Elsewhere correction for multiple comparisons was made using a cluster threshold of P < 0.01.
For all whole-brain analyses, SPMs were corrected for multiple comparisons by using a cluster threshold of Z > 2.3, and a corrected cluster significance threshold of P < 0.05.
Whole-brain analyses for the pain > no-pain contrast are reported using a cluster-forming threshold of p <.001 (uncorrected, cluster size > 10), with cluster-level family-wise error (FWE) correction.
The starting point of the technique is the definition of the output centroids using a clustering method compatible with most thresholding techniques in the literature.
Evaluation of the conformer model accuracy using RMSD is an intuitive and convenient choice, as the conformer model clustering in PubChem3D uses an RMSD value as a clustering threshold; however, in practice, PubChem3D primarily uses three measures in 3-D similarity comparison between molecules: shape-Tanimoto (ST), color-Tanimoto (CT), and combo-Tanimoto (ComboT) [37 40].
For the whole brain comparisons, we used a cluster-level threshold corrected for multiple comparisons of P < 0.05 and an uncorrected voxelwise threshold of P < 0.001 (Fig. 2 and Supplementary Fig. 5, Table 1 and Supplementary Table 3).
In the case of cluster-level tests, we used a cluster-defining threshold of P < 0.001 (uncorrected).
We use a minimum cluster size Cmin=5, a clustering threshold Z θ =0.3, and a distance threshold d θ =150 m.
Sequence headers were parsed to retain the scaffold/chromosome information and the two datasets concatenated and clustered by peptide sequence identity using CD-HIT v4.6 [ 50], with a clustering threshold of 75%% amino acid identity.
The resulting clusters of differences in GM volume were corrected for multiple comparisons using random field theory with a cluster threshold of t >2.3 and a reliability of p < 0.05 for extend of the cluster.
