Rebecca M Mitchell1,2,3, Zhiyong Zhou1, Mili Sheth4, Sheila Sergent1, Michael Frace4, Vishal Nayak5, Bin Hu5, John Gimnig1, Feiko Ter Kuile6, Kim Lindblade1, Laurence Slutsker1, Mary J Hamel1, Meghna Desai1, Kephas Otieno7, Simon Kariuki7, Ymir Vigfusson8, Ya Ping Shi9. 1. Division of Parasitic Diseases, Center for Global Health, Centers for Disease Control and Prevention (CDC), Atlanta, USA. 2. Department of Computer Science, Emory University, Atlanta, USA. 3. School of Nursing, Emory University, Atlanta, USA. 4. Biotechnology Core Facility Branch, Division of Scientific Resources, CDC, Atlanta, USA. 5. Office of Infectious Diseases, National Center for Emerging and Zoonotic Infectious Diseases, CDC, Atlanta, USA. 6. Liverpool School of Tropical Medicine, Liverpool, UK. 7. Kenya Medical Research Institute, Centre for Global Health Research, Kisumu, Kenya. 8. Department of Computer Science, Emory University, Atlanta, USA. ymir.vigfusson@emory.edu. 9. Division of Parasitic Diseases, Center for Global Health, Centers for Disease Control and Prevention (CDC), Atlanta, USA. yps0@cdc.gov.
Abstract
BACKGROUND: Simultaneous infection with multiple malaria parasite strains is common in high transmission areas. Quantifying the number of strains per host, or the multiplicity of infection (MOI), provides additional parasite indices for assessing transmission levels but it is challenging to measure accurately with current tools. This paper presents new laboratory and analytical methods for estimating the MOI of Plasmodium falciparum. METHODS: Based on 24 single nucleotide polymorphisms (SNPs) previously identified as stable, unlinked targets across 12 of the 14 chromosomes within P. falciparum genome, three multiplex PCRs of short target regions and subsequent next generation sequencing (NGS) of the amplicons were developed. A bioinformatics pipeline including B4Screening pathway removed spurious amplicons to ensure consistent frequency calls at each SNP location, compiled amplicons by SNP site diversity, and performed algorithmic haplotype and strain reconstruction. The pipeline was validated by 108 samples generated from cultured-laboratory strain mixtures in different proportions and concentrations, with and without pre-amplification, and using whole blood and dried blood spots (DBS). The pipeline was applied to 273 smear-positive samples from surveys conducted in western Kenya, then providing results into StrainRecon Thresholding for Infection Multiplicity (STIM), a novel MOI estimator. RESULTS: The 24 barcode SNPs were successfully identified uniformly across the 12 chromosomes of P. falciparum in a sample using the pipeline. Pre-amplification and parasite concentration, while non-linearly associated with SNP read depth, did not influence the SNP frequency calls. Based on consistent SNP frequency calls at targeted locations, the algorithmic strain reconstruction for each laboratory-mixed sample had 98.5% accuracy in dominant strains. STIM detected up to 5 strains in field samples from western Kenya and showed declining MOI over time (q < 0.02), from 4.32 strains per infected person in 1996 to 4.01, 3.56 and 3.35 in 2001, 2007 and 2012, and a reduction in the proportion of samples with 5 strains from 57% in 1996 to 18% in 2012. CONCLUSION: The combined approach of new multiplex PCRs and NGS, the unique bioinformatics pipeline and STIM could identify 24 barcode SNPs of P. falciparum correctly and consistently. The methodology could be applied to field samples to reliably measure temporal changes in MOI.
BACKGROUND: Simultaneous infection with multiple malaria parasite strains is common in high transmission areas. Quantifying the number of strains per host, or the multiplicity of infection (MOI), provides additional parasite indices for assessing transmission levels but it is challenging to measure accurately with current tools. This paper presents new laboratory and analytical methods for estimating the MOI of Plasmodium falciparum. METHODS: Based on 24 single nucleotide polymorphisms (SNPs) previously identified as stable, unlinked targets across 12 of the 14 chromosomes within P. falciparum genome, three multiplex PCRs of short target regions and subsequent next generation sequencing (NGS) of the amplicons were developed. A bioinformatics pipeline including B4Screening pathway removed spurious amplicons to ensure consistent frequency calls at each SNP location, compiled amplicons by SNP site diversity, and performed algorithmic haplotype and strain reconstruction. The pipeline was validated by 108 samples generated from cultured-laboratory strain mixtures in different proportions and concentrations, with and without pre-amplification, and using whole blood and dried blood spots (DBS). The pipeline was applied to 273 smear-positive samples from surveys conducted in western Kenya, then providing results into StrainRecon Thresholding for Infection Multiplicity (STIM), a novel MOI estimator. RESULTS: The 24 barcode SNPs were successfully identified uniformly across the 12 chromosomes of P. falciparum in a sample using the pipeline. Pre-amplification and parasite concentration, while non-linearly associated with SNP read depth, did not influence the SNP frequency calls. Based on consistent SNP frequency calls at targeted locations, the algorithmic strain reconstruction for each laboratory-mixed sample had 98.5% accuracy in dominant strains. STIM detected up to 5 strains in field samples from western Kenya and showed declining MOI over time (q < 0.02), from 4.32 strains per infected person in 1996 to 4.01, 3.56 and 3.35 in 2001, 2007 and 2012, and a reduction in the proportion of samples with 5 strains from 57% in 1996 to 18% in 2012. CONCLUSION: The combined approach of new multiplex PCRs and NGS, the unique bioinformatics pipeline and STIM could identify 24 barcode SNPs of P. falciparum correctly and consistently. The methodology could be applied to field samples to reliably measure temporal changes in MOI.
Entities:
Keywords:
Haplotype and strain; MOI estimation; Multiplicity of infection; Plasmodium falciparum; STIM; StrainRecon; Transmission
Authors: Maxwell Kilama; David L Smith; Robert Hutchinson; Ruth Kigozi; Adoke Yeka; Geoff Lavoy; Moses R Kamya; Sarah G Staedke; Martin J Donnelly; Chris Drakeley; Bryan Greenhouse; Grant Dorsey; Steve W Lindsay Journal: Malar J Date: 2014-03-21 Impact factor: 2.979
Authors: Wesley Wong; Allison D Griggs; Rachel F Daniels; Stephen F Schaffner; Daouda Ndiaye; Amy K Bei; Awa B Deme; Bronwyn MacInnis; Sarah K Volkman; Daniel L Hartl; Daniel E Neafsey; Dyann F Wirth Journal: Genome Med Date: 2017-01-24 Impact factor: 11.117
Authors: Standwell C Nkhoma; Rachel L Banda; Stanley Khoswe; Tamika J Dzoole-Mwale; Stephen A Ward Journal: Infect Genet Evol Date: 2018-08-23 Impact factor: 3.342
Authors: Emily LaVerriere; Philipp Schwabl; Manuela Carrasquilla; Aimee R Taylor; Zachary M Johnson; Meg Shieh; Ruchit Panchal; Timothy J Straub; Rebecca Kuzma; Sean Watson; Caroline O Buckee; Carolina M Andrade; Silvia Portugal; Peter D Crompton; Boubacar Traore; Julian C Rayner; Vladimir Corredor; Kashana James; Horace Cox; Angela M Early; Bronwyn L MacInnis; Daniel E Neafsey Journal: Mol Ecol Resour Date: 2022-05-03 Impact factor: 8.678