statistics, and more!
Based at the Penryn campus of University of Exeter. Please share with anyone you think might be interested! Deadline for applications is the 4th November. Happy to field questions. Details on how to apply here: www.exeter.ac.uk/study/fundin...www.swbio.ac.uk#microsky
Obviously @jburnmurdoch.bsky.social stuff. I think it's the text annotations in plot that really make them so much more understandable than the classic plots in a paper.
How will you know the AI written code is recreating the idea in the paper? Or just subtly changing something?