Ubunit of the18S rRNA exist in several copies (7) in the Plasmodium genome. One of the major advantages of the method previously reported by Shokoples et al. [7] over other approaches [13,18,19], is that the primers designed target all copies thus increasing the inhibitor sensitivity of the reaction. The use of inhibitor species-specific oligonucleotides that could accurately detect all four malaria-causing Plasmodium species (Pf, Pm, Po and Pv) without significant competition between the oligonucleotides designed for the different templates was one of the majorFigure 3. Absolute and relative quantification of Plasmodium DNA in mosquitoes. This figure shows a not significant difference was observed in the P. falciparum densities between the two Anopheles species (P-value = 0, 2197). doi:10.1371/journal.pone.0052719.gadvantages of this approach. Here, the multiplexing of the reaction was optimized for the simultaneous detection of the four Plasmodium species at a time in two reaction tubes. This method was tested on plasmid preparations and showed good amplification efficiencies (E.90 ). We also noticed a good sensitivity with the ability of detecting and quantifying down to 10 copies of Plasmodium 18S rDNA in 5 mL DNA used per reaction, meaning that at least 200 copies, approximately 30 sporozoites, are necessary in DNA preparation for a positive reaction to be quantified. Targets copy number detected below this threshold were still considered positive but unquantifiable as this falls outside the linearity range of external standards. The specificity of the real-time PCR was demonstrated by the absence of cross-reactivity between different primer-probe systems on artificial mixtures of plasmid preparations. The analytical sensitivity of the assays for P. malariae, P. ovale and P. vivax as the minor species in cases of mixed infection with P. falciparum showed, like in the data reported by Shokoples et al. [7] that we could reproducibly detect minors populations at a greater fold down to 1:1000 ratio. This performance was optimized by the formulation of the multiplexing that we have defined (Plasmo/Pf and Pm/Po). With these modifications, we implemented this assay as a confirmatory test for malaria species identification in anopheline vectors (An. gambiae and An. funestus). Plasmodium DNA was consistently amplified from frozen mosquito homogenates initially prepared for ELISA. This suggests that parasite target DNA will likely remain detectable by PCR in mosquito homogenates for longer periods, from the time they are stored at 220uC. This preservation condition of fieldcollections for subsequent target PCR-detection of Plasmodium DNA is 24786787 highly amenable to field work and does not seem to promote biological degradation processes which are favored by the release of nucleases after grinding. In comparison with traditional ELISA-CSP; the real-time PCR assay 
was more useful for the identification of Plasmodium species in the vectors. From the 70 positive mosquitoes for P. falciparum by ELISA-CSP the presence of Plasmodium could be PCR-confirmed in 62 samples. Of important diagnostic significance, 11 samples were misdiagnosed by ELISACSP. Among these, 8 samples that were positive by ELISA-CSP were not confirmed by real-time PCR. The absence of Plasmodium DNA was further ascertained in those samples by using the conventional nested PCR described by Snounou et al [14]. These results are concordant with the hypothesis that ELISA-CSP may be compromised by overdiagnosi.Ubunit of the18S rRNA exist in several copies (7) in the Plasmodium genome. One of the major advantages of the method previously reported by Shokoples et al. [7] over other approaches [13,18,19], is that the primers designed target all copies thus increasing the sensitivity of the reaction. The use of species-specific oligonucleotides that could accurately detect all four malaria-causing Plasmodium species (Pf, Pm, Po and Pv) without significant competition between the oligonucleotides designed for the different templates was one of the majorFigure 3. Absolute and relative quantification of Plasmodium DNA in mosquitoes. This figure shows a not significant difference was observed in the P. falciparum densities between the two Anopheles species (P-value = 0, 2197). doi:10.1371/journal.pone.0052719.gadvantages of this approach. Here, the multiplexing of the reaction was optimized for the simultaneous detection of the four Plasmodium species at a time in two reaction tubes. This method was tested on plasmid preparations and showed good amplification efficiencies (E.90 ). We also noticed a good sensitivity with the ability of detecting and quantifying down to 10 copies of Plasmodium 18S rDNA in 5 mL DNA used per reaction, meaning that at least 200 copies, approximately 30 sporozoites, are necessary in DNA preparation for a positive reaction to be quantified. Targets copy number detected below this threshold were still considered positive 
but unquantifiable as this falls outside the linearity range of external standards. The specificity of the real-time PCR was demonstrated by the absence of cross-reactivity between different primer-probe systems on artificial mixtures of plasmid preparations. The analytical sensitivity of the assays for P. malariae, P. ovale and P. vivax as the minor species in cases of mixed infection with P. falciparum showed, like in the data reported by Shokoples et al. [7] that we could reproducibly detect minors populations at a greater fold down to 1:1000 ratio. This performance was optimized by the formulation of the multiplexing that we have defined (Plasmo/Pf and Pm/Po). With these modifications, we implemented this assay as a confirmatory test for malaria species identification in anopheline vectors (An. gambiae and An. funestus). Plasmodium DNA was consistently amplified from frozen mosquito homogenates initially prepared for ELISA. This suggests that parasite target DNA will likely remain detectable by PCR in mosquito homogenates for longer periods, from the time they are stored at 220uC. This preservation condition of fieldcollections for subsequent target PCR-detection of Plasmodium DNA is 24786787 highly amenable to field work and does not seem to promote biological degradation processes which are favored by the release of nucleases after grinding. In comparison with traditional ELISA-CSP; the real-time PCR assay was more useful for the identification of Plasmodium species in the vectors. From the 70 positive mosquitoes for P. falciparum by ELISA-CSP the presence of Plasmodium could be PCR-confirmed in 62 samples. Of important diagnostic significance, 11 samples were misdiagnosed by ELISACSP. Among these, 8 samples that were positive by ELISA-CSP were not confirmed by real-time PCR. The absence of Plasmodium DNA was further ascertained in those samples by using the conventional nested PCR described by Snounou et al [14]. These results are concordant with the hypothesis that ELISA-CSP may be compromised by overdiagnosi.
