Background: Imputation of partially missing or unobserved genotypes is an indispensable tool for SNP data analyses. However, research and understanding of the impact of initial SNP-data quality control on imputation results is still limited. In this paper, we aim to evaluate the effect of different strategies of pre-imputation quality filtering on the performance of the widely used imputation algorithms MaCH and IMPUTE. Results: We considered three scenarios: imputation of partially missing genotypes with usage of an external reference panel, without usage of an external reference panel, as well as imputation of ompletely un-typed SNPs using an external reference panel. We first created various datasets applying different SNP quality filters and masking certain percentages of randomly selected high-quality SNPs. We imputed these SNPs and compared the results between the different filtering scenarios by using established and newly proposed measures of imputation quality. While the established measures assess certainty of imputation results, our newly proposed measures focus on the agreement with true genotypes. These measures showed that pre-imputation SNP-filtering might be detrimental regarding imputation quality. Moreover, the strongest drivers of imputation quality were in general the burden of missingness and the number of SNPs used for imputation. We also found that using a reference panel always improves imputation quality of partially missing genotypes. MaCH performed slightly better than IMPUTE2 in most of our scenarios. Again, these results were more pronounced when using our newly defined measures of imputation quality. Conclusion: Even a moderate filtering has a detrimental effect on the imputation quality. Therefore little or no SNP filtering prior to imputation appears to be the best strategy for imputing small to moderately sized datasets. Our results also showed that for these datasets, MaCH performs slightly better than IMPUTE2 in most scenarios at the cost of increased computing time.

Background: Imputation of partially missing or unobserved genotypes is an indispensable tool for SNP data analyses. However, research and understanding of the impact of initial SNP-data quality control on imputation results is still limited. In this paper, we aim to evaluate the effect of different strategies of pre-imputation quality filtering on the performance of the widely used imputation algorithms MaCH and IMPUTE. Results: We considered three scenarios: imputation of partially missing genotypes with usage of an external reference panel, without usage of an external reference panel, as well as imputation of ompletely un-typed SNPs using an external reference panel. We first created various datasets applying different SNP quality filters and masking certain percentages of randomly selected high-quality SNPs. We imputed these SNPs and compared the results between the different filtering scenarios by using established and newly proposed measures of imputation quality. While the established measures assess certainty of imputation results, our newly proposed measures focus on the agreement with true genotypes. These measures showed that pre-imputation SNP-filtering might be detrimental regarding imputation quality. Moreover, the strongest drivers of imputation quality were in general the burden of missingness and the number of SNPs used for imputation. We also found that using a reference panel always improves imputation quality of partially missing genotypes. MaCH performed slightly better than IMPUTE2 in most of our scenarios. Again, these results were more pronounced when using our newly defined measures of imputation quality. Conclusion: Even a moderate filtering has a detrimental effect on the imputation quality. Therefore little or no SNP filtering prior to imputation appears to be the best strategy for imputing small to moderately sized datasets. Our results also showed that for these datasets, MaCH performs slightly better than IMPUTE2 in most scenarios at the cost of increased computing time.

