Filter Results
1033 results
  • Minimum working example (MWE) of the a modified training and scoring autoencoder that features Cummulative Error Scoring (CES) and Percentile Loss (PL) Training.
    Data Types:
    • Software/Code
  • Designing a novel encoding method suited for Voice Activity Detection using Spiking Neural Networks
    Data Types:
    • Software/Code
  • The benefit and safety of exercise training for patients with neuromuscular disorders has long been a contentious topic. This is, in part, due to recognised challenges associated with rare diseases including small and heterogenous patient populations. We performed a systematic review and meta-analyses to evaluate the effectiveness and safety of interventional exercise and establish minimal clinically important differences (MCID) in outcome measures in neuromuscular disorders to facilitate clinical interpretation.
    Data Types:
    • Software/Code
  • Reproducable code for the article Vogel, D., Reuber, A., & Vogel, R. Developing a short scale to assess public leadership. Public Administration.
    Data Types:
    • Software/Code
  • Code and Data to reproduce the analysis of the pilot study for the Epistemic Trust Assessment
    Data Types:
    • Software/Code
  • We present both algorithms of strict and approximate OrthDNNs, and for the later ones we propose a simple yet effective algorithm called Singular Value Bounding (SVB), which performs as well as strict OrthDNNs, but at a much lower computational cost. We also propose Bounded Batch Normalization (BBN) to make compatible use of batch normalization with OrthDNNs.
    Data Types:
    • Software/Code
  • SPHERA v.9.0.0 (RSE SpA): Smoothed Particle Hydrodynamics Computational Fluid Dynamics research code. Applications: floods (with transport of solid bodies, bed-load transport, damage on electrical substations, flood-control works); fast landslides (in rocks and soils) and wave motion; hydroelectric plants; fuel sloshing tanks. Copyright: RSE SpA Author list: available within the documentation file
    Data Types:
    • Software/Code
  • A C-library for modeling the energy consumption of BLE. Given certain parameter values, the energy consumption of a BLE radio can be predicted. This Code Ocean Capsule will execute a simple example (see folder "examples"), which can be adjusted before run.
    Data Types:
    • Software/Code
  • Generative machine learning models sample drug-like molecules from chemical space without the need for explicit design rules. A deep learning framework for customized compound library generation is presented, aiming to enrich and expand the pharmacologically relevant chemical space with new molecular entities ‘on demand’. This de novo design approach was used to generate molecules that combine features from bioactive synthetic compounds and natural products, which are a primary source of inspiration for drug discovery. The results show that the data-driven machine intelligence acquires implicit chemical knowledge and generates novel molecules with bespoke properties and structural diversity. The method is available as an open-access tool for medicinal and bioorganic chemistry.
    Data Types:
    • Software/Code
  • Detecting epistatic interactions in GWAS (genome-wide association studies) data is of great significance in studying common and complex diseases; however, the ability to detect high-order epistatic interactions in GWAS data is still insufficient. Existing methods are usually used to identify two-order interactions, and they cannot detect a large number of interactions. In this article, we propose a novel stochastic approach named SHEIB-AGM (stochastic approach for detecting high-order epistatic interactions using bioinformation with automatic gene matrix). SHEIB-AGM utilizes bioinformation to construct a gene matrix. In each iteration, it randomly generate a high-order SNP combination based on the gene matrix. SHEIB-AGM utilizes k2 (the Bayesian network scoring criterion) and G-test to detect epistasis in the generated combination and automatically update the gene matrix. We have compared SHEIB-AGM with six other methods, i.e., DECMDR, SNPHarvester, MACOED, AntEpiSeeker, HS-MMGKG and SEE, on simulated data including 108 epistatic models and 17,600 files. The results demonstrate that SHEIB-AGM greatly outperforms the above methods in terms of F-measure and power. We utilized SHEIB-AGM (with and without bioinformation) to analyze a real GWAS dataset from the Wellcome Trust Case Control Consortium. The results indicate that SHEIB-AGM with bioinformation can detect 33.94~3069.40-times more epistatic interactions. We have found numerous genes and gene pairs that may play an important role in seven complex diseases. Some of them have been found in the CTD database (the Comparative Toxicogenomics Database).
    Data Types:
    • Software/Code
2