The PEAXACT Desktop App is an interactive chemometrics software for the creation of models (methods) for the quantitative multivariate analysis of UV/VIS/IR/Raman/NMR-spectra. It covers all steps of data visualization and modeling in a unified workflow: data pretreatment, spectral modeling with mechanistic and statistical methods, as well as calibration and classification. PEAXACT addresses both first-time users of quantitative spectroscopy and experienced users who wish to have access to a balanced portfolio of different analytical methods when creating models.
PEAXACT Desktop App
Visual Data Inspection
The Data Inspector is a graphical tool for the editing and visualization of spectra.
- Organisation of big data (spectra and other measurements, e.g. concentrations, temperature, measurement time)
- Explorative analysis with customizable 2D, 3D, and 4D (color-coded) diagrams
- Principal Component Analysis (PCA)
- Export of rendered and vector images (PNG, JPEG, TIFF, PDF, EPS, FIG)
Data Pretreatment
- Immediate preview of all data manipulations
- Region selection (global and local)
- Resampling (thinning, re-calculation by interpolation)
- Alignment
- Baseline correction (offset, linear trend, rubber band)
- Smoothing
- 1st and 2nd derivatives
- Standardization (maximum, area, SNV, peak)
- NMR-specific: FID to spectrum, apodization, phase correction
Spectral Modeling
- Band integration
- Peak fitting / peak deconvolution
- Component fitting
- Spectral Hard Modeling
- Direct Hard Modeling
- Indirect Hard Modeling (IHM)
- Complemental Hard Modeling (CHM)
- Hard Modeling Factor Analysis (HMFA)
Calibration
- Simple regression of Integration Models and Hard Models
- Ratiometric regression for relative concentrations
- Multivariate regression (PLS)
- Cross-validation (leave-out, k-fold, group-wise)
- Test-set validation
- Easy comparison of alternative calibrations
- Comprehensive reporting for finding the best calibration
Classification
- Database Lookup
- PCA-based Discriminant Analysis
- Cross-validation (leave-out, k-fold)
- Test-set validation
- Easy comparison of alternative Classification Models
- Comprehensive reporting
Overview of analysis methods
- Principal Component Analysis (PCA)
- Cluster Analysis
- Peak Integration
- Peak Fitting / Peak Deconvolution
- Component Fitting
- Multivariate Curve Resolution (MCR)
- Projection to Latent Structures (PLS)
- Identification by Database Lookup
- Indirect Hard Modeling (IHM)
- Complemental Hard Modeling (CHM)
- Hard Modeling Factor Analysis (HMFA)
