Limits...
Identification of cryptic Anopheles mosquito species by molecular protein profiling.

Müller P, Pflüger V, Wittwer M, Ziegler D, Chandre F, Simard F, Lengeler C - PLoS ONE (2013)

Bottom Line: The approach also classifies specimens from different laboratory colonies; hence proving also very promising for its use in colony authentication as part of quality assurance in laboratory studies.While being exceptionally accurate and robust, MALDI-TOF MS has several advantages over other typing methods, including simple sample preparation and short processing time.As the method does not require DNA sequence information, data can also be reviewed at any later stage for diagnostic or functional patterns without the need for re-designing and re-processing biological material.

View Article: PubMed Central - PubMed

Affiliation: Department of Medical Services and Diagnostic, Swiss Tropical and Public Health Institute, Basel, Switzerland. pie.mueller@unibas.ch

ABSTRACT
Vector control is the mainstay of malaria control programmes. Successful vector control profoundly relies on accurate information on the target mosquito populations in order to choose the most appropriate intervention for a given mosquito species and to monitor its impact. An impediment to identify mosquito species is the existence of morphologically identical sibling species that play different roles in the transmission of pathogens and parasites. Currently PCR diagnostics are used to distinguish between sibling species. PCR based methods are, however, expensive, time-consuming and their development requires a priori DNA sequence information. Here, we evaluated an inexpensive molecular proteomics approach for Anopheles species: matrix assisted laser desorption/ionization time-of-flight mass spectrometry (MALDI-TOF MS). MALDI-TOF MS is a well developed protein profiling tool for the identification of microorganisms but so far has received little attention as a diagnostic tool in entomology. We measured MS spectra from specimens of 32 laboratory colonies and 2 field populations representing 12 Anopheles species including the A. gambiae species complex. An important step in the study was the advancement and implementation of a bioinformatics approach improving the resolution over previously applied cluster analysis. Borrowing tools for linear discriminant analysis from genomics, MALDI-TOF MS accurately identified taxonomically closely related mosquito species, including the separation between the M and S molecular forms of A. gambiae sensu stricto. The approach also classifies specimens from different laboratory colonies; hence proving also very promising for its use in colony authentication as part of quality assurance in laboratory studies. While being exceptionally accurate and robust, MALDI-TOF MS has several advantages over other typing methods, including simple sample preparation and short processing time. As the method does not require DNA sequence information, data can also be reviewed at any later stage for diagnostic or functional patterns without the need for re-designing and re-processing biological material.

Show MeSH

Related in: MedlinePlus

Model selection and cross-validation to discriminate between species of the Anopheles gambiae complex (Model 1).(A) The graph shows the error rate from the cross-validation plotted as a function of the number of the ranked peaks included in the SDA model that discriminates between members of the A. gambiae species complex. The peaks were ranked (left to right) according to the correlation-adjusted t-scores (CAT scores). The vertical, red line shows the 68 peaks chosen for the SDA model. (B) List with the 68 ranked peaks (top equals highest rank) their corresponding CAT scores. The length and direction of the horizontal blue bars represents the CAT scores of the centroid versus the pooled mean and show the influence of a particular peak in differentiating between the groups (Table S2). For example the top peak, M12369.6 has a strong influence in separating A. merus from all the other species, emphasised by the length of the bar and the opposite direction from the bars of the other species. In contrast, the tenth peak, M12527.3 has a stronger influence in separating A. gambiae s.s. from A. arabiensis. AR: A. arabiensis; GA: A. gambiae s.s.; ME: A. merus; QD: A. quadriannulatus.
© Copyright Policy
Related In: Results  -  Collection


getmorefigures.php?uid=PMC3585343&req=5

pone-0057486-g003: Model selection and cross-validation to discriminate between species of the Anopheles gambiae complex (Model 1).(A) The graph shows the error rate from the cross-validation plotted as a function of the number of the ranked peaks included in the SDA model that discriminates between members of the A. gambiae species complex. The peaks were ranked (left to right) according to the correlation-adjusted t-scores (CAT scores). The vertical, red line shows the 68 peaks chosen for the SDA model. (B) List with the 68 ranked peaks (top equals highest rank) their corresponding CAT scores. The length and direction of the horizontal blue bars represents the CAT scores of the centroid versus the pooled mean and show the influence of a particular peak in differentiating between the groups (Table S2). For example the top peak, M12369.6 has a strong influence in separating A. merus from all the other species, emphasised by the length of the bar and the opposite direction from the bars of the other species. In contrast, the tenth peak, M12527.3 has a stronger influence in separating A. gambiae s.s. from A. arabiensis. AR: A. arabiensis; GA: A. gambiae s.s.; ME: A. merus; QD: A. quadriannulatus.

Mentions: In an attempt to overcome the poor performance of the unsupervised cluster analysis in discriminating between the A. gambiae sibling species, a SDA classification model was evaluated as an alternative. The model (Model 1) was trained using 110 specimens, 5 individual mosquitoes from each of 22 laboratory colonies including 5 A. arabiensis, 13 A. gambiae s.s., 2 A. merus and 2 A. quadriannulatus colonies (Table 2). When ranked by the CAT scores, including the top 68 peaks gave a model with zero remaining total error rate in the cross-validation (Figure 3 and Table S2). For estimating the generalised classification error of the final model the other 110 specimens, not used for model building and cross-validation, from the laboratory colonies plus an additional set of 125 field-caught female mosquitoes were classified using Model 1.


