Our Work

This report describes the workshop "Mapping cross-project outreach and user journeys for the BFSP community", held during the ELIXIR All Hands Meeting 2026. The workshop brought together contributors …

This report describes the workshop "Mapping cross-project outreach and user journeys for the BFSP community", held during the ELIXIR All Hands Meeting 2026. The workshop brought together contributors from across the ten BFSP 2024-26 programme work packages. It addressed two objectives: (1) evaluating the outputs of the four concluding work packages (WP2–5: E-PAN, FAIRyMAGs, HARVEST, Odyssey) from both researcher and developer perspectives; and (2) mapping user journeys and identifying technical and training gaps connecting these outputs to the active work packages (WP6–10).In Part 1, the depth and detail of the presentations exceeded the original schedule, occupying approximately 75 minutes. On behalf of E-PAN, Pedro Barros presented completed pan-genome standards and a published consortium white paper; On behalf of FAIRyMAGs, Bérénice Batut described FAIR Galaxy workflows, benchmarking infrastructure and new GTN training materials; On behalf of HARVEST, Sebastian Beier reported on RDMKit updates, submitted manuscripts, and an active FAIRification pipeline; and Odyssey was presented by Evangelos Pafilis (GR-HCMR) via Zoom, substituting for Gareth Gillard, covering the completed R package, training lesson, and post-project sustainability plans.Part 2 comprised approximately 15 minutes of onsite group work in four thematic areas (Plant Sciences, Metagenomics, Biodiversity, Pathogens). The key cross-cutting finding across all groups: researchers lack direct motivation for FAIRness even when infrastructure exists, and “FAIR-by-design” service approaches were identified as the most practical structural response. Specific community inputs were captured for WP7 (controlled-access pathogen data and PI-level training), WP8 (MetaBuddy for AI-assisted experimental design annotation), and WP10 (KYBELE entity prioritisation for Odyssey). Findings will inform the 2027–28 BFSP work plan.

• Small to quite large mammals with rounded head and ratherflat face, facial whiskers, and large eyes and ears; sleek and streamlined body with muscular legs. • 48.6399 cm. • Holarctic, Neotropical, A…

• Small to quite large mammals with rounded head and ratherflat face, facial whiskers, and large eyes and ears; sleek and streamlined body with muscular legs. • 48.6399 cm. • Holarctic, Neotropical, Afrotropical, and Oriental regions. • From desert through forest to mountain areas, from cold temperate zone to tropics. • 14 genera, 37 species, at least 228 extant taxa. • 1 species Critically Endangered, 6 species Endangered, 9 species Vulnerable; 5 subspecies Extinct since 1600.

Legal regulations, financial criticality, and low trust in the evolving use of AI in engineering project management make Explainable AI (XAI) essential. Current research in XAI is focused on feature-a…

Legal regulations, financial criticality, and low trust in the evolving use of AI in engineering project management make Explainable AI (XAI) essential. Current research in XAI is focused on feature-attribution tools like SHapley Additive exPlanations (SHAP) and Local Interpretable Model-agnostic Explanations (LIME). Their outputs align with the expectations of data scientists but are not readily understandable to most engineering project managers, who work with Gantt charts and earned value scores. This paper introduces the Tri-Modal Explanation Architecture (TMEA) that generates three coordinated outputs from a single decision-tree surrogate of a Proximal Policy Optimization (PPO) control policy: a Gantt schedule overlay, a KPI partial-derivative panel, and a Structured Decision Rationale (SDR). Temporal, magnitude, and causal alignment between the three outputs is enforced by a Cross-Modal Consistency Mechanism (CMCM); explanations that fail alignment are withheld rather than released. A simulation environment built from an empirical project database (187 real projects) produces scenarios that are statistically indistinguishable from real earned-value trajectories, confirmed by K-S tests (D ≤ 0.032, p ≥ 0.419). Across 1,400 bootstrapped scenarios and a comparison against GPT-4o with and without retrieval augmentation (RAG), the Tri-Modal configuration delivers higher surrogate fidelity and cross-modal alignment than any single- or bi-modal alternative.

Standalone build of Hamelin 0.2.1 for Linux x86-64. CPU-only version: it does not use the GPU, even if the computer has one. No Python required. What's new since 0.2.0 Updated the authors list (About…

Standalone build of Hamelin 0.2.1 for Linux x86-64. CPU-only version: it does not use the GPU, even if the computer has one. No Python required. What's new since 0.2.0 Updated the authors list (About section of Settings, README, citation file and screenshots). Training on a GPU that also drives the screen no longer fails every trial. This fix applies when running Hamelin from source with a CUDA-enabled PyTorch (uv sync); this executable always trains on the CPU. Usage tar xzf hamelin-0.2.1-linux-x86_64-cpu.tar.gz ./hamelin Requires a graphical desktop and a few Qt system libraries; see DISTRIBUTION.md. Example datasets are in datasets/. To train on an NVIDIA GPU, install Hamelin from the source code (uv sync, then uv run hamelin); the default install includes CUDA support. See the README.

