Malaria haplotype frequency estimation

Abstract

We present a Bayesian approach for estimating the relative frequencies of multi-single nucleotide polymorphism (SNP) haplotypes in populations of the malaria parasite Plasmodium falciparum by using microarray SNP data from human blood samples. Each sample comes from a malaria patient and contains one or several parasite clones that may genetically differ. Samples containing multiple parasite clones with different genetic markers pose a special challenge. The situation is comparable with a polyploid organism. The data from each blood sample indicates whether the parasites in the blood carry a mutant or a wildtype allele at various selected genomic positions. If both mutant and wildtype alleles are detected at a given position in a multiply infected sample, the data indicates the presence of both alleles, but the ratio is unknown. Thus, the data only partially reveals which specific combinations of genetic markers (i.e. haplotypes across the examined SNPs) occur in distinct parasite clones. In addition, SNP data may contain errors at non-negligible rates. We use a multinomial mixture model with partially missing observations to represent this data and a Markov chain Monte Carlo method to estimate the haplotype frequencies in a population. Our approach addresses both challenges, multiple infections and data errors.

Estimation of malaria haplotype and genotype frequencies: a statistical approach to overcome the challenge associated with multiclonal infections

“…First, in contrast to all previously published methods, the model makes use of all available data, including those that are incomplete due to unsuccessful genotyping outcomes or study design. Second, in contrast to the recently published Bayesian method [26] and the model underpinning the freely available online software MalHaploFreq [11], this model is not reliant upon per-patient measurements of the MOI. Third, in contrast to some existing approaches [11,24,27], it enables rapid analysis of data from three or more markers.…”

“…Third, in contrast to some existing approaches [11,24,27], it enables rapid analysis of data from three or more markers. Superior assumptions regarding detectability and experimental error are incorporated into alternative models [11,26]. It is especially important to take into account the suboptimal detectability of minority clones, which was addressed by Hastings and colleagues using an indicator function [11,22], when the data are derived from polymerase chain reaction (PCR) methods and individual per-patient MOI measurements are regarded as fixed.…”

“…In fact, the majority of statistical methods use maximum likelihood estimation to generate point estimates of frequencies and accompanying confidence intervals [11,24,25,27]. Alternatively, a Bayesian framework, such as that developed by Wigger et al [26], can be used to impute the unobserved variables, allow the propagation of uncertainty and enable the incorporation of specialist knowledge [28,29]. Since the aforementioned Bayesian model is reliant upon patient-level measurements of the MOI, an alternative Bayesian model, which is not reliant upon patient-level MOI measurements, was developed.…”

Markov chain Monte Carlo and expectation maximization approaches for estimation of haplotype frequencies for multiply infected human blood samples

“…Simple counting of the number of different alleles at each loci provides a minimum MOI. However, this may underestimate the population MOI if clones share alleles at hyper-variable loci purely by chance, or if they are low density clones missed during genotyping [3]. …”

On the Effects of Malaria Treatment on Parasite Drug Resistance – Probability Modelling of Genotyped Malaria Infections

“…Wigger et al [1] uses Markov chain Monte Carlo (MCMC) methods where one step simulates the true state. Hastings and Smith [2], present a computer package for the calculations.…”