Benchmarking Study of Parameter Variation When Using Signature Fingerprints Together with Support Vector Machines

Abstract

QSAR modeling using molecular signatures and support vector machines with a radial basis function is increasingly used for virtual screening in the drug discovery field. This method has three free parameters: C, γ, and signature height. C is a penalty parameter that limits overfitting, γ controls the width of the radial basis function kernel, and the signature height determines how much of the molecule is described by each atom signature. Determination of optimal values for these parameters is time-consuming. Good default values could therefore save considerable computational cost. The goal of this project was to investigate whether such default values could be found by using seven public QSAR data sets spanning a wide range of end points and using both a bit version and a count version of the molecular signatures. On the basis of the experiments performed, we recommend a parameter set of heights 0 to 2 for the count version of the signature fingerprints and heights 0 to 3 for the bit version. These are in combination with a support vector machine using C in the range of 1 to 100 and γ in the range of 0.001 to 0.1. When data sets are small or longer run times are not a problem, then there is reason to consider the addition of height 3 to the count fingerprint and a wider grid search. However, marked improvements should not be expected.

Deep-learning: investigating deep neural networks hyper-parameters and comparison of performance to shallow methods for modeling bioactivity data

“…Scikit-learn implementation of SVM is based on the Libsvm and the Liblinear libraries [66, 67]. When optimizing the SVM using the non-linear RBF kernel the values for hyper-parameters gamma ( γ ) and Cost ( C ) where selected with similar range to those reported by Alvarsson et al [68]. Here values for gamma ( γ ) tested were 10e−1, 10e−2, 10e−3, 10e−4, 10e−5 and for cost ( C ) 1, 10, 100, 1000.…”

Identifying new clotting factor XIa inhibitors in virtual high‐throughput screens using PCA‐GA‐SVM models and signature

“…The two major categories of vHTS approaches are molecular simulation (e.g., AutoDock, DOCK, Flex, AMBER, GROMACS, CHARMM), and ligand‐based scoring . Molecular simulation uses models and concepts from physics, chemistry, and mathematics to make optimized ligand/substrate interaction predictions.…”

“…The two major categories of vHTS approaches are molecular simulation (e.g., AutoDock, DOCK, Flex, AMBER, GRO-MACS, CHARMM), [41][42][43][44][45][46][47][48][49][50][51][52] and ligand-based scoring. 49,[53][54][55] Molecular simulation uses models and concepts from physics, chemistry, and mathematics to make optimized ligand/substrate interaction predictions. Although molecular simulations require minimal experimental data, it is computationally expensive and requires numerical solution convergence for confidence.…”

Towards agile large-scale predictive modelling in drug discovery with flow-based programming design principles

“…Often, the QSAR model algorithms come with free parameters that need to be determined, e.g., support vector machines based on the radial basis function has the free parameters and cost  [47] and k-nearest neighbour has k  [46]. A common way of determining actual values for parameters such as these is a grid search or “parameter sweep”.…”