The Bacille Calmette-Guérin (BCG) vaccine against tuberculosis is the most widespread vaccine in the world. Discovered by French investigators Albert Calmette and Camille Guérin at the Pasteur Institute, it remains the only effective vaccine against tuberculosis infection. This report describes the recognition and identification of a previously unknown French handwritten laboratory notebook prepared by Drs. Calmette and Camille Guérin recording their experiments performed during the development of the BCG vaccine. The notebook was examined, translated into English, photographed and the experiments analyzed. The manuscript laboratory notebook consists of 69 leaves written in 2 hands, one of which corresponds to that of Albert Calmette. It contains details of experiments that were performed during the development of the BCG vaccine at the Pasteur Institute by Drs. Calmette and Guérin. These include experimental inoculations of rabbits and guinea pigs; descriptions of the pathology of skin lesions, inflammatory reactions, and organ pathology; and survival outcome. The experiments describe varying inoculative dosages of the bacteria, and different routes of administration including intraperitoneal and subcutaneous injections, and administration of bacilli in the ear. In those cases where the animal had died following inoculation of tubercle bacilli, necropsy was performed and the organs examined and the pathology findings described. Details of culture experiments and vaccine passage are listed. This previously unknown notebook is a highly organized and detailed record of investigations using tuberculosis in animal experiments and microbiological culture to produce a safe and effective vaccine, first used in humans in 1921.
Large language models (LLMs) offer promise for systematic review data extraction, but performance in complex multidisciplinary domains and utility for clinical statement generation remain insufficiently described. To evaluate Google NotebookLM for AI-assisted data extraction and RAND/UCLA consensus statement generation in a systematic review of IBD, obesity, and cardiometabolic comorbidities. Studies were organized into domain-specific notebooks; structured prompts generated standardized evidence tables. Two independent reviewers validated outputs against full-text articles using a four-category error classification. Cell-level accuracy and critical accuracy (cells free of major factual errors) were the primary metrics; workflow time was compared against a published conventional extraction benchmark. Concordance between AI-generated and expert-finalized statements was assessed. Across 57 articles, 1,710 data cells were extracted; 151 (8.83%) were flagged, yielding 91.17% cell-level accuracy. Major factual errors occurred in only 4 cells (0.23%), for a critical accuracy of 99.77%. Most errors were minor omissions (59.6%) or incomplete extractions (30.5%); domain error rates ranged from 7.08% to 11.33%. The pipeline required 17.7 versus a projected 165.1 person-hours (89.3% reduction). PICO-structured prompting generated 70 candidate statements; 58 of 112 finalized panel statements (51.8%) were AI-derived, and 85.7% were retained in the finalized set. Google NotebookLM demonstrates feasibility as a primary extraction and synthesis tool in a multidisciplinary systematic review, with extractive incompleteness as the principal limitation and substantial time savings over conventional approaches. Its novel application to RAND/UCLA consensus statement generation extends AI-assisted evidence synthesis to clinical consensus generation workflow.
Visualization design is often a demanding process that involves trying different encodings, exploring different data transformations, and refining details. While there have been important studies of these workflows, the lower-level, code-intensive practices remain underexplored. Exploratory notebook tools have allowed designers to rapidly iterate on visualizations. We use publicly-available version histories of notebooks to study how users work in these environments, observing both how they build new visualizations from existing templates or previous work and how they refine visualizations over time. We examine the interplay between data manipulation and visualization, and classify the types of changes made when updating visual encodings. We also analyze the impact of different frameworks by comparing two code-oriented libraries and a chart wizard. Finally, we examine how interactions with notebooks have changed over the years, including after the widespread availability of AI. These analyses help us understand how users iterate to produce visualizations over time using different frameworks.
