In the third step, the one conformer with the lowest stored similarity is selected as a new reference shape.
As a consequence, reference shapes generated from structures with less common features tend to cluster fewer database conformers than those coming from compounds with more common features.
In this work, we seek to expand upon these earlier two efforts by exploring in more depth the rate of growth of shape space as a function of reference shape count, conformer volume, and ST value with a much larger data set of 16.4 million biologically relevant small molecules and their 1.46 billion diverse conformers.
Instead, finding a common reference shape triggers computing, via Eq. 2, the shape Tanimoto between the query conformer and database conformer, as may be performed by a typical brute-force ROCS approach.
A scheme to specify planar C2 Pythagorean-hodograph (PH) quintic spline curves by control polygons is proposed, in which the "ordinary" C2 cubic B-spline curve serves as a reference for the shape of the PH spline.
Different signals including sinusoidal, square, and linear waves are input into the control system as reference shape of the mirror and three outputs of the mirror are tracked simultaneously.
Different wave signals are input into the control system as reference shape and outputs (transverse deflection) of the mirror are monitored to track the desired membrane deformation with recognizable accuracy.
Also, for efficiency purposes, as conformers are deleted from the database, care must be taken to ignore reference shapes that no longer represent any database conformer to prevent unnecessary comparisons to a redundant reference shape.
In other words, where A is the reference shape (i.e. a set of pixels) and B is the shape detected via our algorithm, our normalized error measure e is calculated as: (8) The same error measure is calculated for the reference shape and the shape detected via manually-initialized ASM.
