Breaking News
admin
May 31 2025
4A growing concern in the global scientific community is the inability to reproduce a significant portion of published research findings—an issue commonly referred to as the reproducibility crisis. This challenge is particularly evident in the biomedical sciences, where studies suggest that fewer than half of research findings can be successfully replicated. In response, researchers in Brazil launched a large-scale initiative to assess and address the reproducibility of experiments conducted within the country.
The Brazilian Reproducibility Initiative (BRI), founded in 2017, was established to examine the extent of this issue at a national level. Led by neuroscientist-turned-meta-researcher Olavo Amaral and joined by meta scientist Clarissa Carneiro, the initiative aims to identify factors contributing to poor reproducibility and suggest local solutions tailored to Brazil's research environment.
Carneiro, who now serves as the co-executive director of the Brazilian Reproducibility Network, first encountered the reproducibility challenge as an undergraduate studying pharmacy at the Federal University of Rio de Janeiro (FURJ). Her attempts to recreate a behavioral model from a published study yielded no results, prompting self-doubt and further investigation. She eventually discovered that such challenges were not unique, with even pharmaceutical companies reporting difficulty in replicating academic experiments.
Carneiro noted that the problem extends beyond method failure; it includes a lack of clear standards to validate methodologies. This ambiguity complicates the interpretation of negative results, making it difficult to determine whether a drug is ineffective or whether there was a deviation in execution.
Recognizing the need for a structured approach, Amaral and Carneiro designed the BRI as a multicenter, one-shot replication effort. Unlike prior studies that focused on a single field across multiple regions, BRI concentrated on experiments conducted by Brazilian researchers, allowing for dialogue with local institutions and funders. The team reviewed over 2,700 publications to select common experimental techniques and ultimately chose to replicate studies involving a cell metabolism test, a gene amplification assay, and a behavioral analysis.
To ensure consistency, Carneiro transcribed each study’s methods section into a working protocol. Participating labs—56 in total—were then recruited from across Brazil to carry out the replications. However, the COVID-19 pandemic disrupted progress, causing staff turnover in many labs. This, in turn, introduced inconsistencies in protocol interpretation, highlighting a broader issue: even experienced researchers often diverged in understanding standard methodological terminology.
For example, Amaral pointed out the ambiguity in sample size descriptions such as “n of three” in cell culture experiments. It remained unclear whether this referred to three independent cultures, three subsamples from a single culture, or repeated measurements over time.
Once the replication data was gathered, a joint review was conducted by BRI team members and replication participants to assess whether deviations in protocol warranted the exclusion of some results. Approximately one-third of replications were excluded for this reason. The final analysis included 97 replications spanning 47 experiments, with reproducibility rates ranging between 15% and 45%, aligning with trends observed in global studies.
Importantly, Amaral emphasized that these low replication rates don’t necessarily imply fraud or error but reflect gaps in how research methods are communicated, especially regarding lab conditions and experimental nuances. Carneiro agreed, highlighting the central issue as one of scientific communication.
Looking forward, the BRI plans to investigate the unique challenges associated with each method they tested and propose targeted improvements. Both Carneiro and Amaral continue to advance these efforts through the Brazilian Reproducibility Network, which encourages nationwide collaboration to develop tools and strategies that enhance the credibility of scientific research.
As the global scientific community grapples with the reproducibility crisis, Brazil’s initiative offers a local model for systematic change, emphasizing transparency, methodological rigor, and cross-institutional collaboration as the path forward.
Source:https://www.the-scientist.com/brazilian-scientists-evaluate-reproducibility-in-biomedical-research-73055
This is non-financial/medical advice and made using AI so could be wrong.
