top of page
Search

SpectraGuruâ„¢ Releases Major Update to Expand Spectral Preprocessing, Analytics, Machine Learning, and Toolbox Workflows

SpectraGuruâ„¢, an open-source ecosystem for Raman and surface-enhanced Raman spectroscopy analysis, has released a major platform update that significantly expands its preprocessing, analytics, machine-learning, database, and toolbox capabilities.

The updated platform is now available at spectraguru.org and provides researchers with a broader set of browser-based tools for spectral analysis, including new smoothing and baseline-correction methods, expanded machine-learning classification models, spectrum simulation tools, peak assignment table collection, and an improved user interface for more accessible data processing.

The update strengthens SpectraGuru’s goal of making advanced spectral analysis more accessible, reproducible, and scalable for the spectroscopy community. By integrating preprocessing, statistical analysis, machine learning, database access, and interpretation tools into a unified platform, SpectraGuru supports a complete workflow from raw spectral data to meaningful scientific insight.


Major New Features


Expanded Analytics and Machine Learning

The platform now supports additional analysis and classification tools, including:

  • Spectral derivation

  • FFT analysis

  • Random Forest classification

  • K-nearest neighbors classification

  • Support vector machine classification

These additions complement existing visualization and analysis tools such as average spectrum and standard deviation plots, confidence interval plots, correlation heatmaps, hierarchical clustering heatmaps, principal component analysis, and t-SNE dimensionality reduction.


Expanded Preprocessing

The update also adds several preprocessing methods that are important for Raman and SERS data quality improvement, including:

  • Median filter smoothing

  • Wavelet denoising

  • SNIP baseline correction

  • Asymmetric least squares baseline correction

These tools expand the existing preprocessing workflow, which includes interpolation, cropping, despiking, smoothing, baseline removal, normalization, and outlier removal.


New Toolbox Functions

A new toolbox section has been added to support additional spectral interpretation and educational workflows. Current tools include:

  • Spectrum simulation

  • Peak assignment table collection

These features are intended to help users better understand spectral features, compare experimental and simulated spectra, and organize peak assignments for Raman and SERS studies.


Improved User Interface and Accessibility

The update also improves the user experience through clearer upload status messages, better user feedback, an updated function-usage table with documentation, and login support.


Growing Worldwide Adoption

SpectraGuru continues to see rapid international adoption. According to the latest platform update, SpectraGuru now includes 3,000+ spectral datasets, supports scalable analysis of 1,000 spectra at a time, has reached users in 98 countries and regions, and has recorded 694,071 total platform visits since release.

The platform includes both a standard spectrum database and a raw experimental database, providing a foundation for reproducible spectral analysis, machine-learning model development, and community-driven data sharing.


Community Use and Citation

Researchers who use SpectraGuru are encouraged to cite the platform in their work:

Ma, F. et al. Comprehensive Open-Source Ecosystem for Raman and SERS Spectroscopy: Introducing SpectraGuru. Analytical Chemistry, 98, 11186–11196 (2026).

Ma, F. et al. SpectraGuru: a community-guided path toward scalable Raman and SERS analysis. SPIE BIOS, Photonics West 2026, pp. 31–41.


Availability

Website: spectraguru.org

GitHub: Available through the SpectraGuru project page

More information: zhao-nano-lab.com


 
 
 
bottom of page