Database

Bacterial pathogens ensure their successful dissemination by employing specialized secretion systems to deliver a wide range of virulence proteins that interact with specific host proteins to modulate the response during infection. These interactions are often mediated by host-like elements, like short linear motifs, which can interfere with host interaction interfaces, enabling bacterial secreted proteins to compete and displace native partners of the targeted proteins. Several resources provide access to bacteria-host protein-protein interaction data. However, some are neither fully functional or up to date. Moreover, most of these databases lack virulence protein annotations, thus hampering hypothesis formulation based on protein interaction analyses. In this context, we have developed BactMentha, a unified resource that integrates bacterial-host protein-protein interactions for three organisms (human, mouse, and rat) with virulence protein annotations and bacterial strain clinical relevance information. BactMentha provides easy access to available experimentally identified binding regions of the interaction partners, complemented with putative interaction interfaces based on predicted structure complexes and on the detection of host-like elements on bacterial protein sequences interacting with human proteins. BactMentha currently contains 11 776 unique protein interactions between 2970 bacterial proteins, of which 684 have at least one virulence annotation, and 5088 host proteins. Experimentally identified binding regions are available for 532 bacteria-host interactions, most of which involve human proteins (440 interactions). Predicted interaction interfaces enabled a 3.7-fold increase in known and putative binding regions in the bacteria-human protein interaction data. By providing easy access to functionally annotated and clinically relevant protein interaction data, enriched with experimentally and predicted interaction interfaces, BactMentha can be instrumental in formulating novel hypotheses and speeding up wet lab research on the molecular mechanisms underlying pathogenic bacterial infections.

Biomedical knowledge graphs can connect information across genes, phenotypes, tissues, pathways, experiments, and clinical resources, but they are difficult to query correctly without detailed knowledge of the graph. Large language models can help write Cypher code for knowledge graphs, yet a query focused on a bioinformatics task that looks reasonable may still use the wrong identifier, relationship direction, source, intermediate node, or output unit if the user is not completely trained on the schema. We developed ddkg.skill, an Agent Skill that can be loaded into a compatible LLM session to support querying of the NIH Common Fund Data Ecosystem Data Distillery Knowledge Graph (DDKG), a biomedical property graph integrating more than 180 ontologies, over 40 genomics datasets, and dozens of additional cross-domain biomedical datasets. Rather than utilizing a simple markdown prompt, ddkg.skill contains a controller, a compositional library of DDKG-specific references and structured tables, primary documentation, validated query patterns, a routing table, and a Python script that checks the internal links among these materials. The skill build for the December 2025 DDKG release contains 38 bundled files and 239 routing relationships and is identified by checksum so users can state the exact build used to compose a query. The evaluation reported here was performed separately on an earlier build with 38 files and 227 routing entries. In that orthogonal nine-test evaluation against a live December 2025 DDKG instance, five of seven tests targeting sources absent from the worked examples produced correct executed results. The evaluation also exposed a failed query, a cross-source comparison that was not biologically well posed, and errors in the skills own reference material that informed later revisions. The skill guides an AI through entity resolution, graph inspection, query construction, and validation for biomedical and bioinformatics queries, while keeping the DDKG itself as the source of returned results. This design provides a portable and versioned method for giving general-purpose AI systems practical knowledge of a complex biomedical knowledge graph to empower complex bioinformatics data integration. Three executed biomedical use cases further demonstrate cross-species phenotype-to-expression querying, genomic overlap between 4DN chromatin-loop anchors and GTEx eQTLs, and integration of congenital heart phenotype associations with ClinVar and GTEx heart evidence.

We combine GeneNetwork's 1TB structured database with its metadata as an RDF graph and an associated literature corpus. We benchmark frontier and local LLMs querying it through four architectures: document, graph and agentic retrieval-augmented generation (RAG), and a combination of the three. We evaluate them on 40 domain specific expert-curated questions with a new metric: the satisfaction frequency psi (psi, i.e, the share of questions answered with precision and recall of at least 70% or an f1 of at least 60%). psi ranks systems the same way precision, recall and f1 -- a machine learning metric built from the harmonic mean of precision and recall -- but is easier to read and assess for humans. System architecture had a large effect: with full use of Claude Opus 4.8 the plain document RAG of our earlier GNQA system had a psi of 36% against 86% for our new combined tools. Delegating most of the work from Claude Opus 4.8 to a local model halved query costs while still reaching a psi of 59%.

