Our Work

This repository contains MD trajectories, structure files, data and scripts employed to analyze data used in the paper "The structural disorder of Dsup and its interaction with the human nucleosome: A…

This repository contains MD trajectories, structure files, data and scripts employed to analyze data used in the paper "The structural disorder of Dsup and its interaction with the human nucleosome: A computational study" submitted to the International Journal of Biological Macromolecules.

CORRECTION NOTICE (added 28 September 2026)The manuscript files in this version (00_manuscript/PAPER6_V5_MANUSCRIPT_FINAL.md and 15_publication_files/PAPER6_MANUSCRIPT_FINAL.pdf) contain three interpr…

CORRECTION NOTICE (added 28 September 2026)The manuscript files in this version (00_manuscript/PAPER6_V5_MANUSCRIPT_FINAL.md and 15_publication_files/PAPER6_MANUSCRIPT_FINAL.pdf) contain three interpretive errors. Corrected versions of the paper are in preparation. The data, code and numerical results in this record are unchanged.1. Temporal validation: the failure on the validation period is mainly caused by the frozen calendar spline, whose final basis coefficient (-1156.29, SE 5714.41) is identified only from the last hours of the development period; it is not evidence that the market or the feature-label relationship changed.2. Label audit: in 42 of the 43 reconstruction violations the graduation was recorded at exactly the poll that the reconstruction had inferred as the first unseen one, so the reconstructed exit time was wrong, not the label; the statement that these tokens were last seen at rank 45 or beyond is not supported by the deposited files.3. Cohort: a token first detected in the last 24 hours of collection enters the cohort only if it is labeled GRADUATED (56 tokens, all GRADUATED), so the last day of the validation cohort is outcome-selected; the manuscript does not disclose this.Also corrected: the two-minute-window odds ratios are 1.44 and 7.55, not 1.32 and 7.36 (the same two entries in 08_provenance/NUMBER_PROVENANCE.csv are wrong); the 183 failed development bootstrap refits are caused by calendar-spline columns left empty by the resampled days, not by category-level mismatches as the output disclaimer states; 30 SOL is a point mass of initial market capitalization, not a graduation threshold; the has_twitter interval includes 1 when tokens above the development 99th percentile of initial market capitalization are removed; and the analysis protocol, frozen after earlier analyses of the same data, is a pre-specified re-analysis protocol rather than a pre-registration. Where this description or the files say "pre-registered", read "pre-specified". RED-PUMP-2026-v1 is a reproducibility package for a pre-registered fintech-microstructure study of 860,213 pump.fun token launches recorded by an off-chain V3 collector between 2026-05-08 and 2026-06-10 (regime data-collection window). The primary analytical cohort is 749,816 unique mints in the V3 primary window (1,597 GRADUATED, 748,219 TIMEOUT). Two pre-registered analyses are conducted: (i) a measurement audit of the collector's terminal classifier (documented at 04_frozen_inputs/OUTCOME_ASCERTAINMENT_MECHANISM.md); and (ii) a locked temporal-holdout evaluation of a pre-registered logistic association model (15-day development split followed by a 14-day validation split). Development discrimination is strong (AUROC 0.8594); validation discrimination is indistinguishable from chance within the bootstrap interval (AUROC 0.4642, percentile interval [0.4112, 0.5196] containing 0.5000), and validation calibration deteriorates substantially (slope 0.013, intercept +2.816). Of nine automated stability-gate evaluations emitted by the corrected pipeline, two pass, six fail, one is not evaluable. This v1.6 release publishes four frozen protocol documents as redacted copies, with system-specific file paths and identifiers replaced by placeholders, and corrects one input checksum in the pipeline that had stopped the rebuild in earlier versions. Data, outputs and results are unchanged; see REDACTION_NOTE.md. Version 1.5 superseded v1.3 (DOI 10.5281/zenodo.21908375) per the deposited preprint correction and withdrawal notice at 11_corrigendum/PHASE14_CORRIGENDUM_FOR_ZENODO_21908375.md. The concept DOI 10.5281/zenodo.20633486 auto-resolves to the latest published version. Package: 72 files (deposit tree plus SHA256SUMS), 348.53 MB. Reproducibility entry point: bash REBUILD_ANALYSIS.sh --launches PATH --outcomes PATH (PowerShell equivalent shipped). Peak resource requirement: approximately 26 GB RAM and 2 hours wall-clock during the 500-replicate day-block bootstrap phase. Frozen input SHA-256 verification is enforced by d1_verify() before execution; internal manifest verifies with sha256sum -c SHA256SUMS (71 files OK). Licenses: manuscript, figures, tables, and provenance documents CC-BY-4.0; code and tests MIT (see LICENSE).

