NCI1

Introduced by Nikil Wale et al. in Comparison of descriptor spaces for chemical compound retrieval and classification

The NCI1 dataset comes from the cheminformatics domain, where each input graph is used as representation of a chemical compound: each vertex stands for an atom of the molecule, and edges between vertices represent bonds between atoms. This dataset is relative to anti-cancer screens where the chemicals are assessed as positive or negative to cell lung cancer. Each vertex has an input label representing the corresponding atom type, encoded by a one-hot-encoding scheme into a vector of 0/1 elements.

Source: Ring Reservoir Neural Networks for Graphs

Homepage