When an ontology-based biomedical knowledge graph (a KG, containing information about genes, pathways, phenotypes, and molecular interactions) is analyzed using a computational reasoning system, the system can return unwanted generic concepts. Such generic concepts, in the context of a query about potential drug treatments for a disease, might include 'pharmaceutical preparations' or 'antibiotics'. Answers can be filtered based on hand-curated block-lists or ontology Information Content (IC), but to our knowledge, (1) neither filtering method has been validated against expert judgments of generality, and (2) IC is not available for the majority of nodes in the representative KG used in this work. In this study, three domain experts rated representative lists of KG concepts for generality. We built a semantic precision score (SPS) from a concept's name and description alone, using a biomedical language model (BioLORD). For the purpose of filtering generic answers from biomedical query results, we compared (using queries for 200 diseases) the SPS against a curated concept block-list and two IC thresholds. IC agreed moderately with the experts (Spearman's {rho} = 0.59), but correlations were negligible or negative within IC bands. The SPS was well-correlated with the experts ({rho} = 0.76) globally and within SPS bands, and remained informative on concepts without IC ({rho} = 0.59). Hard filtering of query results using the SPS removed 1,502 of 33,161 query answers, nearly three times the block-list's 513, while removing fewer of the query-specific DrugCentral indications. Hard filtering by IC removed either few answers or many approved drugs. We additionally compared the filtering methods' performance on an independent evaluation set of queries for 200 diseases. The SPS filter removed 5,135 of 91,203 answers compared with the block-list's 1,419 and lost none of 1,296 indications compared with 17. Further, SPS filtering removed approved drugs no more often from the half of DrugCentral not used in training than from the half used in training. Semantic meaning captures expert judgments of specificity more faithfully than ontology position, and semantic scoring can replace IC and manually maintained block-lists for filtering drug-repurposing results.

There are many databases of post-translational modifications, but each has different protein and PTM coverage, making the current state of post-translational modifications difficult to obtain, assess, or harmonize. ProteomeScout has been a relational database that harmonized a myriad of resources to address this challenge. In this update, we wished to create a lightweight, sustainable open source resource for the community that retained the power of harmonization, protein visualization, and Python tools for interacting with the database. ProteomeScout-v4 presents an ecosystem of open-source tools for aggregating and updating proteins and post-translational modifications and presenting the web-based and code-based access tools for interacting with the data. ProteomeScout has moved from using a complex relational database to a flat-text file format. The database file update is version controlled on FigShare and all features have processes for continuous updating, including migrating PTMs to updated protein records, when external databases are updated. The open-source nature and flexibility allow for user-based control and community-based feature integration. Fast and easy analysis is possible with the API, with example analyses provided here.

Agentic artificial intelligence (AAI) is increasingly used by biomedical scientists, where it has substantially lowered the difficulty of integrating computational, statistical and data-science approaches into day-to-day research activities, assisting with project design, analysis and manuscript preparation. Significant barriers remain, however, which are unlikely to be removed solely by releasing more capable AI models. Meaningful use still demands a technical fluency that many biomedical researchers lack: a PI and their lab members must learn how to use the AAI system, establish standard operating procedures (SOPs) for managing, sharing and curating data, and enforce rules that address data privacy and security. Existing commercial AAI systems, which were designed primarily for software engineering purposes, need to be significantly re-purposed for biomedical-related data science. We present Murmurent, shared software that sits beneath the agentic AI and provides the skills, procedures and protections to address the concerns above, removing the need for a lab to develop them independently. This includes (1) infrastructure to support projects and so-called choreographies} involving multiple group members; (2) a set of specialized agents each dedicated to typical biomedical data science tasks including software and statistical development, data visualization, literature search, equity, diversity, inclusion, and decolonization (EDID) review, and others; (3) a tiered memory which can retain important research decisions, intermediate, derived outputs and sensitive data, and which can exploit this information to automatically build better contexts in AAI sessions; (4) traceability records which are used by the AAI session to better plan and execute data analyses and software builds by, for example, helping to avoid repeating decisions that lead to dead-ends; (5) enforcement of SOPs for data maintenance and data governance across all lab members; and (6) multi-user capacity to allow groups to interact and collaborate. We use the system to identify putative inhibitors of Peptidyl-prolyl cis-trans Isomerase NIMA-interacting 1 (Pin1), a protein with a shallow catalytic site making it difficult to target. We describe the construction of several approaches to identify Pin1 inhibitors and the results they yield. Murmurent is open source.

