Data-driven extraction of relative reasoning rules to limit combinatorial explosion in biodegradation pathway...

Area of Science:

• Environmental chemistry
• Computational toxicology
• Biotechnology

Background:

• The University of Minnesota Pathway Prediction System (UM-PPS) is a rule-based expert system for predicting organic compound biodegradation pathways.
• Iterative rule application in UM-PPS leads to combinatorial explosion, hindering efficient pathway prediction.
• Existing methods require improvement to manage the complexity of biodegradation pathway generation.

Purpose of the Study:

• To develop and implement relative reasoning rules to prioritize biotransformations within the UM-PPS.
• To mitigate the combinatorial explosion issue in biodegradation pathway prediction.
• To enhance the efficiency and accuracy of predicting environmental fate of organic compounds.

Main Methods:

• Data from known biotransformation pathways were used to derive 112 relative reasoning rules.
• These rules were integrated into the UM-PPS to guide prediction steps.
• The system's performance was evaluated using internal and external validation datasets.

Main Results:

• Relative reasoning rules reduced predicted biotransformations by over 25% for rule-generation compounds and ~15% for external xenobiotics.
• The accuracy of predicting known products remained high at 75% with the implemented rules.
• The number of predicted transformation products was effectively reduced without sacrificing prediction quality.

Conclusions:

• Relative reasoning rules successfully address the combinatorial explosion problem in UM-PPS.
• The enhanced UM-PPS provides more efficient and accurate biodegradation pathway predictions.
• This approach improves the prediction of environmental fate and transformation products for various organic compounds.