ProbOnto: ontology and knowledge base of probability distributions

Area of Science:

• Mathematical and Statistical Modeling
• Computational Biology
• Bioinformatics

Background:

• Probability distributions are fundamental to mathematical and statistical modeling.
• Existing resources lack a detailed ontology and programmatic access for probability distributions.
• A unified reference is needed for encoding, annotating, and exchanging probability models.

Purpose of the Study:

• To develop ProbOnto, an ontology-based knowledge base for probability distributions.
• To provide a comprehensive resource for defining and characterizing probability distributions.
• To facilitate the annotation and encoding of distribution-based models.

Main Methods:

• Developed an ontology-based knowledge base named ProbOnto.
• Included over 80 uni- and multivariate probability distributions.
• Detailed defining functions, characteristics, relationships, and re-parameterization formulas for each distribution.

Main Results:

• ProbOnto contains more than 80 probability distributions.
• The knowledge base includes defining functions, characteristics, and relationships.
• Re-parameterization formulas are provided for enhanced usability.

Conclusions:

• ProbOnto serves as a valuable resource for probability distributions.
• It simplifies model annotation and the encoding of distribution-based models.
• Enables programmatic access to detailed information on probability distributions.