Small Molecule Identification with MOLGEN and Mass Spectrometry

Area of Science:

• Computational chemistry
• Cheminformatics
• Structure elucidation

Background:

• The CASMI (Critical Assessment of Small Molecule Identification) contest benchmarks computational methods for identifying chemical structures.
• Structure generation approaches are crucial for predicting molecular structures from experimental data.

Purpose of the Study:

• To detail the application of MOLGEN software for small molecule identification in the 2012 CASMI contest.
• To demonstrate the utility of structure generation approaches in chemical structure elucidation.
• To provide insights into optimizing MOLGEN settings for different identification categories.

Main Methods:

• Utilized various MOLGEN programs (MOLGEN-MS/MS, MOLGEN 3.5, MOLGEN 5.0, MOLGEN-MS) tailored to specific CASMI categories.
• Focused on Categories 1 and 2, which had the highest participant numbers.
• Described detailed settings and justifications for their selection.
• Employed evaluations to contextualize the obtained results.

Main Results:

• MOLGEN entries were submitted for multiple CASMI categories, demonstrating structure generation capabilities.
• Detailed descriptions of program settings and their rationale were provided.
• Results were analyzed and presented in perspective through various evaluations.
• The semi-automated nature of the approaches and the impact of user experience were noted.

Conclusions:

• Structure generation approaches, exemplified by MOLGEN, offer a viable pathway for unknown small molecule identification.
• User experience significantly influences the performance of these semi-automated methods.
• The presented insights can guide users in applying MOLGEN for enhanced structure elucidation in similar challenges.