Glycan X-Ray Diffraction (XRD) Data Interpretation, Integration and Visualization
Designed to Bring Efficiency to Your Glycan XRD Data Visualization
XRD is an important tool for analyzing the stereo conformations of polysaccharides and elucidating conformational relationships. Combining automated experiments and high-throughput algorithms, CD BioGlyco provides XRD pattern classification, training, data interpretation, integration, and visualization services. Our Glycoinformatics-assisted Analysis/Prediction team aims to provide standardized, predictive, and operational data analysis services to clients worldwide. In particular, we provide high-quality Glycan Data Visualization services.
Data collection and algorithm optimization
Our researchers provide efficient XRD mode simulations and an equivalent number of virtual pair distribution functions (PDFs). Relying on linear combinations of single modes, we provide Fourier transform computation and training services for virtual PDFs.
To address the poor signal-to-noise ratio in images, our team provides professional post-imaging processing services utilizing deep neural networks. This network model combines potential physical features in the image and detects new features in the image by self-supervision.
Glycan XRD data visualization
Based on advanced tools, our professional algorithm team provides raw XRD and post-analysis XRD data processing services. We help our clients seamlessly import data by streamlining the analysis process.
The analysis tools we offer include an interactive chart with well-designed x- and y-axes that display your data. Moreover, multiple traces and datasets from different information packages such as texture and preferred orientation, peak broadening, and peak displacement are displayed on the same chart.
Whether it's adjusting x and y values, normalizing data for better comparisons, or switching to a waterfall view for hierarchical analysis, our high-quality tools meet your needs.
Publication
Technology: Deep learning, XRD pattern simulation, Variational autoencoder (VAE) model
Journal: Npj Computational Materials
Published: 2021
IF: 9.4
Results: This study presents visualization and novelty detection of XRD datasets through deep learning. Combined with XRD pattern simulation and VAE, it quickly interprets the potential information of crystal structures. It accurately recognizes data distributed in addition to the training set. The developed dynamic visualization tool has the ability to identify patterns under multiple constraints.
Fig.1 Visualization of the VAE latent space representation. (Banko, et al., 2021)
Applications
- Glycan XRD data visualization points out a new way for automated multimodal characterization.
- Glycan XRD data visualization can be used for real-time analysis of carbohydrate XRD data.
- Glycan XRD data visualization can be used for on-the-fly analysis of a dataset distribution across different structures.
Advantages of Us
- Our researchers offer data visualization services for various diffraction modes such as X-ray single crystal diffraction and X-ray fiber diffraction.
- We offer analytical tools with powerful data compatibility and versatile visualization.
- Depending on the specific needs of our clients, we offer machine learning model development services to customize the data model that best suits your needs.
Frequently Asked Questions
- Can the XRD patterns and data obtained by the analysis tool be directly exported for use?
- Sure! Our excellent calculation team provides the service of exporting data graphs in a defined format for direct use by our clients.
- Is it possible to compare multiple glycan XRD datasets?
- Our tool supports overlaying multiple trajectories from different datasets and generating comparative plots to better elucidate the data.
CD BioGlyco has a well-trained team and advanced computers to provide clients with glycan XRD visualization services. We always maintain a passionate and professional philosophy to serve every client. Feel free to contact us for any reason whatsoever.
Reference
- Banko, L.; et al. Deep learning for visualization and novelty detection in large X-ray diffraction datasets. Npj Computational Materials. 2021, 7(1): 104.
For research use only. Not intended for any diagnostic use.
Related Services
