Nicola Romanò @nicolaromano@social.cleverlibre.org
Senior lecturer at ZJE and Edinburgh university.
I teach #imageanalysis & #dataanalysis with #RStats & #python. I study #heterogeneity in #pituitary (and other) cells.
Joined Oct 2022
Nicola Romanò
boosted
Also, if you build Shiny apps, you might want to consider submitting a proposal to be a speaker at ShinyConf2023 (@ShinyConf)
More details here: shinyconf.appsilon.com/become-a-speaker
Nicola Romanò
boosted
Digital publications without a print issue - "it will never work".
Open Access - "it will never work".
#Preprints - "it will never work"-
Open peer review - "it will never work".
#OpenData - "it will never work."
If I would write it today, I'd add:
Open source social infrastructure, hosted by academics. - "It will never work."
Nicola Romanò
boosted
Love, love, love to see the old "remote work is ineffective" argument being made after having just lived through two years of remote work. 🤪
How about you just say the truth? "I don't trust you and can't monitor you unless you're in the office."
Nicola Romanò
boosted
#️⃣ This video helped me add a column to Mastodon that contains various hashtags I want to follow.
⚙️ First you have to go to your preferences and enable advanced web interface.
@davead@fosstodon.org Looks very interesting, thank you!
Also, I realise I have visited the lab of the last author a few years ago, it's a small world!
@Aschniedermann Oh absolutely, not an easy topic... and terminology seems to be field- dependent, just to confuse matters even more! But it's great to have resources like the Turing handbook!
@artologica Well... grey squirrels are actually an invasive species here in Scotland and threaten the native red squirrels.
@mcaleerp As an Italian if feels wrong to pronounce it any other way than the second... anyway, however you pronounce it, I'm sure you're going to enjoy it!
@askennard Thank you! That's exactly the type of resource I was looking for! Now on to trying doing that!
Nicola Romanò
boosted
@nicolaromano apologies if this is already on your radar, but as someone outside the field I appreciated the clarity and detail in this paper https://elifesciences.org/articles/66747
I am wondering if anyone has experience in doing comparative analysis of #scRNAseq datasets from different species, or if anyone knows of good resources about that esp with an #evolutionaryBiology focus?
I'd like to point out this great little table about reproducible research.
This is from a great resource "The Turing Way handbook to reproducible, ethical and collaborative data science", which you can find here: https://the-turing-way.netlify.app/welcome.html
Nicola Romanò
boosted
@nicolaromano @datamaps @arnaudlerouzic @smach @MichaelIngrisch @rstats Plus you can easily write "base R" code that will break without the right system dependencies, that relies on a specific file structure, will only work on one OS, etc.. None of which CRAN is going to help you with. And personally I see those kind of problems in analysis scripts far more often than broken dependencies.
@datamaps @arnaudlerouzic @smach @MichaelIngrisch @rstats
That is exactly what I'm saying.
Reproducibility does not depend on whether you use external packages or not.
Let me clarify this further. Just because you use R base it doesn't mean the script is reproducible, just like using tidyverse doesn't make it non reproducible.
If you don't provide the information you mention (package version etc) then you're not guaranteed to be running in the same environment.
But, there is more to this.
Reproducibility is not just "been able to run the script without errors".
Many times I've seen an interesting paper, downloaded the code and got a different result. Or maybe found no instructions on how to use the scripts.
Or, very often only the script for performing a particular part of the analysis is given and how to get to that part is a mystery.
In my experience these other issues are way more common. Indeed I can only remember one or two instances where I had package version issues.
Nicola Romanò
boosted
Probably one of the most simple & interesting (for its utterly basic qualities) #cellularAutomata I've ever written...
Probabilistic (1% mutation chance per frame)
Lower mutation threshold (non-probabilistic)
Absolutely, what @Patrick Anker says.
Also, I think most problems of reproducibility (at least, those I encountered) are not dependent on package versions. The script might be incomplete, or the data to be fed to the script needs to be processed somehow and there are no clear instructions of how to do it etc.
These are much more common issues, at least in my field.
Nicola Romanò
boosted
@arinbasu1 @nicolaromano @rstats It is true that using base R has a higher likelihood of reproducibility in the near term than the tidyverse suite. However, there is still a chance that a change in R causes breaking changes (e.g. R 4.0 with the `stringsAsFactors` default switch). What Nicola's point is that it's better to avoid that situation altogether with renv, which generates a lockfile and restores pkg versions using MRAN, taking care of random breaking changes
@arnaudlerouzic @datamaps @smach @MichaelIngrisch @rstats
I think we're conflating two problems here.
Using base R does not ensure reproducibility, just like using tidyverse won't make a script less reproducible. For reproducibility, use renv, or similar solutions.
Loading tidyverse is a convenient way to have access to lots of functions without having to remember which package they come from. I like to be lazy like that of I'm just testing something... but will probably include single packages for a more important script (eg for a publication or a package)
Senior lecturer at ZJE and Edinburgh university.
I teach #imageanalysis & #dataanalysis with #RStats & #python. I study #heterogeneity in #pituitary (and other) cells.
Joined Oct 2022
