Pharmacological affinity fingerprints derived from bioactivity data for the identification of designer drugs

Table 2 Clustering hyperparameters investigated

Hyperparameters	Parameter	Values explored
Hierarchical clustering
Linkage	Ward	Minimizes the variance of the clusters being merged
	Complete	Maximum distances between all observations of the two sets
	Average	Average of the distances of each observation of the two sets
	Single	Minimum distances between all observations of the two sets
Spectral clustering
Fully connected graph (RBF)	\(\gamma\)	[1–5]
Fully connected graph (RBF)	eigen_tol	[0.1, 0.01, 0.001,0.0001,0.00001, 0.000001]
\(k\)-nearest neighbor graph	n_neighbors	[7, 9, 11, 13, 15, 17, 19]
\(k\)-nearest neighbor graph	eigen_tol	[0.1, 0.01, 0.001,0.0001,0.00001, 0.000001]

The fcluster and dendrogram in scipy.cluster.hierarchy package are used for hierarchical clustering, the SpectralClustering in sklearn.cluster package are used for spectral clusterings

ISSN: 1758-2946