This repository contains the image dataset, reference proxy masks, generative reconstructions, model-derived binary vegetation masks, and analysis code used in the study “Generative foundation m…

This repository contains the image dataset, reference proxy masks, generative reconstructions, model-derived binary vegetation masks, and analysis code used in the study “Generative foundation models for agricultural image annotation: Reliability assessment in semi-arid rangelands.” The dataset comprises 98 proximal RGB field images representing 10 indigenous forage shrub and rangeland vegetation classes collected in the arid and semi-arid lands of Marsabit County, Kenya. The images represent heterogeneous field conditions, including exposed soil, dry litter, shadows, sparse and dense canopies, woody stems, open crowns, partial occlusion, and visually complex vegetation–background boundaries. Each source image was processed using four deployed generative image systems: gpt-image-2 gpt-image-1.5 gemini-3-pro-image qwen-image-edit-2511 A common prompt constrained reconstruction protocol was used to preserve the target vegetation while reducing visually distracting background information. The archive therefore contains 392 reconstructed image outputs together with corresponding binary vegetation masks. Reference proxy masks associated with the original RGB images are also provided. These masks were generated from the original imagery using a SAM 3.0 annotation workflow and were used as consistent comparative references for evaluating reconstruction-derived vegetation priors. They should be interpreted as reference proxy masks rather than absolute pixel-level ground truth. The accompanying analysis notebook provides a reproducible implementation of the study's multi metric reliability assessment. It includes: Dice and Intersection over Union agreement foreground area difference boundary F1 and boundary IoU structural and skeleton based measurements foreground loss and generated foreground background leakage internal canopy gap inpainting colour and appearance alteration pairwise model agreement multi model consensus analysis model disagreement mapping human review priority ranking The consensus analysis identifies vegetation regions consistently retained across models and regions where the reconstructed masks diverge. These disagreement regions are intended as candidate signals for human review, rather than calibrated predictive uncertainty or validated error maps.Repository structure:1_Original_images_classified2_original_masks_SAM33_masked_images_generated_ChatGPT4_ChatGPT_binary_masks5_masked_images_generated_Copilot6_Copilot_binary_masks7_masked_images_generated_Gemini8_Gemini_binary_masks9_masked_images_generated_Qwen10_Qwen_binary_masksmultimodal_generative_AI_Zenodo.ipynbREADME.mdrequirements.txtCITATION.cff All images and masks used for paired quantitative analysis were standardised to a spatial grid of 1184 × 864 pixels. The included notebook automatically validates the expected folder structure, performs reference v/s model and model v/s model comparisons, generates consensus statistics, and exports reproducible CSV results and figures. This repository is intended to support reproducibility, benchmarking, and further research on generative foundation models, agricultural image annotation, rangeland vegetation analysis, and human-in-the-loop annotation workflows. Authors: Shubham Rana, Oliver Hensel, Abozar NasirahmadiAffiliations: University of Kassel, Germany; Swedish University of Agricultural Sciences, SwedenProject: InfoRange — Increasing Efficiency in Rangeland-Based Livestock Value Chains through Machine Learning and Digital Technologies.AcknowledgmentsThis research was conducted within the InfoRange project, funded by the German Federal Ministry of Research, Technology andSpace (BMFTR) under grant “01LL2201B”. We thank the funding bodies for their support, Compwiz Creations (https://compwiz.io/) for assisting with app development used in data collection, and our German and Kenyan InfoRange partners for their contributions to image collection. This research did not receive any specific grant from funding agencies in the public, commercial, or not-for-profit sectors.

