Phylogenomics to structure: evolutionary and clinical signals in the TP53 DNA-binding core through LOOCV-validated...

Area of Science:

• Evolutionary genomics
• Structural biology
• Cancer genomics

Background:

• TP53 is a critical tumor suppressor gene.
• Understanding evolutionary constraints on TP53 is essential for interpreting pathogenic variants and their clinical significance.
• Identifying functionally critical residues aids in understanding cancer development and potential therapeutic strategies.

Purpose of the Study:

• To systematically prioritize functionally critical residues in the TP53 gene.
• To integrate evolutionary genomic signals, structural information, and clinical variant data.
• To develop a generalizable computational framework for residue-level prioritization in cancer genes.

Main Methods:

• Genome-wide alignment and phylogenetic estimation using fixed effects likelihood and fast unconstrained Bayesian approximation across 19 vertebrate orthologs.
• Mapping evolutionary selection signals onto AlphaFold-predicted TP53 structures.
• Integrating human variant data from ClinVar and UniProt with structural and evolutionary information.
• Machine learning validation using leave-one-out cross-validation (LOOCV) for predictive performance assessment.

Main Results:

• Identified five TP53 sites under positive or diversifying selection, with human codon 129 consistently detected by multiple methods.
• Pathogenic variants cluster at the DNA-binding interface, with residues 239-248 identified as a primary pathogenic hotspot.
• A composite scoring system integrating evolutionary constraint, pathogenic burden, and domain importance effectively prioritized critical residues.
• Machine learning models demonstrated robust predictive performance, with ensemble methods achieving high accuracy in phylogenetic-distance regression and clade classification.

Conclusions:

• Evolutionary signals and clinical variants converge within the DNA-binding core of TP53.
• Codon 129 is a significant positive-selection site, and residues 239-248 represent a major pathogenic hotspot.
• The developed AlphaFold-anchored, LOOCV-validated framework provides a systematic approach for residue prioritization.
• This framework can guide mechanistic studies and inform precision oncology applications pending experimental validation.