Patching Science - amending the literature through version control

Kane, A.; Amin, B.

doi:10.1101/2022.04.13.487348

Home/Archive/bioRxiv

bioRxiv · 10.1101/2022.04.13.487348

Patching Science - amending the literature through version control

Kane, A.·Amin, B.

View original source ↗Export metadata ↓

Abstract

The ideal of self-correction in science is not well served by the current culture and system surrounding amendments to published literature. Here we report on a survey (N = 132) that highlights academics dissatisfaction with the status quo and their support for an alternative approach. We then describe our view of how amendments could and should work by drawing on the idea of an author-led version control system. Here authors would include a link in their published manuscripts to an updatable website (e.g. a GitHub repository or similar) that could be disseminated in the event of any amendment. Such a system is already in place for computer code and, as such, requires nothing but buy-in from the scientific community - a community that is already evolving towards various open science frameworks. This would remove a number of frictions that discourage amendments thus leading to an improved scientific literature and a healthier academic climate.

Explore related subjects

scientific communication and education Preprint

Keep this discovery

Explore connections, maps & timelines

BibTeX RIS

Kane, A., Amin, B.. 2022-04-14. Patching Science - amending the literature through version control. https://doi.org/10.1101/2022.04.13.487348

Cite the original work for its findings. Save a collection to share your selection of sources.

KEEP EXPLORING

Related preprints

bioRxiv2026

Evaluating Large Language Models as Tools to Navigate Researchers in Rapidly Evolving Research Landscapes: A Case Study in Cancer Drug Response Prediction

Large Language Models (LLMs) have emerged as promising tools for assisting researchers in automating and accelerating the synthesis of literature reviews. However, their reliability is a significant concern due to issues like factual inaccuracies and hallucinations. The key question is whether LLMs can reliably provide comprehensive, up-to-date overviews and analyses. This study evaluates the performance of three leading LLMs (OpenAI's ChatGPT, Google's Gemini, and DeepSeek) on the complex task of generating a comprehensive survey paper on deep learning for cancer Drug Response Prediction (DRP). By testing both standard and Deep Research (DR) / Deep Think (DT) modes of LLMs with prompts of varying detail, this paper assesses key academic dimensions, including reference management, content quality, and analytical depth. Key findings reveal that while DR modes of LLMs significantly improve reliability by eliminating hallucinations, performance variations exist across models and prompts. A trade-off between reference quantity and integration quality was observed, and even the best-performing models lacked the analytical depth of human experts, often requiring extensive human supervision. The study concludes that LLMs currently serve as powerful assistive tools but still cannot replace the critical validation and synthesis provided by human researchers. Choosing the best LLM to use depends on the task in hand, while several strategies can be implemented to improve the produced output.

scientific communication and education↗

bioRxiv2023

Insights from a survey of mentorship experiences by graduate and postdoctoral researchers

Mentorship is vital for early career researchers in training positions, allowing them to navigate the challenges of work and life in research environments. However, the quality of mentorship received by trainees can vary by investigator and by institution. One challenge faced by those hoping to improve trainee mentorship is that the extent to which mentorship is offered to and experienced by research trainees is not well characterized. To address this knowledge gap, we conducted a survey to examine the quality of mentorship received by trainees in research environments, to identify characteristics of positive and negative mentorship, and to highlight best practices to improve trainee mentorship. We received 2,114 responses from researchers at graduate and postdoctoral career stages worldwide. Quantitative analysis showed that at least [~]25-45% of respondents were dissatisfied with some aspects of their mentorship. Qualitative responses revealed that common issues in mentorship include unclear expectations in research and mentoring interactions, lack of guidance, and inadequate support of trainee independence and career goals. Our findings also identified key mentorship elements desired by trainee mentees. Based on trainee suggestions, we describe strategies for individual mentors, departments, and institutions to improve the training experience for graduate and postdoctoral researchers.

scientific communication and education↗

bioRxiv2023

Supporting diamond open access journals. Interest and feasibility of direct funding mechanisms

More and more academics and governements consider that the open access model based on Article Processing Charges (APC) is problematic, not only due to the inequalities it generates and reinforces, but also because it has become unsustainable and even opposed to open access values. They consider that scientific publishing based on a model where both authors and readers do not pay - the so-called Diamond, or non-APC model - should be developed and supported. However, beyond the display of such a support on an international scale, the landscape of Diamond journals is rather in the form of loosely connected archipelagos, and not systematically funded. This article explores the practical conditions to implement a direct funding mechanism to such journals, that is reccurent money provided by a funder to support the publication process. Following several recommendations from institutional actors in the open access world, we consider the hypothesis that such a funding would be fostered by research funding organizations (RFOs), which have been essential to the expansion of the APC model, and now show interest in supporting other models. Based on a questionnaire survey sent to more thant 1000 Diamond Open Access journals, this article analyzes their financial needs, as well as their capacity to interact with funders. It is structured around four issues regarding the implementation of a direct funding model : Do Diamond journals really make use of money, and to what end ? Do they need additional money? Are they able to engage monetary transactions? Are they able to meet RFOs visibility requirements? We show that a majority of OA Diamond journals could make use of a direct funding mechanism with certain adjustments. We conclude on the challenges that such a financial stream would spur.

scientific communication and education↗