Evaluation of Large Language Model Performance and Reliability for Citations and References in Scholarly Writing: Cross-Disciplinary Study.

BACKGROUND: Large language models (LLMs) have gained prominence since the release of ChatGPT in late 2022. OBJECTIVE: The aim of this study was to assess the accuracy of citations and references generated by ChatGPT (GPT-3.5) in two distinct academic domains: the natural sciences and humanities. METHODS: Two researchers independently prompted ChatGPT to write an introduction section for a manuscript and include citations; they then evaluated the accuracy of the citations and Digital Object Identifiers (DOIs). Results were compared between the two disciplines. RESULTS: Ten topics were included, including 5 in the natural sciences and 5 in the humanities. A total of 102 citations were generated, with 55 in the natural sciences and 47 in the humanities. Among these, 40 citations (72.7%) in the natural sciences and 36 citations (76.6%) in the humanities were confirmed to exist (P=.42). There were significant disparities found in DOI presence in the natural sciences (39/55, 70.9%) and the humanities (18/47, 38.3%), along with significant differences in accuracy between the two disciplines (18/55, 32.7% vs 4/47, 8.5%). DOI hallucination was more prevalent in the humanities (42/55, 89.4%). The Levenshtein distance was significantly higher in the humanities than in the natural sciences, reflecting the lower DOI accuracy. CONCLUSIONS: ChatGPT's performance in generating citations and references varies across disciplines. Differences in DOI standards and disciplinary nuances contribute to performance variations. Researchers should consider the strengths and limitations of artificial intelligence writing tools with respect to citation accuracy. The use of domain-specific models may enhance accuracy.

Exploring stem cell frontiers: definitions, challenges, and perspectives for regenerative medicine.

Each year, the European Summer School on Stem Cell Biology and Regenerative Medicine (SCSS) attracts early-career researchers and actively practicing clinicians who specialise in stem cell and regenerative biology. The 16th edition of this influential course took place from 12th to 19th September 2023 on the charming Greek island of Spetses. Focusing on important concepts and recent advances in stem cells, the distinguished faculty included experts spanning the spectrum from fundamental research to clinical trials to market-approved therapies. Alongside an academically intensive programme that bridges the various contexts of stem cell research, delegates were encouraged to critically address relevant questions in stem cell biology and medicine, including broader societal implications. Here, we present a comprehensive overview and key highlights from the SCSS 2023.

Tools and methods for high-throughput single-cell imaging with the mother machine.

Despite much progress, image processing remains a significant bottleneck for high-throughput analysis of microscopy data. One popular platform for single-cell time-lapse imaging is the mother machine, which enables long-term tracking of microbial cells under precisely controlled growth conditions. While several mother machine image analysis pipelines have been developed in the past several years, adoption by a non-expert audience remains a challenge. To fill this gap, we implemented our own software, MM3, as a plugin for the multidimensional image viewer napari. napari-MM3 is a complete and modular image analysis pipeline for mother machine data, which takes advantage of the high-level interactivity of napari. Here, we give an overview of napari-MM3 and test it against several well-designed and widely used image analysis pipelines, including BACMMAN and DeLTA. Researchers often analyze mother machine data with custom scripts using varied image analysis methods, but a quantitative comparison of the output of different pipelines has been lacking. To this end, we show that key single-cell physiological parameter correlations and distributions are robust to the choice of analysis method. However, we also find that small changes in thresholding parameters can systematically alter parameters extracted from single-cell imaging experiments. Moreover, we explicitly show that in deep learning-based segmentation, 'what you put is what you get' (WYPIWYG) - that is, pixel-level variation in training data for cell segmentation can propagate to the model output and bias spatial and temporal measurements. Finally, while the primary purpose of this work is to introduce the image analysis software that we have developed over the last decade in our lab, we also provide information for those who want to implement mother machine-based high-throughput imaging and analysis methods in their research.

Ending "domestic helicopter research".

"Helicopter research" refers to a practice where researchers from wealthier countries conduct studies in lower-income countries with little involvement of local researchers or community members. This practice also occurs domestically. In this Commentary, we outline strategies to curb domestic helicopter research and to foster equity-centered collaborations.

Professional quality of life in animal research personnel is linked to retention & job satisfaction: A mixed-methods cross-sectional survey on compassion fatigue in the USA.

BACKGROUND: Working with research animals can be both rewarding and challenging. The rewarding part of the work is associated with understanding the necessity for animal research to improve the health of humans and animals and the knowledge that one can provide care and compassion for the animals. Challenges with animal research include witnessing stress/pain in animals necessitated by scientific requirements, end of study euthanasia, and societal stigmatization about animal research. These challenges could be compounded with more general workplace stresses, in turn, impacting job retention and satisfaction. However, these factors have yet to be formally evaluated. Therefore, the purpose of this survey was to comprehensively evaluate professional quality of life's correlation with key workplace metrics. METHODS: Six institutions were recruited to participate in a longitudinal intervention trial on compassion fatigue resiliency. This manuscript reports key baseline metrics from this survey. A cross-sectional mixed methods survey was developed to evaluate professional quality of life, job satisfaction, retention, and factors influencing compassion fatigue resiliency. Quantitative data were analyzed via general linear models and qualitative data were analyzed by theme. RESULTS: Baseline data was collected from 198 participants. Personnel who reported higher compassion satisfaction also reported higher retention and job satisfaction. Conversely, personnel who reported higher burnout also reported lower job satisfaction. In response to open-ended questions, participants said their compassion fatigue was impacted by institutional culture (70% of participants), animal research (58%), general mental health (41%), and specific compassion fatigue support (24%). CONCLUSIONS: In conclusion, these results show that professional quality of life is related to important operational metrics of job satisfaction and retention. Furthermore, compassion fatigue is impacted by factors beyond working with research animals, including institutional culture and general mental health support. Overall, this project provides rationale and insight for institutional support of compassion fatigue resiliency.

Do research collaborations age like wine? Absolute and relative measures of CANZUK research partnerships' strength since the 1950s.

Although previous studies of today's globalised and competitive research landscape have mentioned the research collaborations of CANZUK countries (i.e., Australia, Canada, New Zealand, and the United Kingdom), none have yet studied them in detail. Further, such studies have used different measures of international research collaboration (IRC), resulting in disparate findings. This paper, therefore, analyses the strengths of CANZUK research collaborations, how those collaborations have changed over time, and assesses the effect of three ways of measures on the results (absolute strength, bilateral similarity, and multilateral similarity). We provide a detailed characterisation of the CANZUK research network and its relationships with partner countries, which reveals that the most collaborative CANZUK countries are the UK and Australia, among other findings. We also confirm that many findings differ depending on which measures are used. We offer an explanation of this difference with reference to the nature of the measures (i.e., what they really measure) and make suggestions for suitable measures in future studies depending on their purpose. Finally, we discuss how this study's findings can be used by research policy makers (in CANZUK and elsewhere) in deciding on research strategy and by researchers in appropriately measuring IRC.

What Is "Key Information"? Consideration of the Reasons People Do or Do Not Take Part in Research.

We performed a qualitative review of 50 consent forms posted on Clinicaltrials.gov, examining the content of key information sections. We found that key information disclosures are typically focused on procedures, risks, potential benefits, and alternatives. Drawing upon reviews of the large literature examining the reasons people do or do not take part in research, we propose that these disclosures should be based more directly on what we know to be the real reasons why people choose to take part or refuse participation. We propose key information language for consideration by researchers and institutional review boards.