Ultra Large Library Virtual Screening in MolSoft ICM


MolSoft's highly accurate virtual screening methodology, in conjunction with the computing capabilities of GPU technology, empowers you to conduct efficient and precise screenings of ultra-large giga-sized chemical databases. Our diverse range of approaches caters to your specific requirements, taking into account your initial starting point of structure and/or ligand based data, available libraries, and computing resources.

The table below summarizes the different options available.

Ligand-Based Ultra-Large Library Screening

Method Approach Target Structure(s) Required High Affinity Lead(s) Speed Notes
RIDE Ligand-Based No Yes ~1.5M confs/sec on latest GPU RTX 4090 Ligand-based screening approach using MolSoft's 3D Pharmacophore method Atomic Property Fields Any prepared giga size library in molt format can be screened.
RIDE-Dock Ligand- and Structure-based hybrid Yes Yes Speed optimized by RIDE Ligand-based RIDE screen followed by Structure-based virtual ligand screening of top scoring chemicals.

Structure-Based Ultra Large Library Screening

Method Approach Target Structure(s) Required High Affinity Lead(s) Speed Notes
V-SYNTHES + ICM-VLS Fragment-based and enumeration Yes No ~2 weeks with a 250 VLS Cluster License Efficiently screen the 21B Enamine Real Space Set using an initial fragment-based filter followed by enumeration and structure-based docking.
RIDGE BETA Structure-based Yes No 100 chemicals/sec on latest GPU RTX 4090 Rapid structure-based GPU docking engine. Any prepared giga sized library in molt format can be screened.
Giga Screen Deep Learning Yes No Speed optimized by RIDGE The GigaScreen method combines machine learning and deep learning tools to tackle the computational intensity of screening very large chemical databases. Any prepared giga sized library in molt format can be screened.

Available databases include the following which are available as a free download here.