BackgroundClinicians increasingly use AI systems to search the medical literature, but current benchmarks do not jointly test whether a response identifies the originating paper and its supporting passage. LitBench evaluates four dimensions: stated versus interpretation-requiring facts, the number and composition of competing papers, one-versus two-paper evidence, and refusal when evidence is absent. MethodsLitBench comprises 1,980 open-access papers (1,000 epilepsy papers holding the answers, 980 decoys from unrelated fields) and 9,472 human-reviewed facts, giving 2,188 single-paper questions and 170 requiring a fact from each of two papers. Difficulty rose by burying the answer among up to 2,000 distractors, then again over live PubMed Central. The same questions were then asked with the answering paper removed, so that refusal was the correct response. Four systems were tested: Gemma-4B, Gemma-12B, Sonnet-5, and DeepSeek-V4-Flash (refusal conditions only). Three model judges scored each answer by majority; two-paper questions counted only when both facts were found. ResultsAcross fixed-corpus single-paper conditions, accuracy ranged from 55.9% to 71.9% for Gemma-4B, 61.2% to 75.5% for Gemma-12B, and 90.0% to 91.3% for Sonnet-5. Similar epilepsy papers were harder than mixed candidate sets for both Gemma configurations. Two-paper accuracy was 5.7%, 17.1%, and 37.8%, respectively. With evidence absent, Gemma-12B refusal fell from 88.3% to 39.9% as candidate sets grew; Gemma-4B almost never refused. Sonnet-5 refused on 94.9% of sampled single-paper and all sampled two-paper cells. DeepSeek-V4-Flash also refused frequently, but on 80.3% of matched answer-present controls. ConclusionsNo system was reliable across every axis tested: accuracy fell as competing papers were added, interpretation-requiring facts were harder than stated ones, two-paper synthesis was rarely achieved, and only some systems recognized absent evidence. What counts as "successful" literature search carries considerable nuance; LitBench can evaluate proposed tools at a fine-grained level.

High-throughput transcriptomics can be used to develop Connectivity Mapping resources for identifying potential reversers and mimickers of disease signatures for drug and target discovery. Recently, several new Connectivity Mapping datasets have emerged, but no efforts were made so far to evaluate and integrate them. Motivated by this opportunity, we have developed Perturb-Seqr, a comprehensive Connectivity Mapping search engine that integrates nine small molecule transcriptomics Connectivity Maps and seven single gene perturbation followed by expression Connectivity Maps. Specifically, Perturb-Seqr contains processed data from LINCS (L1000), Tahoe-100M, Novartis Institute for Biomedical Research (DRUG-seq), the original CMap (cDNA microarrays), Ginkgo Bioworks (DRUG-seq), SciPlex (scRNA-seq), DeepCover MoA (PISA proteomics), Perturb Atlas (RNA-seq), Replogle et al. (Perturb-seq), CREEDS (cDNA microarrays), CM4AI (Perturb-seq), and RummaGEO (RNA-seq). To serve these Connectivity Maps in one unifying search engine, we converted the expression signatures from these resources into 450,271 gene sets of up- and down-regulated genes measuring the effects of 9,856 unique drug and 17,643 unique gene perturbations, targeting 917 cell lines and cell types. We demonstrate how Perturb-Seqr, and the gene sets it serves, can be used to identify targets for drugs and drugs for targets, uncover new pathways, and identify potential side effects for preclinical compounds. Overall, Perturb-Seqr is a comprehensive up-to-date Connectivity Mapping search engine and database, enabling the discovery of new drugs and targets. Perturb-Seqr is accessible from: https://perturbseqr.maayanlab.cloud/.