This chapter presents a reproducible, cloud-based framework for integrating Python-driven data analysis into a large-enrollment undergraduate biochemistry course. Rapid growth in technologies has transformed biochemistry into a data-intensive discipline, yet most life-science students receive little formal training in computational skills. To address this gap, we developed a set of interactive Jupyter notebooks that embed introductory coding, protein structure visualization, metabolic modeling, and simple multi-omics analysis directly within the core biochemistry curriculum. The workflow is delivered entirely through the CyVerse Discovery Environment, which provides browser-based JupyterLab sessions with preconfigured Python environments, version-controlled notebooks, and standardized datasets. This infrastructure eliminates the need for local installation, supports hundreds of simultaneous users, and ensures consistent execution across diverse student hardware. The notebooks follow a scaffolded learn-apply-reflect design that links biochemical concepts with incremental coding tasks, enabling students with no prior experience to manipulate real datasets, generate visualizations, and interpret quantitative patterns. Assessment using validated pre/post surveys across more than 1000 matched responses demonstrated substantial gains in students' confidence with coding and their ability to interpret omics data. The protocol outlined here offers a scalable, equitable model for incorporating computational literacy into biochemistry education and can be readily adapted for related life-science courses.
Japan is seeing a rapid increase in the numbers of migrant workers from Asian countries amidst a rapidly aging population and declining birthrate. The notebook of personal health record (nPHR) system is a Japanese occupational health service that supports those formerly employed in hazardous work, including migrant workers. This review outlines the structure, scale, and importance of the nPHR scheme and explores Japanese medical institutions that conduct health checkups for former workers with nPHR, offering insights into the potential applicability of similar systems in migrant workers' home countries. Information on Japanese and international laws and Japanese statistical data was obtained from government websites. Studies of workers with nPHRs were reviewed using the ICHUSHI database. This review outlines the nPHR scheme established under Japanese legislation, summarizing its framework, issuance requirements, and post-retirement health checkups for 15 designated agents/works. In terms of scale, a gradual decline in newly issued nPHRs for major agents/works has been observed in recent years. The system's importance is highlighted through a review of studies involving former workers with nPHRs and an international comparison of post-retirement health checkup systems. Because of the relatively small number of patients with nPHRs, public hospitals are candidates that could be contracted as medical institutions in migrant workers' home countries. Over the past 2 decades, three additional agents/works have been included in the system, and the requirements for asbestos-related nPHRs have been expanded, indicating that the system is actively adapting to evolving occupational health needs.
In the era of data-intensive science, managing and verifying research processes that include raw data, analysis scripts, workflows, and documentation, remains a major challenge, particularly in complex fields like neuroimaging were multiple processing and analyses stages can take place. Fragmented tools, inconsistent documentation, and poor platform integration continue to undermine reproducibility. While current Electronic Lab Notebooks (ELNs) aid organization, they often lack cross-platform interoperability, authorship certification, and immutable auditability. To address these gaps, this study proposes a novel ELN framework IntegriLAB, that consolidates traditional data, document and code repositories into a centralized, web-based system for end-to-end project tracking. As a proof of concept, our case study integrates DataLad, Overleaf and GitHub. It allows real-time monitoring across the research cycle while preserving familiar workflows. A key innovation is the use of LabTrace built on green blockchain technology to certify research activities through cryptographically signed, immutable records, ensuring data integrity and verifiable authorship with minimal environmental impact. By unifying project management, secure data handling, and blockchain-based verification, the proposed ELN advances reproducibility, fosters trust, and strengthens collaborative research practices.
Obtaining, Evaluating, and Communicating Information is one of the eight Science and Engineering Practices in the National Research Council framework that underlies Next Generation Science Standards (NGSS). Scientific inquiry requires evaluation and communication of research findings. These skills are part of scientific inquiry itself. The goal of this study was to characterize the communication practices students used during Science and Engineering Fairs (SEFs). To examine these practices, we included the following question in our online anonymous and voluntary national high school SEF surveys carried out during 2021-22 and 2022-23, "What types of communication and presentation skills did you use in your science fair project?" The possible answers were literature review; research notebook; software to prepare tables, graphs, or images; written report; poster board preparation; PowerPoint presentation; and interview with the judges . Literature review and research notebook are part of developing research questions and data collection. Software to prepare tables, graphs, or images, and PowerPoint presentation contribute to analyzing research and developing a presentation. Written reports, poster board preparation, and interviews with the judges provide the opportunity to present the findings. Overall, 1789 students answered the question. The percentage of students utilizing these skills ranged from 17.3% doing a literature review to 67.6% preparing a poster board. On average, students indicated the use of three skills. Poster board preparation and interview with the judges were selected by students more than twice as frequently as literature review and research notebook. More positive SEF outcomes were associated with a greater number of skills used by students. Among the different skills, use of a research notebook showed the strongest association. In this paper, we present the findings and discuss implications regarding use of notebooks and other communication and presentation skills in students' SEF experiences and engagement with scientific inquiry.
