Abstract Objective: Accurate nematode instance segmentation is critical in biomedical informatics, enabling rapid detection and localization of nematodes to provide data support for biomedical research on human aging mechanisms, drug screening, and neurological disorders. Despite the impressive performance of visual foundation models in general image segmentation tasks, their accuracy declines with nematodes due to the organisms' complex characteristics, including multi-scale variations, curled morphologies, and frequent occlusions. Therefore, we propose a multilevel-prompt SAM architecture, enhancing segmentation precision.
Methods: This study leverages the Segment Anything Model (SAM) as a foundational framework and systematically investigates its few-shot learning performance under diverse prompting strategies across three public nematode datasets and a proprietary dataset from the China Manned Space Engineering program. The proposed novel architecture sequentially integrates global point prompts and local box prompts, offering enhanced multi-view guidance for segmenting multi-scale nematode instances. To evaluate the efficacy of our approach, we conduct comparative experiments against the original point-prompt SAM and fine-tuned single-level prompt SAM model. We also conducted comparative experiments with baseline models.
Results: The results demonstrate the superior performance of the multilevel-prompt SAM, with substantially improvements in segmentation accuracy across all datasets. Using mAP 50:95 , mAP 50 , mAP 75 , Dice Score and Hausdorff Distance as evaluation metrics, our approach notably achieved 15.21% and 3.51% higher mAP 50:95 on the Mating dataset compared to Mask2former and OMGSeg for small-scale nematode targets.
Conclusion: This study validates the practical applicability of the proposed methods for analyzing experimental data from the biomedical experiments and the Chinese Space Station using computer vision technologies, highlighting its potential for advancing fundamental life science research. Keywords: Instance Segmentation, Nematode Image Segmentation, Multilevel-prompt.

BioNatura Journal celebrates its third anniversary by denoting a transition from an emerging publication to a consolidated scientific platform, driven by foundational efforts in 2024 and 2025. This period emphasized operational maturity and moral standards, highlighted by a constant publication cadence of four issues per year. Technological evolution was achieved through a formal migration to Open Journal Systems (OJS) to standardize metadata and ensure digital archiving via PKP PN and LOCKSS/CLOCKSS. Furthermore, the journal implemented an "Online First" policy to eliminate artificial delays in knowledge dissemination while retaining persistent DOIs. The success of its Open Access model is evidenced by over 1.16 million PDF downloads across 116 countries in 2025, as well as its inclusion in the OpenAlex ecosystem to foster candid communication. With a rigorous double-blind peer-review process and a diverse editorial board representing 22 countries, BioNatura published 180 articles involving authors from 27 nations. These results support the journal's role as a mature, international bridge for scientific collaboration in biotechnology and life sciences.

Abstract In this research article, we present the first cross-disciplinary descriptive analysis of the use of contribution statements. Our main objective is to obtain further insight on contributions by a variety of fields (Multidisciplinary, Health, Life, Physical, and Social Sciences) from the largest dataset used up to now. We examine more than 700,000 articles published between 2018 and 2023 in Elsevier and PLOS journals, in combination with bibliometric data extracted from the Scopus database. The descriptive analysis of the dataset focuses on the overall coverage of the merged data, the distribution of authorship and disciplines at paper level, and the interactions between contribution statements, author order, and disciplines. Our two main findings indicate that, on the one hand, looking at contributions and authorship order can enrich the way we understand science as a social endeavor. On the other hand, delving deeper into contributorship differences by field is key. We underscore the value of the Contributor Role Taxonomy (CRediT) in unveiling nuanced research dynamics and offering a more equitable framework for evaluation.

Enantioselective recognition of chiral molecules, particularly cysteine enantiomers (L/D-Cys), is crucial in life sciences, drug development, and clinical diagnostics due to their distinct physiological effects. However, conventional methods often lack sufficient selectivity and sensitivity. Herein, a molecularly imprinted electrochemical sensor based on three-dimensional (3D) heterostructured Mn1/3Zn2/3CO3-rGO-CNTs was developed for highly selective detection of cysteine enantiomers. The Mn1/3Zn2/3CO3-rGO microspheres were successfully synthesized via a two-step co-precipitation-hydrothermal strategy, and combined with carbon nanotubes to construct a 3D carrier with high specific surface area and excellent conductivity. XPS, SEM, and TEM characterization studies demonstrated its suitability as an ideal platform for molecular imprinting. Using L-Cys or D-Cys as the template molecule, the chiral recognition sites were constructed on the modified electrode interface via electropolymerized polypyrrole, achieving efficient enantiomer differentiation. Combined with a potential dynamic rapid-switching detection mode, the sensor exhibited significantly enhanced detection efficiency with a response time of 2 s. This proposed sensor demonstrated a wide linear range of 1 × 10-7 to 1 × 10-3 mol L-1 for L-Cys detection, and exhibited significant enantioselectivity at the same concentration (L/D = 4.248 ± 0.01; D/L = 5.033 ± 0.50). This method offers high sensitivity and stability, and applicability for complex biological sample analysis, providing a robust technical platform for chiral recognition and pharmaceutical research.