Published version of: Кобзєв, Володимир; Ховрат, Артем. Паралельне Прогнозування Ринкових Індикаторів в Умовах Соціальних Катастроф на Основі Авторегресійних Моделей. Інформаційні Технології: Наука, Т…

Published version of: Кобзєв, Володимир; Ховрат, Артем. Паралельне Прогнозування Ринкових Індикаторів в Умовах Соціальних Катастроф на Основі Авторегресійних Моделей. Інформаційні Технології: Наука, Техніка, Технологія, Освіта, Здоров'я: матеріали XXXI Міжнародної науково-практичної конференції MicroCAD-2023 (Харків, Україна, 17–20 травня 2023 року). НТУ «ХПІ», 2023. pp. 1084.Original publication: http://science.kpi.kharkov.ua/wp-content/uploads/2023/05/Zbirnik-tez-MicroCAD-2023-new_compressed-1.pdf

Code used to produce the results and figures in the manuscript Infrastructure-Aware Urban Flood Susceptibility from Multi-Event High-Water Marks: A Multi-Model Comparison.

Sound detection amidst noise presents an important challenge in audition. Many naturally occurring sounds (rain, wind) can be described and predicted statistically, so-called sound textures. Previous …

Sound detection amidst noise presents an important challenge in audition. Many naturally occurring sounds (rain, wind) can be described and predicted statistically, so-called sound textures. Previous research has demonstrated humans' ability to leverage this statistical predictability for sound recognition, but the neural mechanisms remain elusive. We trained mice to detect vocalizations embedded in sound textures with different statistical predictability, while recording and optogenetically modulating the neural activity in the auditory cortex. Mice showed improved performance and neural representation if they sampled the statistics longer per trial. Textures with more exploitable structure, specifically higher cross-frequency correlations (CFCs) improved performance and vocalization decoding. Activating parvalbumin-positive (PV) interneurons had an asymmetric effect, improving detection and neural representation of vocalizations for low, and vice versa for high CFCs. Thus, mice can exploit stimulus statistics to improve the sound detection in noise, reflected in performance and neural activity, while relying on PV interneurons.

Diagnosis of cystic echinococcosis is complex and has to be confirmed by the combination of immunological tests and imaging techniques. In this study heat shock proteins were induced and their immunor…

Diagnosis of cystic echinococcosis is complex and has to be confirmed by the combination of immunological tests and imaging techniques. In this study heat shock proteins were induced and their immunoreactivity was assessed by ELISA. Sera were collected from 34 hydatid patients who were admitted to the Rizgary Teaching Hospital through October 2013 to July 2017, in addition to 29 healthy donors and 18 non-hydatid cases. For heat shock response, two batches of 25000 protoscoleces (Ps) were incubated separately at 42°C and 45 °C for 4 hours. Heat treated and normal Ps were disrupted and the extracts were divided into two parts. One part was directly used as source of antigens (PE, PE42 and PE45) and the other one was partially purified on Sephadex G150. The immunoreactivity of these antigens, as well as hydatid fluidwas assessed by ELISA. The cutoff value to differentiate positive from negative sera was established by receiver operating characteristic (ROC) analysis. Two peaks of PE42 and three peaks of PE45 resulted by Sephadex G150. Extracts of 42ºC treated Ps resulted in two protein peaks and were used as PE42P1 and PE42P2 antigens. For 45°C treated Ps, the chromatography patterns resulted in three protein peaks and were used as PE45P1, PE45P2 and PE45P3 antigens. Highest rates of sensitivity, specificity and diagnostic accuracy were detected with PE42P2 (91.2%) and PE45P2 (91.2%). Sensitivity of ELISA was consistent for liver cysts with all applied antigens. Hydatid antigens extracted from heat treated Ps markedly raised the sensitivity of ELISA to detect anti-hydatid IgG.

What's Changed Add a from host radiative transfer solver by @K20shores in https://github.com/NCAR/tuv-x/pull/217 Update version to 0.17.0 by @boulderdaze in https://github.com/NCAR/tuv-x/pull/220 Fu…

What's Changed Add a from host radiative transfer solver by @K20shores in https://github.com/NCAR/tuv-x/pull/217 Update version to 0.17.0 by @boulderdaze in https://github.com/NCAR/tuv-x/pull/220 Full Changelog: https://github.com/NCAR/tuv-x/compare/v0.16.0...v0.17.0