Virtual Screening Metrics

Computes set of popular performance estimations for a virtual screening. The input is expect to be sorted according to the ranking criteria (performance estimate), such that the most promising compound is in the first row. The measure computed are the enrichment factor in the top X%, the area under the ROC curve (AUROC), the robust initial enhancement (RIE, Sheridan et al, JCICS, 2001, 41, 1395−1406) and the Boltzmann-enhanced discrimination of ROC (BEDROC, Truchon and Bayly, JCIM, 2007, 47 (2), pp 488–508)


Activity column
Numerics column containing the true activity class. A compound is regarded as active if its respective property is above zero.
BEDROC alpha
The alpha parameter used to compute the RIE (Robust Initial Enhancement) and the BEDROC
% Enrichment Factor
The top X% of the ranked data is used to compute the enrichment of true actives in that part.

Input Ports

Table sorted with respect to the screening score (e.g. similarity)

Output Ports

VS Metrics


This node has no views




You want to see the source code for this node? Click the following button and we’ll use our super-powers to find it for you.