New Workflow for QSAR Model Development from Small Data Sets: Small Dataset Curator and Small Dataset Modeler. Integration of Data Curation, Exhaustive Double Cross-Validation, and a Set of Optimal Model Selection Techniques

Quantitative structure–activity
relationship (QSAR) modeling
is a well-known in silico technique with extensive
applications in several major fields such as drug design, predictive
toxicology, materials science, food science, etc. Handling small-sized
datasets due to the lack of experimental data for specialized end
points is a crucial task for the QSAR researcher. In the present study,
we propose an integrated workflow/scheme capable of dealing with small
dataset modeling that integrates dataset curation, “exhaustive”
double cross-validation and a set of optimal model selection techniques
including consensus predictions. We have developed two software tools,
namely, Small Dataset Curator, version 1.0.0, and Small Dataset Modeler, version 1.0.0, to effortlessly execute the proposed workflow. These tools are
freely available for download from https://dtclab.webs.com/software-tools. We have performed case studies employing seven diverse datasets
to demonstrate the performance of the proposed scheme (including data
curation) for small dataset QSAR modeling. The case studies also confirm
the usability and stability of the developed software tools.