Details
The discovery of novel therapeutic agents requires exploring increasingly expanded and complex chemical space, resulting in rapidly growing data proliferation. Computational techniques that scale favorably with the expanding chemical space and provide efficient insights have become critical to any lead discovery and lead optimization effort.
Cheminformatics techniques such as fingerprint-based similarity searching and substructure matching can screen millions of compounds in seconds; clustering and diversity selection can analyze and improve the content of real and virtual compound libraries; principal components analysis and self-organizing maps reduce complex, high dimensional information into easily visualized relationships in a small number of dimensions; and supervised learning techniques offer quantitative models that elucidate structure-activity relationships and provide insights into new compounds' activities.
Software link: Canvas - Cheminformatics Software