Jupyter Notebooks have transformed the way we conduct computational research thanks to a versatile platform that combines code, annotations, plots, and a lightweight format that enables versioning. However, steps for installing dependencies and downloading data for a specific notebook may require significant time and failure to establish identical environments can compromise reproducibility of environments. Deploying a JupyterHub server not only provides a containerized environment for each notebook to ensure fast execution and exact reproducibility but also provides a computational engine that makes these applications available from any device. By providing all the steps for deploying a JupyterHub server, and providing our experience deploying and maintaining Netbooks, a server for academic investigations in network biology, we want to encourage other groups to deploy JupyterHub servers for their own computational work and to recognize their value in improving reproducibility, accelerating manuscript review, and supporting education. © 2026 The Author(s). Current Protocols published by Wiley Periodicals LLC. Basic Protocol: Setting up a JupyterHub server Support Protocol 1: Compiling a collection of Jupyter Notebooks Support Protocol 2: Securing HTTPS access Support Protocol 3: Adding parameters for JupyterHub Support Protocol 4: Data management for JupyterHub Support Protocol 5: Customizing the website homepage Support Protocol 6: Importing/Exporting conda environments Alternate Protocol: Running a JupyterHub server as a virtual machine.
We developed OpenStats, a user-friendly web application that brings the power of the R language to researchers through a high-level interface and broad support for statistical methods such as t-tests and ANOVA. OpenStats was integrated into our electronic lab notebook Chemotion ELN via its third-party API, enabling direct data exchange with it. A pivotal feature of OpenStats is its ability to record each analysis step in a structured history. This record allows users to retrace their work and enables automatic replay of the entire analysis, promoting reproducibility and long-term data integrity. In the modern research landscape, Research Data Management (RDM) tools like electronic lab notebooks (ELNs) are crucial for generating reproducible, repeatable, and transparently documented data. However, integrating statistical analysis tools into RDM systems remained a challenge. We demonstrated how OpenStats can be seamlessly linked to Research Data Management (RDM) systems, using Chemotion ELN as a reference implementation, by our application to a standard dose-response assay scenario. This linkage allows the integration of statistical data analysis in the form of a closed, traceable workflow into the RDM world. It therefore complements systems that are aiming for reproducible and standardized workflows in the scientific environment.
Structure Integration with Function, Taxonomy and Sequences (SIFTS) provides residue-level mappings between UniProt Knowledgebase sequences and Protein Data Bank structures and has historically been generated through internal Protein Data Bank in Europe (PDBe) pipelines. Here, PDBe-SIFTS is presented as a fully open-source, locally deployable implementation of this mapping framework. The pipeline combines fast, scalable sequence search using MMseqs2, an improved bounded scoring scheme for ranking candidate mappings, and residue-level mapping refinement based on backbone connectivity. PDBe-SIFTS is distributed as a Python package with command-line tools for 1) building a sequence search database, 2) identifying the best sequence-structure match, 3) one-to-one mapping at the residue level, and 4) generating SIFTS annotations in PDBx/mmCIF format. Benchmarking on the complete Protein Data Bank archive showed that MMseqs2 reduced archive-scale UniProtKB searches from hours with BLASTP to minutes, approximately 22-36 times faster, while curated mappings were recovered at top rank in 93.1% of cases. The remaining discrepancies mainly involved biologically ambiguous cases such as highly conserved proteins, chimeric constructs, or closely related orthologs. These results show that PDBe-SIFTS enables fast mapping, improving structural coherence in residue-level alignments while delivering the most up-to-date and accurate mappings, comparable to expert curation. Tool: https://github.com/PDBeurope/SIFTS Quick start notebook with example: https://github.com/PDBeurope/SIFTS/tree/master/notebooks.