Biomedical artificial intelligence (AI) increasingly relies on heterogeneous data sources, yet differences in data structure, terminology, evidence quality, and provenance complicate their integration into reliable knowledge resources. Here, we develop an evidence-aware framework for processing and integrating heterogeneous pharmacovigilance data from FDA Adverse Event Reporting System (FAERS) reports, biomedical literature, clinical trials, and electronic health records in Mass General Brigham. The framework combines large language model (LLM)-based information extraction, ontology-driven semantic normalization, cross-source evidence assessment, statistical and machine learning-based prioritization, and ingredient-based adverse drug event (ADE) relationship inference to transform heterogeneous evidence into a structured and evidence-based knowledge resource. The constructed knowledge system contains 11,588 drug nodes, 44,843 ingredient nodes, 3,275 ADE nodes, 993 indication nodes, and more than 2.47 million relationships, with each drug-ADE relationship assigned a confidence score. We further developed a drug-level web interface for accessing the knowledge resource and validated its utility in downstream AI tasks. Specifically, incorporating the developed knowledge resource improved the comprehensiveness, relevance, and overall quality of LLM-based drug-specific ADE synthesis and improved the precision and F1 score of clinical ADE relation extraction. These results demonstrate that AI-assisted processing combined with evidence-aware curation can transform heterogeneous biomedical evidence into reusable, AI-ready knowledge with explicit provenance and evidence strength information. The framework provides a general strategy for integrating heterogeneous biomedical data to support reproducible knowledge construction and downstream biomedical AI applications.

Ensuring the safety and suitability of proteins considered for food and biotechnology applications is an important challenge. Computational approaches can facilitate the prioritisation of candidate proteins for further investigation, particularly given the rapid growth of genomic and metagenomic sequence datasets. In this study, we developed computational models to distinguish edible plant tissue-associated (EPTA) proteins from non-EPTA proteins. The positive class comprised 939 proteins with experimentally validated evidence of expression in plant tissues commonly consumed by humans, hereafter referred to as EPTA proteins. The negative class comprised 939 proteins annotated in UniProt as toxic, allergenic, or antinutritional, hereafter referred to as non-EPTA proteins. Models were evaluated using five-fold cross-validation with a cluster-based partitioning strategy, ensuring that no protein in the validation set shared more than 40% sequence identity with proteins in the corresponding training set. First, composition-based sequence features were evaluated to establish a baseline for EPTA or non-EPTA classification, with the Extra Trees (ET) classifier achieving a validation AUROC of 0.848. To incorporate evolutionary information beyond amino acid composition, PSSM profiles were subsequently evaluated, yielding an AUROC of 0.842 with an SVM classifier. Finally, to capture richer sequence patterns and contextual information using protein language modelling, we evaluated ESM2-based models. The fine-tuned ESM2-T33 model achieved the highest validation AUROC of 0.925 and was selected as the final predictive model. The resulting approach was implemented as the EdiProPred web server (https://webs.iiitd.edu.in/raghava/edipropred/) and standalone software for computational classification of plant proteins. The EdiProPred framework provides a useful computational tool for prioritising candidate plant proteins and facilitating the exploration of novel protein resources for food and biotechnology applications.

Genetic association analyses of exome sequencing data pose distinct methodological challenges in small cohorts, which are common in rare disease research. Data filtering prior to association testing requires particular care, especially when analyzing datasets generated under different study designs. Most of current genetic association tools need to be integrated into complex data processing pipelines, whose implementation requires technical expertise and is prone to error. To alleviate mentioned issues we built a comprehensive, high performance Nextflow workflow for genetic association testing, which includes all commonly performed data filtering and preparation steps, controlled by a rich set of modifiable parameters. We compared our software against existing tools incorporating quality control and aggregated association testing steps and found our solution to offer the most complete suite of data preparation, association testing and reporting modules, making small-cohort association analyses accessible to teams with limited technical expertise.

