Drug discovery workflow setups often require a generator for creating libraries of chemical compounds with prefered features such as, e.g., certain physicochemical properties (e.g., polar surface area, logP, molecular weight, and others). Such a generator should be aware of chemical rules, and should be able to create novel compounds and search chemical molecule space effectively. Finally, it should fit well into a workflow which requires seamless integration with other software tools processing the molecules generated. 

CIDRUX’ Molecule Commander^TM, a command-line driven library generator and molecule optimizer, satisfies all of those requirements. The Molecule Commander^TM software package generates libraries of any size, with physicochemical properties as requested by the user. 

The Molecule Commander^TM uses powerful evolutionary algorithms for creating and optimizing molecule structures. It combines the evolutionary search for molecular innovations with automatic options for creating molecules with certain physicochemical properties, such as polar surface area, logP, solubility, molecular weight, and others. Through an iterative process of molecule creation, selection, modification and refinement the system generates a library of candidate molecules with the desired properties.

The Molecule Commander^TM is invoked from the command line, as shown in the left figure above. By specifying the range or the target values for the physicochemical properties, the user requires the Molecule Commander^TM to generate compounds with property values in that range. An example is shown in the right figure above, where a quick run of the Molecule Commander^TM found the compounds shown. Of course, the total number of structures generated can be arbitrarily large and be chosen by the user.

Key features:

·    Automatic generation of NCEs using evolutionary principles (mutation and recombination of molecule structures) ·    Automatic library creation, either based on existing structures or completely De Novo ·    Intelligent operators for generating only chemically feasible molecules ·    Automatic calculation of physicochemical properties such as polar surface area (PSA), logP, molecular weight, number of hydrogen donors and acceptors, number of rotatable bonds, and numbers of aromatic systems and substituents ·    Optional automatic filtering of molecules to generate only molecules according to certain chemical rules (Bredt’s rule, avoidance of certain unstable structures)

·    Command-line based user interface, facilitating seamless integration into drug discovery workflows ·    Import and export of molecular structures and libraries using standard file formats (.mol and .sd file format) for easy export to other software and databases

·    Knowledge-based mutation and recombination of molecules ·    Efficient search through chemical space

·    Automatic library generation by evolutionary search ·    Automatic optimization of molecules towards specified properties ·    “Front-load” program for any platform that incorporates software components into a drug discovery workflow