Statistical Inference and Biological Interpretation via Comparatively Realistic Models of Molecular Evolution

Gene Ontology;MCMC;Bayes factor;indel models;protein structure impact;phenotype-genotype mapping;molecular fitness;population genetic interpretation;ancestral allele prediction

Choi, Sang Chul ; David M. Bird, Committee Member,Eric A. Stone, Committee Member,Jeffrey L. Thorne, Committee Chair,Brian M. Wiegmann, Committee Member,Choi, Sang Chul ; David M. Bird ; Committee Member ; Eric A. Stone ; Committee Member ; Jeffrey L. Thorne ; Committee Chair ; Brian M. Wiegmann ; Committee Member

University:North Carolina State University

关键词: Gene Ontology;    MCMC;    Bayes factor;    indel models;    protein structure impact;    phenotype-genotype mapping;    molecular fitness;    population genetic interpretation;    ancestral allele prediction;   

Others  :  https://repository.lib.ncsu.edu/bitstream/handle/1840.16/5865/etd.pdf?sequence=1&isAllowed=y

美国|英语

来源: null

【 摘 要 】

Recently, advances in statistical inference techniques haveallowed analyses of molecular evolution to proceed without the biologicallyimplausible assumption of independent change among DNA sequence sites.These techniques permit incorporation of molecular phenotypes suchas RNA secondary and protein tertiary structure directly into themodels of DNA sequence evolution, and they thereby facilitate assessmentof the impact of molecular phenotype on the rates of sequence evolution.Our analysis of 1,195 non-redundant protein-coding sequences suggeststhat solvent accessibility and pairwise interactions among amino acidshave important and roughly comparable impacts on the rates of evolution.We show how solvent accessibility and pairwise amino acid interactionscan be used with protein-coding single nucleotide polymorphism (SNP)data to predict which SNP allele is ancestral and which is derived.Our analysis of 142 non-synonymous SNPs indicates that ancestral allelesare more selectively advantageous with respect to tertiary structurethan are derived alleles. In other work, we show how recently developedmodels of molecular evolution with dependent change among sites canbe adapted to generate stationary distributions that match a desiredvariable length Markov model or profile hidden Markov model for proteinsequence organization. Departures between a neutral model for proteinevolution and the variable length Markov model or profile hidden Markovmodel are attributed to natural selection. We show how these departureslead to a crude approximation of the product of effective populationsize and the difference in relative fitnesses between sequences.