Table 1 The advantages and limitations of bioinformatics tools current available for analysis of polyclonal infections and estimation of MOI
From: Molecular approaches to determine the multiplicity of Plasmodium infections
Software/approach | Implementation | Data source | Advantage | Limitation | References |
|---|---|---|---|---|---|
Fws | Fws statistics | SNPs | Provide a summary statistic of relative inbreeding and correlation with MOI | Does not provide a direct estimate of COI | Manske et al. [108] |
estMOI | Perl scripts | TDS | estimate MOI, SNP combinations and proportion in a specific genomic region | Partial genomic region, sufficient density of SNPs in an amplicon is required | Assefa et al. [107] |
COIL | Perl scripts web interface | SNPs | Use genome-wide SNPs to infer the number of strains and their proportions | Does not infer haplotypes; sequence read counts are assumed to be unbiased and the SNPs are unlinked | Galinsky et al. [93] |
SeekDeep | Perl scripts | TDS | No reference sequence required, direct estimation of amplicon haplotype cluster and frequency | Partial genomic region, threshold for haplotype cluster calls needs to be determined by empirical methods in each study | Hathaway et al. [104] |
HaplotypR | R package | TDS | Direct estimation of haplotype cluster and frequency based quality filtering of sequencing reads | Partial genomic region, enough high polymorphism in the amplicon is required and unable to deal with indel of amplicon | Lerch et al. [102] |
pfmix | R package | SNPs | Infer inbreeding coefficients using WGS to assess within-sample parasite infections | Does not infer haplotypes, sequence read counts are assumed to be unbiased and SNPs need to be unlinked | O’Brien et al. [110] |
THE REAL McCOIL | Perl scripts web interface | SNPs | Infer the number of strains and their proportions | Does not infer haplotypes, sequence read counts are assumed to be unbiased and SNPs need to be unlinked | Chang et al. [94] |
DEploid | R package | SNPs | Estimate the number of strains, their relative proportions, and the haplotypes | Reference panel required, more reference strains are better. Multiallelic variant not considered | Zhu et al. [111] |