Use this URL to cite or link to this record in EThOS:
Title: Natural selection on two linked loci : Wright-Fisher perspectives and applications
Author: He, Zhangyi
ISNI:       0000 0004 6057 8592
Awarding Body: University of Bristol
Current Institution: University of Bristol
Date of Award: 2016
Availability of Full Text:
Access from EThOS:
In the age of next-generation sequencing, an increasing amount of high-quality genomic time-series data is becoming available. This has created new opportunities to provide a more precise inference of population genetic parameters and more accurate hypotheses-testing about the recent action of natural selection. However, most existing population genetic inference procedures are only applicable to one-locus problems or multi-locus problems assuming no linkage between different loci due to the difficulty of computing the likelihood of the time-series data of allele frequencies sampled from multiple linked loci. To address this, we propose a hidden Markov model to characterise the time-series data of allele frequencies and develop an efficient simulated based method to evaluate the likelihood. In the hidden Markov model, the underlying population is assumed to evolve under natural selection at two linked loci according to the Wright-Fisher diffusion, and the observations are implicitly modelled through independent multinomial sampling from the underlying population at each given time point. We evaluate the transition probability density of the Wright-Fisher diffusion and the likelihood of the time-series data of allele frequencies sampled from two linked loci through importance sampling. Our simulation-based method enables us to jointly estimate the selection coefficient, recombination rate, and other population genetic quantities of interest from the two-locus time-series data of allele frequencies and can be naturally extended to analyse the multi-locus time-series data of allele frequencies.
Supervisor: Not available Sponsor: Not available
Qualification Name: Thesis (Ph.D.) Qualification Level: Doctoral
EThOS ID:  DOI: Not available