Background. High-throughput drug screening (HTS) is increasingly used in precision oncology across cell lines, patient-derived models such as organoids (PDO) and xenograft organoids (PDxO), and ex vivo patient samples, with multiple active clinical trials incorporating HTS for individualizing treatment. Yet HTS efforts predominantly rely on ad hoc, spreadsheet-based workflows, with no standardized infrastructure for storing, analyzing, and comparing screens across patients, cohorts, or institutions, limiting reproducibility and slowing clinical data exchange. Existing tools address isolated pieces of the workflow, handling either data management or analysis and visualization, but none integrate standardized storage, reproducible dose-response analysis, individualized drug prioritization, and clinician and molecular tumor board friendly reporting in a single open-source framework. Results. We present DS5, an open-source Python framework that operates on an HDF5-based DS5 file format that unifies standardized storage, quality control, dose-response analysis, cohort-level drug prioritization, and automated reporting in a single lightweight package. DS5 enforces raw-data immutability, integrates RxNorm-based drug name standardization, and scales from small clinical cohorts to large pharmacogenomics datasets on a standard laptop. We validated DS5 against two public benchmarks spanning over 428,000 drug, cell line pairs, recovering published LN IC50 values from GDSC2 (Pearson r = 0.973) and Emax values from CTRP (r = 0.961). Using a published breast cancer PDxO dataset (45 compounds by 16 models), we reproduced GR-based profiles via a custom-metric extension and used DS5's cohort-level prioritization to predict in vivo tumor response with ROC AUC = 0.907 across 30 drug PDX pairs. Finally, we demonstrate the utilities of DS5 in an ongoing pediatric brain tumor precision oncology initiative, where its cohort-normalized drug rankings and automated reports directly support molecular tumor board discussions. Conclusion. DS5 provides the cancer basic and translational research community with a standardized file format and reproducible computational analysis methods to support functional drug screening experiments; and its reporting features facilitate bridging raw drug screening measurements to translational decisions. DS5 is released open-source under the permissible MIT license.

PURPOSE: To demonstrate an end-to-end research workflow combining a large electronic health record (EHR) resource and a multi-agent artificial intelligence (AI) system governed by Human-Guided Orchestration (HGO), using sequential sacituzumab govitecan (SG) and trastuzumab deruxtecan (T-DXd) orders in breast cancer as the clinical use case. METHODS: A June 2026 EHR release contained 4.95 million patients, including 156,051 with breast cancer codes. Muse Spark supported cohort construction, executable statistical analysis, data-quality checks, literature retrieval, and reporting executed through Symbiotic Agent Teaming (SAT) under investigator direction. The main clinical measure was the second-drug record span, from first to last order. Analyses were descriptive and unadjusted; orders did not confirm administration or progression. RESULTS: The workflow identified 106 patients: 73 had T-DXd ordered first and 33 had SG ordered first. Median second-drug spans were 63 days (interquartile range [IQR], 29-116) for SG after T-DXd and 49 days (IQR, 21-126) for T-DXd after SG; the median difference was -14 days (95% bootstrap CI, -41 to 21; distributional P = 0.42). Reported outputs additionally included cohort definitions, aggregate analyses, evidence context, figures, and a preliminary draft manuscript. CONCLUSION: This study provides a concrete model for integrating large EHR resources, agentic AI, and human expertise across the research cycle. The ADC use case illustrates how computational execution and Expert-in-the-Loop (EITL) adjudication can produce interpretable research outputs and refine their scientific meaning, providing a foundation for extending and evaluating this approach across clinical questions.