Reliable predictions of protein-protein binding affinities are essential for molecular biology and therapeutic discovery. However, most computational methods rely on three-dimensional structural models, which are often unavailable for many complexes. We introduce BindPred, a structure-agnostic input framework that predicts affinities directly from amino acid sequences by combining embeddings from large protein language models with gradient boosting trees. On the protein-protein binding (PPB)-Affinity benchmark, which comprises 11 919 diverse complexes, BindPred achieves a Pearson correlation coefficient of 0.86 in random split five-fold cross-validation. Ablation analysis indicates that evolutionary embeddings alone capture most of the predictive signals, while augmenting with physics-based energy terms from PyRosetta and BindCraft increases the correlation only by 0.01. A more stringent protein-level split that places entire protein families (wild-type and all mutants) exclusively in either training or testing sets, resulting in only a modest decline in performance, demonstrating robust generalization to novel interaction pairs. Because BindPred operates exclusively on sequence input, it enables rapid inference [approximately 3 million complexes per GPU (T4) hour], making proteome-scale screening computationally feasible. The pretrained model and inference pipeline are available in a Google Colab notebook: BindPred Colab notebook. The training dataset, code, and model weights are available on the hugging face: https://huggingface.co/hbp5181/BindPred.
Parameterised quantum circuits are commonly assessed using measures such as expressibility, gradient behaviour, and entanglement. While useful, these measures do not indicate whether a circuit respects the symmetry it was designed to respect. This is especially important in equivariant quantum machine learning, where symmetry is central to the model. We introduce PsiAudit, an open-source Python toolkit for auditing symmetry-aware quantum neural network ansätze before training. Given an ansatz, a target symmetry, and a state trajectory, PsiAudit reports how the circuit occupies symmetry sectors, maintains coherence between them, fluctuates across the trajectory, and complies with the target symmetry. These outputs are combined into a configurable dashboard-style summary. PsiAudit supports phase, spin, and permutation symmetries, with the permutation audit implemented using Hamming-weight orbits. Tests on five ansatz families, across twenty random seeds and four to eight qubits, show, in the tested setting, that PsiAudit can identify inactive equivariant circuits, recover structure when multiple symmetry sectors are activated, and distinguish ansätze that appear similar under standard diagnostics. The toolkit also includes a unitary-level compliance check, reproducible examples, and a notebook for regenerating the reported results.
Deep learning methods for protein structure generation, sequence design, and structure and property prediction have created unprecedented opportunities for protein engineering and drug discovery. However, using these tools often requires navigating incompatible software environments, diverse input/output formats, and high-performance computing infrastructure, any of which may hinder adoption by primarily experimental chemical biology laboratories. Here, we present BioPipelines, an open-source Python framework that allows researchers to define multistep computational design workflows in a few lines of code. Additionally, its robust yet modular architecture provides a straightforward way to expand the tool kit with different functionalities, particularly by leveraging coding agents, with little effort. The framework currently integrates over 40 tools encompassing structure generation, sequence design, structure prediction, compound screening, and analysis. The same workflow code can be prototyped interactively in a Jupyter notebook and then submitted for production-scale runs without modification. We demonstrate applications in inverse folding, gene synthesis, de novo protein design, compound library screening, iterative binding site optimization, and fusion-protein linker optimization. We hope that this framework will empower researchers, allowing them to focus on the scientific question rather than computational logistics. BioPipelines is available under the MIT license at https://github.com/locbp-uzh/biopipelines.