In den letzten Jahren gab es rasche Weiterentwicklungen auf dem Gebiet der Nukleinsäure-Erkennung. Von microRNA-Quantifizierung zur Untersuchung von Zelltods, --Teilung und -Regulation bis zur Bewertung genetischer Variabilität in Hinblick auf Krankheitsentstehung und -Behandlung: Die Analyse von Nukleinsäuren wird in der zukünftigen Medizin eine zentrale Rolle zukommen. Vor allem die Erkennung von SNPs als Hauptquelle der genetischen Vielfalt, aber aus Analysesicht auch eine der herausforderndsten Mutationen, stellt in dieser Hinsicht einen wesentlichen Aspekt dar. Methoden zur SNP-Erkennung müssen nicht nur sensibel, selektiv und stabil, sondern auch vielfältig sein und eine der wachsenden Analyseanzahl gerecht werdende hohe Verarbeitungsmenge bieten. Im Rahmendieser Arbeit wurde ein chemisches Prüfverfahren zur Erkennung von Nukleinsäuren und Einzelnukleotid-Polymorphismen (SNPs) entwickelt. Das Reaktionssystem zur Nukleinsäuren- Erkennung beruht hierbeiauf der Interaktion zweier modifizierter Peptid-Nukleinsäure (PNS) Oligonukleotiden. Das Erste beinhaltet einen C-terminalen Thioester (Donor-Sonde), die zweite einen N-terminalen Cysteinyl-Rest (Akzeptor-Sonde). Zusätzlich ist die Donor-Sonde durch einenmakrocyclischen Metall Chelatkomplex aus 1,4,7,10-tetraazacyclododecan-1,4,7,10-tetraessigsäure(DOTA) mit einem gebundenen lanthanoid-tag funktionalisiert. In die Akzeptor-Sonde wurde, zurReinigung mit magnetischen Streptavidin Partikeln, Biotin integriert. Der Ziel-DNA-Strang bringt beideSonden in räumliche Nähe zueinander und ermöglicht so eine chemische Reaktion. Das so gewonneneLigationsprodukt beinhaltet den Lanthanoid-Tag und Biotin, über welches das Produkt gereinigt wird,bevor die Detektion mittels ICP-MS erfolgt. Die Lanthanoid Konzentration dient als Indikator desLigationsprodukts welches wiederum den Reporter des Ziel-DNS-Strangs darstellt. Die, mithilfe diesesSystems erreichte, methodische Nachweisgrenze lag bei 29 pM mit einem RSD von 6,8% bei 50 pM(n=5). Zur Erkennung von SNPs wurde das Experiment mit einer Kombination zweier-Sets PNS Sonden mit unterschiedlichen Lanthanoid Tags durchgeführt. Das erste Set zielte auf die SNP beinhaltende Sequenz (Reportersystem) ab, während das zweite an eine benachbarte Sequenz (Kontrollsystem) binden sollte. Zur Erkennung der SNP wurden die Signale bei der Lanthanoide wurden ins Verhältnis gesetzt. Mithilfe dieses Verfahrens konnte durch Messung von sechs Lanthaniden bei einer Konzentration von 5 nM erfolgreich simultan zwischen den Allelen dreier SNPs unterschieden werden. / The field of nucleic acid detection has evolved swiftly in recent years. From quantification of micro RNA for the study of cell death, proliferation, and regulation, to the assessment of the influence of genetic variability towards disease development and treatment, the analysis of nucleic acids will play a central role in future medicine. In that regard, the detection of SNPs, as the primary source of genetic variability and the most challenging mutation from the analytical point of view, will be at the forefront of the discussion. Methods for the detection of SNPs not only require sensitivity, selectivity and robustness, but they should also allow multiplexing and offer high throughput in order to face the growing analysis demand In this work an assay for the detection of nucleic acids and single nucleotide polymorphisms (SNPs) was developed. The reaction system for the detection of nucleic acids is based on the interaction between two modified peptide nucleic acid (PNA) oligonucleotides. The first incorporated a C-terminal thioester (donor probe), and the second one a N-terminal cysteinyl residue (acceptor probe). In addition, the donor probe is functionalized with a metal-tag, which consist of a macrocyclic metal chelate complex of 1,4,7,10-tetraazacyclododecane-1,4,7,10-tetraacetic acid (DOTA) with a chelated lanthanoide. A biotin tag for purification by streptavidin magnetic particles was incorporated in the acceptor probe. The target DNA strand brings together the reporter probes allowing the chemical reaction. The resulting ligation product contains the metal-tag and the biotin, which is used to purify the product before measurement in the ICP-MS system. The lanthanoid concentration is used as an indicator of the ligation product, which at the same time serves as reporter of the target template. The methodological limit of detection achieved with this system was 29 pM with RSD of 6.8% at 50 pM (n=5). Detection of SNPs was performed using a combination of two sets of PNA probes labeled with different lanthanoid metal tags. The first probe set targeted the sequence where the SNP was present (reporter probe system), while the second set of probes was designed to bind to a neighboring sequence (control probe system). The signals of both lanthanides were used to establish a ratio that allowed the detection of the SNP. This assay was successfully used to simultaneously differentiate between alleles of 3 SNPs by measuring six lanthanoids at 5 nM concentration.