Abstract Previous research has shown that journal article quality ratings from the cloud based Large Language Model (LLM) families ChatGPT and Gemini and the medium sized open weights LLM Gemma3 27b correlate moderately with expert research quality scores. This article assesses whether other medium sized LLMs, smaller LLMs, and reasoning models have similar abilities. This is tested with Gemma3 variants, Llama4 Scout, Qwen3, Magistral Small and DeepSeek R1 on a dataset of 2780 medical, health and life science papers in 6 fields, with two different gold standards, one novel. Few-shot and score averaging approaches are also evaluated. The results suggest that medium-sized LLMs have similar performance to ChatGPT-4o mini and Gemini 2.0 Flash, but that 1b parameters may often, and 4b sometimes, be too few. Reasoning models did not have a clear advantage. Moreover, averaging scores from multiple identical queries seems to be a universally successful strategy, and there is weak evidence that few-shot prompts (four examples) tend to help. Overall, the results show, for the first time, that smaller LLMs > 4b have a substantial capability to rate journal articles for research quality, especially if score averaging is used, but that reasoning does not give an advantage for this task; it is therefore not recommended because it is slow. The use of LLMs to support research evaluation is now more credible since multiple variants have a similar ability, including many that can be deployed offline in a secure environment without substantial computing resources.

Parameter-wise predictions and sensitivity analysis for random walk models in the life sciences.

Recent developments and applications of photothermal AFM-IR in characterization of amyloids and amyloids aggregation processes: Mini-review.

Biological models with nonlocal terms: Future scopes of research: Comment on "Nonlocal models in biology and life sciences: Sources, developments, and applications" by S. Pal & R. Melnick.

Multiple approaches revealed ImKC cells as a RAW 264.7 derivative rather than a Kupffer cell line.

Mitochondria-targeted fluorescent probes based on the FRET principle for the detection of sulfite.

Double-stranded endonuclease activity exploration through scanning substrate assay and its biosensing application.

Recent advances in spectroscopic chiral analysis of carboxylic acids.

The physical world is composed of graphs, such as the protein structures in life science, the patient relations in medical diagnosis, the user connections in social media, etc. Graphs help both build the world itself and understand the semantics behind the data for humans. However, how such graph structures work toward semantic representation is still unclear, where existing attempts focus on employing the graphs for special tasks. In this work, we first introduce two measures to evaluate graph quality, namely structural complexity and homophily. Structural complexity describes the quantity of graph structural information representing the graph structure's symmetry, and homophily describes the percentage of intra-class edges to quantify edge consistency. Using these two measures, we then discover the relationship between the graph quality and the corresponding performance for general tasks, that is the performance positively correlates with the structural complexity, and "J"-shaped correlates with homophily, which are proved mathematically. Based on these, we design a graph augmentation tool Graph$^+$+. Graph$^+$+ can enhance the natural graph structure and accordingly improve the general tasks. Empirical validation on tasks including Alzheimer's diagnosis and breast cancer subtype identification shows Graph$^+$+'s ability to improve both graph structure and task performance, revealing the underlying data semantics.

On implementation of nonlocal terms in models in ecology and evolution. Comment on 'Nonlocal models in biology and life sciences: Sources, developments, and applications' by S. Pal & R. Melnick.

Undergraduate students in physiology and other life science programs commonly intend to pursue careers in research or health care professions. Often, however, they are poorly prepared for research and are discouraged when they are not accepted into professional programs. To address this, we created Research Readiness and Advancing Biomedical Discoveries, a third-year course for life science students. The course includes a scaffolded research proposal assignment along with online preclass modules and extensive in-class group work on topics such as career planning, project management, and commercialization. To assess the course, we surveyed students both during and 1-3 yr after course completion. Students agreed or strongly agreed that the course prepared them for research opportunities (4.17 ± 0.07, mean ± SE, 5 = strongly agree) and prompted them to consider flexible career paths (4.26 ± 0.06). Of the 63 former students (alumni) who completed the survey (a 29% response rate) almost two-thirds were pursuing advanced degrees, with most of the others in science-related positions. Alumni indicated that the course improved their skill set to achieve future goals (4.14 ± 0.10), consider flexible career paths (4.16 ± 0.10), and prepare for research opportunities (4.00 ± 0.12). The most common themes to the question "What aspects of the course helped with your current career pursuits?" were working in teams (68%) and developing and writing an original research proposal (68%). Our findings will encourage others to contemplate a similar course or activities/assignments and illustrate the value of surveying former students.NEW & NOTEWORTHY To better prepare students for research and a range of science careers, we designed and delivered a course to develop teamwork, communication, creative, critical thinking, and career planning skills and an understanding of how research is conducted and translated. Student feedback obtained during the course and from alumni who had completed the course 1-3 yr before demonstrates the value of course content, assignments, and group work in fostering learning and shaping career trajectories.

Pioneers: Glucose Sensing and Control of Health-span and Lifespan.

Evaluating short-term study abroad programs with self-reported growth and objective testing.

Spatial transcriptomics: integrating platforms and computational approaches for clinical insights.

Photon FLASH spares radiation-induced changes in cardiac function, remodelling and arrythmia in a preclinical model.

Pioneers: Academic Career in Cryo-EM Structural Biology.