ABSTRACT: The geometry of cutting tools significantly influences machining quality, dimensional accuracy and tool life. The manufacturing and reconditioning of cutting tools require the maintenance of…

ABSTRACT: The geometry of cutting tools significantly influences machining quality, dimensional accuracy and tool life. The manufacturing and reconditioning of cutting tools require the maintenance of the geometrical parameters established during the design stage. This paper presents the design, development and experimental validation of a universal sharpening device intended for lathe tools and indexable carbide inserts used in profile turning operations. The proposed solution enables accurate generation of thread-form profiles such as trapezoidal, worm, metric and Whitworth geometries by means of a simple mechanical adjustment system. The sharpening process is performed on a conventional RP250 grinding machine equipped with a diamond wheel. Experimental investigations demonstrated accurate profile generation, good repeatability and surface roughness values between Ra = 0.069 and 0.104 μm. The developed device represents an economical alternative to universal tool grinding machines and contributes to the recovery and reuse of worn carbide inserts.

Photony Unification Compendium, Volume V develops the unification framework of Photony Theory and examines how its foundational concepts may connect spacetime, energy, quanta, particles, fields, force…

Photony Unification Compendium, Volume V develops the unification framework of Photony Theory and examines how its foundational concepts may connect spacetime, energy, quanta, particles, fields, forces, quantum mechanics, and relativity within a common physical model. The volume presents Photony's proposed "Big Theory of Everything" approach, built around the interaction of energy filaments with quasi-static Planck Space and the relationships among the Photony unification classes SGEMT, QPC, and EMFF. Rather than replacing established physics wholesale, the book frames Photony as a mechanistic and kinematic foundation intended to support and connect elements of classical physics, quantum theory, relativity, and the Standard Model. Photony Compendium- Volume V The volume reviews major historical and contemporary approaches to unification, including general relativity, quantum mechanics, string theory, Grand Unified Theories, quantum electrodynamics, quantum chromodynamics, quantum field theory, quantum gravity, electroweak theory, and the Standard Model. These subjects are examined alongside Photony interpretations of classical photons, electrons, electromagnetic fields, the Higgs field, wave-particle duality, the Schrödinger equation, the Heisenberg uncertainty principle, and quantum entanglement. Photony Compendium- Volume V At the center of the proposed framework are energy filaments, elemental charge photons, chained entities, the Photony Motion/Spin Rule, the Manhattan metric, Planck Space compression, and the relationship between energy density and spacetime curvature. Photony treats mass as an emergent manifestation of spacetime curvature associated with concentrated energy and seeks to connect gravitational and electromagnetic phenomena through energy's interaction with the underlying Planck-scale structure of space. Photony Compendium- Volume V Volume V also places the unification effort within the larger Photony Compendium series. Earlier volumes develop quasi-static Planck Space and energy, Photony quanta, particles and chains, electromagnetic fields and breakdown phenomena, and energy science. Volume V brings these concepts together through their relationships with relativity, quantum mechanics, QFT, QED, QCD, quantum gravity, electromagnetic forces, the Higgs field, and the Standard Model. Photony Compendium- Volume V The appendices include a condensed Photony Foundations Primer, the Photony Postulates, an Einstein field-equation derivation, and a discussion of the double-slit experiment. Together, these materials make Volume V both the unification volume of the Photony Compendium series and a standalone reference for examining the proposed relationship between Photony Theory and major frameworks of modern physics.

Abstract Swing trading occupies a unique niche within financial markets, leveraging medium-term price momentum over holding periods typically spanning two to fourteen days. Despite its growing adoptio…

Abstract Swing trading occupies a unique niche within financial markets, leveraging medium-term price momentum over holding periods typically spanning two to fourteen days. Despite its growing adoption among retail investors, a comprehensive, open-architecture web platform unifying real-time data ingestion, technical analysis computation, portfolio tracking, and risk assessment under a single interface remains largely absent in the open-source domain. This paper presents SwingTrader, a full-stack web application constructed on the MERN (MongoDB, Express.js, React.js, Node.js) technology stack. The system integrates live market data feeds, computes widely adopted technical indicators—including the Relative Strength Index (RSI), Moving Average Convergence Divergence (MACD), Bollinger Bands, and Average True Range (ATR)—and presents actionable buy/sell signals through an intuitive, responsive dashboard. A Socket.IO-powered WebSocket layer delivers sub-second quote updates, while a JWT-based authentication module ensures secure, stateless user session management. Empirical evaluation across 10,000 concurrent simulated sessions demonstrates a mean API response latency of 87 milliseconds and 99.7% uptime over a 90-day observation window. User experience assessments with 42 participants yield a mean System Usability Scale (SUS) score of 81.4, categorising the application as ‘excellent’. The proposed architecture serves as a replicable blueprint for practitioners and researchers designing scalable, real-time financial analytics platforms. Keywords swing trading; MERN stack; React.js; Node.js; MongoDB; technical analysis; WebSocket; real-time finance; portfolio management; REST API