Open genomic repositories have become critical infrastructure for pathogen surveillance, variant tracking, outbreak reconstruction, and public health decision making. However, many of these repositories were designed to support rapid scientific data sharing rather than adversarial verification, creating potential security gaps. This paper introduces Advanced Persistent Biological Threats (APBTs), as a framework for understanding how malicious cyber biological actors can gradually disrupt genomic surveillance without entering a laboratory or releasing a biological agent. Instead of altering biological material, an attacker targets the sequence data, metadata, reference datasets, and analytical models used by surveillance systems. We present, to our knowledge, the first integrated multilayer taxonomy of APBTs, structured around a four phase lifecycle, and investigate whether the metadata weaknesses identified by this framework are present in an operational genomic repository. To assess this, we audited 100,000 SARS-CoV-2 BioSample records obtained from the NCBI. A framework of 23 checks was applied across five metadata layers: temporal, geographic, host and specimen, provenance, and technical metadata. Every record triggered at least one APBT relevant metadata vulnerability indicator, with a mean of 7.34 indicators per record (SD = 1.34). Overall, 97.3% of records were classified in the High or Critical severity categories. Every record also contained at least one indicator in both the provenance and technical metadata layers. These results largely reflect the optional status of several fields in the current BioSample submission model rather than isolated errors by individual data contributors. Temporal and geographic information was comparatively complete, while fields needed for provenance reconstruction, chain of custody assessment, technical validation, and analytical interpretation were frequently missing. These findings do not indicate deliberate manipulation. However, the structural incompleteness of BioSample metadata creates vulnerabilities that an adversary can exploit to introduce poisoned, misleading, or weakly traceable metadata into genomic repositories. As a result, maliciously altered metadata can appear plausible, remain undetected, and influence downstream genomic analysis and surveillance. The findings therefore show that widely used pathogen genomic repositories already contain structural weaknesses that can support metadata poisoning as part of an APBT operation.

Panvascular intervention research requires evidence workflows that preserve source identity, outcome definitions and observation windows. We developed PanVasc Research, an executable research framework, and evaluated a fixed local Qwen3-4B model using two complementary tasks. Fifty ClinicalTrials.gov records from five vascular query strata generated 200 source-fidelity tests with intact evidence or controlled removal of the requested source, primary outcome or timeframe, plus 50 clean controls. A separate 100-statement sample from the official NLI4CT test set assessed clinical-trial entailment and evidence selection against existing expert labels. Generic and checklist prompts used identical evidence and maximum generation budgets; each output was also evaluated with an input-only deterministic contract. Registry exact accuracy was 51/200 (25.5%) with the generic prompt and 77/200 (38.5%) with the checklist; intact-case agreement was 50/50 and 48/50. The rule baseline recovered 200/200 tuples. NLI label accuracy was 48/100 (48.0%) and 50/100 (50.0%), respectively. The source contract retained 35 and 31 incorrect NLI labels in the two arms. Registry references establish fidelity to a registration snapshot, while NLI4CT concerns breast-cancer trials and does not validate vascular expertise. The framework also retains provenance-recorded literature retrieval, structured research planning and local numerical analysis. These experiments support a bounded assessment of source handling and semantic failure, rather than a new foundation model or autonomous scientific discovery. A proposed endpoint representation identifies the additional domain annotation and independent validation required for a panvascular research model.

Precise identification depends on highly resolved classification. Therefore, whole genome-based strain typing schemes have been implemented for identification of the most important microbial human pathogens. However, there are no dedicated typing schemes for the majority of prokaryotic human, animal, and plant pathogens, or for non-pathogenic microbes, often limiting their identification to taxonomic classification tools that have the species as the smallest unit. To address this deficiency, we created the genomeRxiv web server which contains over 350,000 genomes and leverages the average nucleotide identity (ANI)-based Life Identification Number (LIN) approach, allowing users to identify prokaryotic genomes in real time from approximately genus-rank to near strain-level resolution. LINs are overlaid with circumscriptions of species and infraspecies classes, integrating taxonomy with strain typing. Metadata and distance trees are visualized to complement inferred phylogenetic relationships with epidemiological and ecological information. A command line tool is provided for users to upload k-mer sketches of unpublished genomes and associated metadata, facilitating epidemiological investigations without the need to share complete DNA sequences. In conclusion, genomeRxiv constitutes a much-needed resource for precise, rapid identification of the many prokaryotes not included in any pathogen-specific typing scheme.

