Anti-aging Compound Dataset and CMAUP Natural Product Candidate Set

This dataset is the correlation research data for the paper "Small Sample Unbalanced Table Learning Method for Natural Anti Aging Compound Screening", which includes two parts: the anti-aging compound dataset and the CMAUP natural product candidate dataset. The anti-aging compound dataset contains a total of 1000 labeled compounds, including 206 samples with positive anti-aging activity and 794 samples with relative negative activity.

Each compound contains a compound name, class label, source information, SMILES representation, and a 100 dimensional numerical molecular descriptor. In the category label, senolytic=1 represents a positive sample with anti-aging activity, while senolytic=0 represents a relatively negative sample. This dataset is used for model training, validation, and testing in the paper. The CMAUP natural product candidate set contains 5148 natural product records, each record containing compound names, SMILES representations, and 100 dimensional numerical molecular descriptors consistent with the anti-aging compound dataset, used for virtual screening of natural anti-aging candidate compounds after training. The molecular descriptors in two datasets were calculated using RDKit.