Balalab-skku

Select Program:

Select Adduct Type:

Data input example:

**Enter data in CSV format (separated by commas). Expected column names must match those in the input example.

Choose File:

Upload a CSV or TXT file with molecular data

Email address:

**Put your email address if you want to get result via email.

About Physicochemical Properties Prediction

Physicochemical Properties Prediction (PPP) is an advanced deep learning tool designed to predict key molecular properties of ginsenosides and related compounds.

This tool can predict:

Collision Cross Section (CCS): A molecular property important for ion mobility spectrometry and structural characterization. Can be predicted with or without adduct ions.
Retention Time (tR): Critical for chromatographic separation and compound identification. Predictions account for adduct formation in mass spectrometry.

The models integrate conventional molecular descriptors with NLP-based embeddings, trained on our comprehensive GinDB-AI database containing data from 1963 to 2024.