Increased interest in the sustainable, protein-rich and plant-based foods has been driving university dining centers to explore innovative plant-based menu options. However, scaling up recipes from be…

Increased interest in the sustainable, protein-rich and plant-based foods has been driving university dining centers to explore innovative plant-based menu options. However, scaling up recipes from benchtop to institutional dining centers presents practical challenges that require a systematic pilot evaluation before conducting a full-scale validation study. This pilot feasibility study evaluated two plant-based meals developed through an academic and dining services partnership at a large public university. The meals were a vegan sunrise bowl and a plant-based lasagna. Both used textured soy protein (TSP) as the main protein, and the Sunrise Bowl also used bulgur wheat as a whole-grain base. Three categories were evaluated including production scale-up, functionality, and preliminary consumer acceptance. Both meals were scaled to institutional batch volume and cooked with institutional kitchen equipment. TSP had a high water-uptake capacity, absorbing 184-196% of its dry weight, while bulgur wheat absorbed 124%. The lasagna had a high-protein, high-fiber profile, supplying 59 g of protein and 15 g of dietary fiber per serving. Using a 9-point hedonic scale, the overall liking for lasagna was 7.89 (n=48), while for the sunrise bowl was 7.37 (n=42). Because the samples were independent convenience panels, these results are descriptive. Based on the observed response variances, we estimate that a definitive study would require roughly 80 to 200 respondents per product. Overall, both meals proved feasible to produce and serve in a university dining setting. These findings justify a confirmatory study. They also offer a reproducible model for academia and foodservice to develop plant-based menus.

Este documento sintetiza la trayectoria biográfica y profesional del tenista español Carlos Alcaraz Garfia (nacido el 5 de mayo de 2003 en El Palmar, Murcia). Se abordan sus inicios en e…

Este documento sintetiza la trayectoria biográfica y profesional del tenista español Carlos Alcaraz Garfia (nacido el 5 de mayo de 2003 en El Palmar, Murcia). Se abordan sus inicios en el tenis, su progresión en el circuito ATP, sus hitos históricos —incluyendo la obtención de títulos de Grand Slam y el ascenso al número uno del ranking mundial a una edad récord— y sus principales datos personales y de rendimiento físico. Asimismo, se incluye un análisis contextual sobre su impacto en el deporte de alto rendimiento y la gestión de su perfil profesional.  

DIROF is a cooperative verification technique for detecting arithmetic overflows in Cprograms. Bounded symbolic reasoning paired with arithmetic overflow assertions provides asystematic symbolic searc…

DIROF is a cooperative verification technique for detecting arithmetic overflows in Cprograms. Bounded symbolic reasoning paired with arithmetic overflow assertions provides asystematic symbolic search for overflow witnesses, while unwinding assertions can justify the absenceof overflow when the chosen unwind bound is sufficient. However, some overflows manifest onlyafter many loop iterations, requiring unwind depths beyond what symbolic reasoning can practicallyhandle. Execution-based analysis can scale to such depths, but frequently fails to generate the inputsequences needed to expose the arithmetic overflow. Dirof bridges this gap by using boundedsymbolic reasoning to construct, rather than fully validate, overflow-seeking executions. Dirof firstperforms bounded model checking with arithmetic overflow assertions. If no overflow is found, itthen checks unwind sufficiency. If the unwind is insufficient, Dirof applies a targeted input encodingthat restricts nondeterministic inputs to finite, type-specific overflow-seeking domains, synthesizinga compact input prefix. The prefix is then used as a concrete input pattern to execute the programunder runtime overflow checks to determine whether it induces an overflow beyond the symbolicallyexplored loop iterations. We instantiate Dirof using CBMC and Clang. On 3,680 SV-COMParithmetic-overflow benchmarks, under the same experimental setup, Dirof detects overflows in3,667 benchmarks compared with 3,556 for CBMC alone, while being about 3.6× faster overall.