The domain-linker-domain (DLD) architecture is commonly found in proteins, where flexible linkers connect consecutive domains and regulate their relative spatial positioning. Often, these linkers present partially structured elements that modulate inter-domain dynamics, directly influencing their function. From a protein design perspective, tuning the relative position and orientation of domains via the linker offers opportunities to modulate biological activity. Despite their relevance, analyzing conformational ensembles of DLD proteins remains a challenge, thereby limiting the structural insights that can be extracted. We present DL3D, a robotics-inspired visualization tool that enables intuitive analysis of the conformational space sampled by DLD proteins. DL3D discretizes the relative positions of the two domains at the linker ends and projects each conformation onto a 3D voxel map, where density is represented in grayscale to highlight the most probable configurations. In addition, quaternion-based operations allow the analysis of relative domain orientations. DL3D facilitates the structural investigation of highly flexible proteins composed of well-folded domains connected by flexible linkers. Beyond visualization, the tool supports downstream analyses such as low-dimensional conformational clustering. DL3D is implemented as a Python package and is available at: https://gitlab.laas.fr/moma/methods/analysis/dl3d/. A Jupyter notebook with usage examples is also provided.
With ever-increasing computational capabilities, robust and automated research workflows have become essential for orchestrating large numbers of interdependent simulations. However, significant technical expertise is still required to configure execution environments, define calculation inputs, interpret outputs, and manage the complexity of parallel code execution on remote machines. To address these challenges, we developed AiiDAlab, a Jupyter-based web platform powered by the AiiDA computational infrastructure that provides a framework for managing and automating computational workflows while ensuring reproducibility through full provenance tracking. Through a collection of open-source user-friendly applications, AiiDAlab enables scientists to set up, execute, and analyze complex computational workflows without interacting directly with the underlying technical details, allowing them to focus on their research questions. In this paper, we discuss how AiiDAlab has matured over the past few years, expanding beyond computational materials science and its AiiDA origins. We present recent developments toward integrating with electronic laboratory notebooks (ELNs) for FAIR-compliant data management, adoption in large-scale facilities for secure access to experimental data and analytical tools, and applications in educational settings. Together with community-driven efforts to simplify onboarding, improve access to computational resources, and support large-scale data workflows, these advancements position AiiDAlab as a powerful platform for accelerating scientific discovery and fostering collaboration across disciplines.
Online data from social media platforms provide an observational window into a wide range of psychological phenomena and real-world social interactions. Accessing these data typically requires the use of application programming interfaces (APIs), which are designed to ensure compliance with the legal and ethical frameworks governing the access, processing, and storage of potentially identifiable information. However, implementing API-based data collection within a dedicated program remains a major barrier for many researchers. This article addresses this challenge by introducing a ready-to-use data collection tool based on the official YouTube Data API (v3) and providing step-by-step guidance for its use. Geared toward researchers and practitioners with little or no prior programming experience, the tutorial enables users to implement transparent and reproducible API-based data collection procedures in their research. The tool and a fully documented interactive Python notebook are freely available via a public GitHub repository.
High-resolution mass spectrometry is a powerful tool for untargeted analysis. However, in-source fragmentation (ISF) could lead to the misidentification of compounds in untargeted metabolomics or exposomics studies. To prevent misidentification and to strengthen compound identification through MS/MS spectral library matching, we developed IMFrag, a Jupyter notebook-based tool that utilizes structural information gained from ion mobility-mass spectrometry (IM-MS) as an orthogonal technique to differentiate independent precursor ions from fragments formed via ISF. We first examined l-tryptophan, which is an essential amino acid that undergoes extensive fragmentation during electrospray ionization (ESI). IM-enabled data-independent acquisition (IM-DIA) analysis revealed distinct mobility signatures for identical fragment ions formed at different instrument sites, enabling discrimination between ISFs generated prior to the IM drift tube and fragments produced via postmobility collision-induced dissociation. Similar patterns were observed for a structurally diverse collection of small molecules. Additional structural information could also be inferred from the IM-DIA workflow, such as unique dimers, protonation sites, and distinct ion types that were not apparent from LC-MS alone. These insights were shown to be useful when applied to healthy human plasma samples, which served as a more complex and biologically relevant matrix that contained ambiguities, such as coeluting, isobaric candidate structures. Thus, IMFrag was developed as an accessible framework for interrogating MS1 and post-IM MS2 chemical features in untargeted data sets and can be integrated into untargeted analysis pipelines or used to support the development of ISF-derived MS/MS spectral libraries.
