Inverse Potts model improves accuracy of phylogenetic profiling

Area of Science:

• Computational biology
• Bioinformatics
• Genomics

Background:

• Phylogenetic profiling is crucial for identifying gene functions.
• Conventional methods suffer from evolutionary and spurious correlation biases.
• Reducing spurious correlation bias in phylogenetic profiling remains a challenge.

Purpose of the Study:

• To develop novel metrics for phylogenetic profiling that reduce spurious correlation bias.
• To improve the accuracy of gene function prediction using these new metrics.
• To evaluate the performance of inverse Potts model (IPM)-based metrics.

Main Methods:

• Developed inverse Potts model (IPM)-based metrics for phylogenetic profiling.
• Created a combined metric integrating IPM and phylogenetic tree information.
• Analyzed empirical datasets to assess prediction performance.

Main Results:

• IPM-based metrics significantly improved prediction performance in phylogenetic profiling.
• A metric combining IPM and phylogenetic tree information showed enhanced accuracy.
• Integrating multiple metrics, including IPM-based ones, outperformed single-metric approaches.

Conclusions:

• Inverse Potts model-based metrics effectively reduce spurious correlation bias in phylogenetic profiling.
• Combined metrics offer superior performance for gene function prediction.
• The developed methods provide a more robust approach to functional genomics.