A multicenter, multidisciplinary initiative to promote transparent and reproducible research in the Brazilian scientific community
Rio de Janeiro, Brazil
Joined June 2018
- Tweets1.1K
- Following376
- Followers1K
- Likes42
Brazilian Reproducibility Network retweeted
My Claude-generated grievance of the @arxiv cap, based on a real human rant:
pracheeac.substack.com/p/arx…
Brazilian Reproducibility Network retweeted
More writing output never did mean more knowledge gained. We knew that. And yet we continued to act like it did. This isn't AIs fault. It's science's.
We have to stop using "paper count" as a useful scientific metric. More writing output no longer means more knowledge gained
washingtonpost.com/education…
Brazilian Reproducibility Network retweeted
Today I'm posting a new blog on a familiar topic... Nearly two hundred papers have confused the names of different proteins and recorded using the wrong antibody! This time it's PERK vs. p-ERK. Researchers from Harvard, the Mayo Clinic and more. Link in the first comment.
Brazilian Reproducibility Network retweeted
The Metacheck team is busy building an improved version of the osf_download function (scienceverse.org/metacheck_b… ). You will soon be able to use metacheck to download all files, wikis, and meta-data. You will also be able to upload everything to Zenodo.
Big news! 🎉
BrRN has just received a grant from @TempletonWorld for a research project evaluating strategies to promote open and reproducible practices in Brazilian graduate programs.
reprodutibilidade.org/en/pos…
Grande novidade! 🎉
A RBR acaba de receber um grant da @TempletonWorld para um projeto de pesquisa que avalia estratégias de promoção de práticas abertas e reprodutíveis nos programas de pós-graduação brasileiros.
reprodutibilidade.org/post/r…
The paper can be found at osf.io/preprints/metaarxiv/8…. We will also be presenting these and other results from the project at the @METRICStanford International Forum (metrics.stanford.edu/metrics…) tomorrow at 9 am Pacific/1 pm Brasília.
We have a new preprint on challenges and lessons learned from the Brazilian Reproducibility Initiative at @metaarxiv!
BRISA, our Brazilian preclinical systematic review collaboration, has a new preprint on the effect of NMDA antagonists on social behaviour. Check it out at biorxiv.org/content/10.64898…
We are pleased to announce the launch of 10 recommendations for open and reproducible research in graduate programs!
The document presents practical and adaptable guidelines, including resources, tools, and examples.
Read more (in Portuguese) here: osf.io/m4csj
Brazilian Reproducibility Network retweeted
Join us on Thursday, March 5, for a free webinar exploring what Diamond Open Access is, why it matters for fair and equitable scholarly publishing and how to transition to community-led, author- and reader-free publishing models.
elifesciences.org/events/942…
We’ll be represented by Abel Packer from @RedeSciELO and @olavoamaral from our coordinating team!
eLife’s Global South Committee is hosting a free webinar exploring Diamond Open Access, featuring a range of industry perspectives and highlighting successful implementations from around the world.
Register free: elifesciences.org/events/942…
It was frustrating to find errors on the initial preprint, but this has taught us a lot both about the current state of data management in biomedical labs and about how hard it is for large collaborations to rely on asynchronous communication for most processes.
We also performed a few other changes in response to reviewers in a journal - particularly a sensitivity analysis using the mean of all experimental units across replications, as opposed to meta-analysis. These are listed at osf.io/6av7k/r5y9x
The combination of both changes take our replication rates to 20-44% (depending on criteria), as opposed to 15-45% in the previous version. Not a major change, but we do believe the new rates are more accurate (especially for criteria based on statistical significance).
Due to insufficient reporting, it's hard to know whether this leads the analysis to be closer or further away from the original one, but we felt it was inadequate to keep the replications underpowered when this could be fixed.
We also changed the analysis of MTT experiments to perform pairing between units measured on the same day by default, as we realized that not doing this led to a major decrease in statistical power, since absolute optical density values were extremely variable across days.
This led to a full revision of data spreadheets, with any doubts cleared up live with the lab performing the experiment, leading to data changes (usually minor) in 18% of replications and changes in validation decisions for 5% of them. These are listed in osf.io/4tu2v
The preprint is online at doi.org/10.1101/2025.04.02.6…. The first important thing is that online meetings with the labs to discuss results uncovered some data errors that had slipped through the asynchronous data checking process we had previously performed.