Introduction: Recently an Illumina Infinium 8K apple SNP-chip was developed by the International RosBREED SNP Consortium for high-throughput genotyping and association studies (Chagne et al, 2012). We have used this SNP-chip to genotype 144 individuals of a Malus x domestica germplasm collection consisting of old Belgian and commercial cultivars, as well as 128 individual progeny of a mapping population. Aims: The genotype data generated was used i) to evaluate the overall performance of the chip and the robustness of the SNP calling both within a group of related individuals, as well as across a wider germplasm and ii) to develop a set of filter parameters to allow accurate automatic calling of the correct SNP genotype. Results: Overall only 9% of the SNPs failed completely. However, when comparing the genotypes of cultivars common to two different datasets, it was clear that differences in population structure between the datasets strongly influences SNP-calling, due to difficulties in clustering fluorescent intensities into the different genotype classes. This in turn is due to the presence of homologous sequences that also hybridize to the chip probes used for the SNP genotyping. This leads to shifts in the observed fluorescence used for genotype calling (AA, AB, BB) and shifts in the allele frequencies depending on the number of homologous loci and the degree of sequence identity. In order to remove those SNPs that show differences between datasets, a bootstrap analysis was done and cut-off values for a range of quality parameters commonly used for filtering SNP-chip data, were established using the variability of the callings over the bootstrap populations. Secondly a visual inspection of the intensity clustering used to genotype each SNP was carried out. This showed that even after automatic filtering, some SNPs showed deviations from the expected distributions due to the presence of the paralogous sequences. The fluorescence plot of these SNPs showed shifted clusters, the presence of additional genotype groups in the plot or a combination of both making it often hard to determine the true genotypes for the different individuals. In order to remove SNPs for which the genotype calling was influenced by the presence of paralogous sequences in the genome, two additional parameters were introduced: one for the cluster position and one for the cluster width since the additional groups were combined into the existing clusters leading to wider clusters. Depending on the stringency of the applied cut-off values for the filter parameters, the number of SNPs that are retained varies between 2,000and 3,500 SNPs. With manual inspection, the proportion of SNPs that were considered to be accurately called rose from 45% without filtering to 85-90.6% depending on the stringency of the filter parameters

Vanderzande, S.; Van de Weg, E.; Troggio, M.; Keulemans, W.; Davey, M. (2013). Accurate SNP-calling with the Illumina Infinium 8K apple SNP-chip. In: ISHS Molecular Markers in Horticulture Symposium, Riva del Garda, TN-Italy, September 25-27, 2013: 62. handle: http://hdl.handle.net/10449/22999

Accurate SNP-calling with the Illumina Infinium 8K apple SNP-chip

Troggio, Michela;
2013-01-01

Abstract

Introduction: Recently an Illumina Infinium 8K apple SNP-chip was developed by the International RosBREED SNP Consortium for high-throughput genotyping and association studies (Chagne et al, 2012). We have used this SNP-chip to genotype 144 individuals of a Malus x domestica germplasm collection consisting of old Belgian and commercial cultivars, as well as 128 individual progeny of a mapping population. Aims: The genotype data generated was used i) to evaluate the overall performance of the chip and the robustness of the SNP calling both within a group of related individuals, as well as across a wider germplasm and ii) to develop a set of filter parameters to allow accurate automatic calling of the correct SNP genotype. Results: Overall only 9% of the SNPs failed completely. However, when comparing the genotypes of cultivars common to two different datasets, it was clear that differences in population structure between the datasets strongly influences SNP-calling, due to difficulties in clustering fluorescent intensities into the different genotype classes. This in turn is due to the presence of homologous sequences that also hybridize to the chip probes used for the SNP genotyping. This leads to shifts in the observed fluorescence used for genotype calling (AA, AB, BB) and shifts in the allele frequencies depending on the number of homologous loci and the degree of sequence identity. In order to remove those SNPs that show differences between datasets, a bootstrap analysis was done and cut-off values for a range of quality parameters commonly used for filtering SNP-chip data, were established using the variability of the callings over the bootstrap populations. Secondly a visual inspection of the intensity clustering used to genotype each SNP was carried out. This showed that even after automatic filtering, some SNPs showed deviations from the expected distributions due to the presence of the paralogous sequences. The fluorescence plot of these SNPs showed shifted clusters, the presence of additional genotype groups in the plot or a combination of both making it often hard to determine the true genotypes for the different individuals. In order to remove SNPs for which the genotype calling was influenced by the presence of paralogous sequences in the genome, two additional parameters were introduced: one for the cluster position and one for the cluster width since the additional groups were combined into the existing clusters leading to wider clusters. Depending on the stringency of the applied cut-off values for the filter parameters, the number of SNPs that are retained varies between 2,000and 3,500 SNPs. With manual inspection, the proportion of SNPs that were considered to be accurately called rose from 45% without filtering to 85-90.6% depending on the stringency of the filter parameters
2013
Vanderzande, S.; Van de Weg, E.; Troggio, M.; Keulemans, W.; Davey, M. (2013). Accurate SNP-calling with the Illumina Infinium 8K apple SNP-chip. In: ISHS Molecular Markers in Horticulture Symposium, Riva del Garda, TN-Italy, September 25-27, 2013: 62. handle: http://hdl.handle.net/10449/22999
File in questo prodotto:
File Dimensione Formato  
2013 ISHS Vanderzande et al.pdf

accesso aperto

Licenza: Tutti i diritti riservati (All rights reserved)
Dimensione 1.26 MB
Formato Adobe PDF
1.26 MB Adobe PDF Visualizza/Apri

I documenti in IRIS sono protetti da copyright e tutti i diritti sono riservati, salvo diversa indicazione.

Utilizza questo identificativo per citare o creare un link a questo documento: https://hdl.handle.net/10449/22999
Citazioni
  • ???jsp.display-item.citation.pmc??? ND
  • Scopus ND
  • ???jsp.display-item.citation.isi??? ND
social impact