An exhaustive database of 8,392 neutral benzenoids, BenzDB, containing up to nine hexagons (all-carbon six-membered rings, 6MR) is presented. This database contains data on structural properties (xyz-optimized geometries, composition, symmetries, irregularity parameters), graph-theoretical derived properties (number of Kekulé structures, Clar cover), vibrational spectra, and magnetic properties (NICS(0) values, IMS maps at 1 Angström) computed at the DFT level. Its main features are illustrated and compared to the largest existing PAH databases in the literature. BenzDB is accessible and easy to handle via the BenzAI software or a companion Jupyter notebook or can be queried directly via a REST API at https://benzenoids.lis-lab.fr/.
The incidence of maternal deaths from preventable pregnancy-related conditions remains alarmingly high at 303,000 annually, with over 800 women dying daily from avoidable causes. Ethiopia is one of eight sub-Saharan African countries that are identified as global hot spots for maternal mortality. Thus, this study aimed to model predictors of incomplete ANC utilization among reproductive-aged women in Ethiopia using explainable machine learning algorithms. This study employed the 2019 Ethiopian Mini Demographic and Health Survey (EMDHS) dataset. Data preparation techniques such as feature engineering, data splitting, handling missing values, resolving imbalanced categories, and outlier removal were used to clean the data. Six popular machine learning classifiers were implemented in R 4.4.2 and Python 3.11.5 via Jupyter Notebook through the Scikit-learn and XGBoost packages and evaluated using multiple permanence matrices. Finally, Shapley Additive exPlanations (SHAP) analysis was used to clarify the impact of the most important predictors on the model's output. This study included 3979 women who had given birth during the five years prior to the survey out of the 8,885 interviewed women. Random forest (RF) was found to be the best model for modeling predictors of incomplete ANC utilization in Ethiopia, with 73% accuracy and 79% area under the ROC curve. Older age (25 and 34), residence area, being in the Benishangul-Gumuz, Tigray region, Harari region and wealth indices were top predictors of incomplete ANC utilization among reproductive-age women in Ethiopia. This study found that young women in rural areas, having low-income indices and low levels of education, as well as those living in the Somali and Harari regions, are more likely to experience incomplete ANC utilization. Policymakers and stakeholders should prioritize these vulnerable groups when designing policies and maternal health services to improve ANC utilization and reduce maternal mortality in Ethiopia.
Intrabody communication (IBC) channels offer physiological diversity that may support future wearable biometric identification. Recent reports of over 99 per cent identification accuracy have frequently resulted from data leakage, where samples from the same subject are seen in both training and evaluation, yielding inflated and unreliable metrics. In this work, we establish a public, leakage-free benchmark for IBC biometrics built on a 30-subject open dataset, using strict subject-wise 80/20 splits repeated five times to ensure reproducibility. We systematically compare frequency-domain and time-frequency representations, including resampled spectra, discrete wavelet transform (DWT) statistics, and their fusion. Under the subject-wise embedded-friendly benchmark, the strongest classical configuration, Scattering + LightGBM, reaches 54.0 per cent accuracy, while db4-DWT and lifting-based wavelet statistics with Random Forest improve over the Simple-3 baseline (49.3 and 51.6 per cent versus 39.0 per cent). Separately, closed-set neural analyses provide exploratory upper bounds rather than leakage-free subject-wise results: a Raw MLP reaches 83.7 per cent accuracy, whereas adding DWT statistics does not improve this result (81.2 per cent for Combined MLP), and SpectralCNN reaches 74 per cent. Confusion matrix analysis reveals that residual errors are concentrated among subject pairs with statistically overlapping signatures, suggesting the presence of intrinsically hard users and a potential biometric ceiling for this modality. Embedded profiling on an STM32F446RE Cortex-M4 microcontroller indicates that lifting-based wavelet features enable low-latency, low-energy scoring, requiring approximately 0.55 ms and 18 micro-J per 256-point spectrum for Lift-bior feature extraction plus Random Forest inference (versus approx. 33 micro-J for the equivalent db4-DWT pipeline). All code, data split scripts, and Jupyter notebooks are released open source to facilitate reproducibility and enable rigorous future comparisons.