Welcome to PARADISe
Traditionally, GC-MS data have been handled through targeted analysis, which is time consuming and often subject to user-variability. Furthermore, it usually escapes finding minor compounds, even if the raw analytical data provides comprehensive information.
PARADISe is a deconvolution and identification software that allows for an optimized untargeted profiling analysis of GC-MS data, ensuring:
-
Capturing the greatest analytical information from the raw analytical signals
Chromatographic co-elutions are not an issue anymore. Even severely co-eluted compounds are easily resolved, improving their identification and quantification.
PARADISe works beyond limits. Peak areas of chemical compounds below limit of detection (LOD) are provided.
-
Efficient analysis time
Large and complex datasets are handled at once, regardless of the number of samples. Indeed, the more samples the better performance of the resulting models.
-
Robust results, little user dependence
PARADISe favors simplicity and automation to turn raw GC-MS data into a Peak Table. Together with the intuitive graphical user interface, this means decision making does not require highly experienced users.
PARADISe is based on scientifically evaluated methods, in particular tensor decomposition (PARAFAC2), which incorporates information from all modes (samples, retention time, and mass bins) of the dataset without needing to correct the raw data.
PARADISe is build for analysing multiple samples at once and the requird user interaction is the same regardless of analysing 20, 200 or 2000 samples. The final product of PARADISe is a peak table with the estimated relative concentration (peak area) in each sample of the identified compounds together with each compounds estimated mass spectrum.
PARADISe is offered in two versions: the Community Edition (for free) and the Professional Edition (license fee applies). An overview of the available features is shown below:
Recommended hardware requirements are as follows:
-
Windows 10/11
-
Between 1 and 512 CPU cores.
-
8 GB of RAM
-
Separately installed NIST MS Search Program (only if spectra matching is desired). PARADISe has been tested against the NIST11, NIST14 and NIST20 MS Search Program, but other versions should also work.
In general, the better the computer features, the better PARADISe will function. In particular, having several cores will allow parallel computations to take place, which can speed up the calculations. Also important, is that there is enough memory on the computer.
Features | Community edition | Professional edition |
|---|---|---|
Automatic model fitting and evaluation | No | Yes |
Select which deep learning algorithm is used for finding chemical peaks | No | Yes |
Read retention indices from NIST (It requires standalone version of NIST MSSearch) | No | Yes |
Apply already established model to generate peak tables from new chromatograms (batch analysis) | No | Yes |
Guided selection of chemicals | Yes | Yes |
Automated pre-alignment of raw signal | Yes | Yes |
Quantification below limit of detection | Yes | Yes |
Improved quantification | Yes | Yes |
Automatic lookup of mass spectra in NIST database (It requires standalone version of NIST MSSearch) | Yes | Yes |
Improved mass spectral estimates | Yes | Yes |