Identification of cryptic Anopheles mosquito species by molecular protein profiling.

Müller P, Pflüger V, Wittwer M, Ziegler D, Chandre F, Simard F, Lengeler C - PLoS ONE (2013)

Model selection and cross-validation to discriminate between species of the Anopheles gambiae complex (Model 1).(A) The graph shows the error rate from the cross-validation plotted as a function of the number of the ranked peaks included in the SDA model that discriminates between members of the A. gambiae species complex. The peaks were ranked (left to right) according to the correlation-adjusted t-scores (CAT scores). The vertical, red line shows the 68 peaks chosen for the SDA model. (B) List with the 68 ranked peaks (top equals highest rank) their corresponding CAT scores. The length and direction of the horizontal blue bars represents the CAT scores of the centroid versus the pooled mean and show the influence of a particular peak in differentiating between the groups (Table S2). For example the top peak, M12369.6 has a strong influence in separating A. merus from all the other species, emphasised by the length of the bar and the opposite direction from the bars of the other species. In contrast, the tenth peak, M12527.3 has a stronger influence in separating A. gambiae s.s. from A. arabiensis. AR: A. arabiensis; GA: A. gambiae s.s.; ME: A. merus; QD: A. quadriannulatus.
© Copyright Policy
Related In: Results  -  Collection

Show All Figures
getmorefigures.php?uid=PMC3585343&req=5

pone-0057486-g003: Model selection and cross-validation to discriminate between species of the Anopheles gambiae complex (Model 1).(A) The graph shows the error rate from the cross-validation plotted as a function of the number of the ranked peaks included in the SDA model that discriminates between members of the A. gambiae species complex. The peaks were ranked (left to right) according to the correlation-adjusted t-scores (CAT scores). The vertical, red line shows the 68 peaks chosen for the SDA model. (B) List with the 68 ranked peaks (top equals highest rank) their corresponding CAT scores. The length and direction of the horizontal blue bars represents the CAT scores of the centroid versus the pooled mean and show the influence of a particular peak in differentiating between the groups (Table S2). For example the top peak, M12369.6 has a strong influence in separating A. merus from all the other species, emphasised by the length of the bar and the opposite direction from the bars of the other species. In contrast, the tenth peak, M12527.3 has a stronger influence in separating A. gambiae s.s. from A. arabiensis. AR: A. arabiensis; GA: A. gambiae s.s.; ME: A. merus; QD: A. quadriannulatus.
Mentions: In an attempt to overcome the poor performance of the unsupervised cluster analysis in discriminating between the A. gambiae sibling species, a SDA classification model was evaluated as an alternative. The model (Model 1) was trained using 110 specimens, 5 individual mosquitoes from each of 22 laboratory colonies including 5 A. arabiensis, 13 A. gambiae s.s., 2 A. merus and 2 A. quadriannulatus colonies (Table 2). When ranked by the CAT scores, including the top 68 peaks gave a model with zero remaining total error rate in the cross-validation (Figure 3 and Table S2). For estimating the generalised classification error of the final model the other 110 specimens, not used for model building and cross-validation, from the laboratory colonies plus an additional set of 125 field-caught female mosquitoes were classified using Model 1.

Bottom Line: The approach also classifies specimens from different laboratory colonies; hence proving also very promising for its use in colony authentication as part of quality assurance in laboratory studies.While being exceptionally accurate and robust, MALDI-TOF MS has several advantages over other typing methods, including simple sample preparation and short processing time.As the method does not require DNA sequence information, data can also be reviewed at any later stage for diagnostic or functional patterns without the need for re-designing and re-processing biological material.

View Article: PubMed Central - PubMed

Affiliation: Department of Medical Services and Diagnostic, Swiss Tropical and Public Health Institute, Basel, Switzerland. pie.mueller@unibas.ch

ABSTRACT
Vector control is the mainstay of malaria control programmes. Successful vector control profoundly relies on accurate information on the target mosquito populations in order to choose the most appropriate intervention for a given mosquito species and to monitor its impact. An impediment to identify mosquito species is the existence of morphologically identical sibling species that play different roles in the transmission of pathogens and parasites. Currently PCR diagnostics are used to distinguish between sibling species. PCR based methods are, however, expensive, time-consuming and their development requires a priori DNA sequence information. Here, we evaluated an inexpensive molecular proteomics approach for Anopheles species: matrix assisted laser desorption/ionization time-of-flight mass spectrometry (MALDI-TOF MS). MALDI-TOF MS is a well developed protein profiling tool for the identification of microorganisms but so far has received little attention as a diagnostic tool in entomology. We measured MS spectra from specimens of 32 laboratory colonies and 2 field populations representing 12 Anopheles species including the A. gambiae species complex. An important step in the study was the advancement and implementation of a bioinformatics approach improving the resolution over previously applied cluster analysis. Borrowing tools for linear discriminant analysis from genomics, MALDI-TOF MS accurately identified taxonomically closely related mosquito species, including the separation between the M and S molecular forms of A. gambiae sensu stricto. The approach also classifies specimens from different laboratory colonies; hence proving also very promising for its use in colony authentication as part of quality assurance in laboratory studies. While being exceptionally accurate and robust, MALDI-TOF MS has several advantages over other typing methods, including simple sample preparation and short processing time. As the method does not require DNA sequence information, data can also be reviewed at any later stage for diagnostic or functional patterns without the need for re-designing and re-processing biological material.

Show MeSH
Related in: MedlinePlus