Adapting CHMTRN (CHeMistry TRaNslator) for a New Use.

Area of Science:

• Computational Chemistry
• Cheminformatics
• Organic Synthesis

Background:

• The CHMTRN language, developed for the LHASA project, was originally designed for describing chemical retro-reactions.
• Existing computational tools for synthetic route design often focus on retrosynthesis.
• There is a need for versatile chemical languages applicable to both forward and retro-synthetic analysis.

Purpose of the Study:

• To adapt and extend the CHMTRN and PATRAN languages for computational chemistry.
• To develop a knowledge base for generating a large database of synthetically accessible, drug-like chemical structures.
• To demonstrate the utility of CHMTRN and PATRAN beyond retrosynthetic route design.

Main Methods:

• Adoption and extension of the CHMTRN language.
• Integration of CHMTRN with the PATRAN chemical pattern description language.
• Development of a computer program utilizing these languages to build the SAVI Database.

Main Results:

• Successful generation of a large database of synthetically accessible, drug-like chemical structures (SAVI Database).
• Demonstrated the capability of the reimplemented CHMTRN and PATRAN languages to describe both forward and retro-reactions.
• Extended the applicability of CHMTRN and PATRAN for broader uses in chemistry.

Conclusions:

• The extended CHMTRN and PATRAN languages provide a robust framework for computational chemistry.
• The SAVI Database represents a significant resource for drug discovery and chemical synthesis.
• These chemical languages have broad potential for various applications in chemistry, including forward-synthetic design.