An Infrastructure to Mine Molecular Descriptors for Ligand Selection on Virtual Screening

Table 1

Example of data mining input file format. Column 1 represents the ligand identification (not used on data mining experiments). Columns MwT, log𝑃, HBD, HBA, and so forth correspond to the molecular descriptors for each ligand, our predictive attributes. The last column is the target attribute FEB.

Ligand	MwT	log𝑃	HBD	HBA	⋯	FEB
1	297.44	4.61	1	2	⋯	−8.50
2	348.47	3.82	2	4	⋯	−7.96
⋯	⋯	⋯	⋯	⋯	⋯	⋯
𝑁	200.19	0.54	2	5	⋯	−6.89