Several tools are available to detect differential gene expression. The predominant approach of different statistical methods is to consider changes in gene expression independent from their genomic context. Previously, DELocal was introduced that takes into consideration gene expression dynamics in the genomic neighbourhood. Whereas DELocal was shown to identify developmentally regulated genes in developing mouse tooth, its performance in other biological conditions remains undetermined. To benchmark DElocal together with other methods, we develop another tool 'broadSeq'. It provides easy to use interface for methods like DELocal , deseq2, limma, edger, EBSeq and many more. Here we report benchmarking results on different data sets that comprise wild type developing mouse organs from different embryonic time points, and also human cancer data. The examined datasets are from the heart, kidney, liver, brain, forebrain, and hindbrain. We found that most methods produce relatively comparable lists of differentially expressed genes. Nevertheless, DELocal identifies partially distinct sets of genes compared to the other methods. Evaluating the identified genes using gene and disease ontology terms show that the uniquely detected genes by DELocal are biologically relevant and also highly expressed compared to the genes identified by the other methods. Because consistently highly expressed genes show relatively subtle fold changes, the neighbourhood concept of DELocal appears to correct for the difficulty in detecting these differential expression of highly expressed genes. Finally, we provide the tools as combined bioconductor packages, allowing researchers to explore and decide the best methods for their data and questions.

Tabular synthetic data generation (SDG) can facilitate research and development in the health domain, as access to health data is restricted. To accommodate for its structural complexity, such as multimodal distributions and fragmentation, transformer models provides an alternative to conventional SDG methods since they are better equipped to infer global spatial coherence due to textual context of tabular records. However, methods for optimising their adaption to tabular SDG are still largely unexplored. We contribute to the understanding of transformer-based SDG by evaluating fidelity and privacy of three prompting regimes, zero shot, one shot and few shot training, for tabular health SDG using small transformer models. We evaluated llama 3.2 (1B and 3B parameters) and Phi (1.3B and 2.7B parameters), on a real clinical dataset. Our results showed that zero shot learning was less prone to hallucinations and performed overall better for fidelity, but at the loss of privacy in comparison to the other regimes. Our results showed no differences between model sizes and families. Future studies should extend the analysis to other models and datasets, and explore strategies to reduce hallucinations.

Public data sharing in immunological and clinical research is essential for transparency, reproducibility, and secondary analysis. However, many datasets in repositories such as ImmPort lack sufficient structured data to enable reuse. ImmPort's Assessments and Lab Tests templates -- which capture critical clinical information -- are often incomplete or omitted due to their complexity and the manual effort required. To address this challenge, an application with a graphical user interface requiring no programming was developed to facilitate population of these templates. The tool streamlines the submission process by reducing time and complexity for data contributors while ensuring template formatting. It integrates ImmPort files, study-specific data files, and a curated data dictionary to automate the mapping of study variables to the required fields within each template. The tool was evaluated on seven ImmPort studies with diverse designs. It successfully generated submission-ready templates that passed ImmPort's internal validation and will be incorporated into the repository. By improving the feasibility and completeness of templates, this approach enhances the quality, findability, and reusability of data within ImmPort.

Reliable extraction of trial characteristics and outcomes is essential for systematic reviews, yet incorrect values can alter evidence synthesis and are difficult to detect without clear source evidence. We evaluate EviSearch, a clinical evidence-curation system that links extracted values to source passages, independently verifies cited evidence, and prioritizes higher-risk cells for review. Using ten trial publications, a 133-column schema, reference annotations, and consistent scoring across model families, we assess answer quality, inter-reading agreement, evidence reproduction, targeted review, and evidence modality. In the Qwen evaluation, EviSearch achieved 91.70% accuracy, cited 100.0% of admitted values, and reproduced 98.9% independently. Reviewing 15.6% of cells increased accuracy to 95.22%.