Twin Research and Human Genetics

Articles

Large-Scale Zygosity Testing Using Single Nucleotide Polymorphisms

Ulf Hanneliusa1 c1, Loreana Ghermana2, Ville-Veikko Mäkeläa3, Astrid Lindstedta4, Marco Zucchellia5, Camilla Lagerberga6, Gunnel Tybringa7, Juha Kerea8 and Cecilia M Lindgrena9

a1 Department of Biosciences and Nutrition, Karolinska Institutet, Huddinge, Sweden. ulf.hannelius@biosci.ki.se

a2 Department of Medical Epidemiology and Biostatistics, Karolinska Institutet, Stockholm, Sweden.

a3 Clinical Research Centre, Karolinska University Hospital, Stockholm, Sweden.

a4 Clinical Research Centre, Karolinska University Hospital, Stockholm, Sweden.

a5 Department of Biosciences and Nutrition, Karolinska Institutet, Huddinge, Sweden.

a6 Department of Medical Epidemiology and Biostatistics, Karolinska Institutet, Stockholm, Sweden.

a7 Department of Medical Epidemiology and Biostatistics, Karolinska Institutet, Stockholm, Sweden.

a8 Department of Biosciences and Nutrition, Karolinska Institutet, Huddinge, Sweden; Clinical Research Centre, Karolinska University Hospital, Stockholm, Sweden; Department of Medical Genetics, University of Helsinki, Helsinki, Finland.

a9 Wellcome Trust Centre for Human Genetics, Oxford University, Oxford, United Kingdom.

Abstract

A requirement for performing robust genetic and statistical analyses on twins is correctly assigned zygosities. In order to increase the power to detect small risk factors of disease, zygosity testing should also be amenable for high throughput screening. In this study we validate and implement the use of a panel of 50 single nucleotide polymorphisms (SNPs) for reliable high throughput zygosity testing and compare it to a panel of 16 short tandem repeats (STRs). We genotyped both genomic (gDNA) and whole genome amplified DNA (WGA DNA), ending up with 47 SNP and 11 STR markers fulfilling our quality criteria. Out of 99 studied twin pairs, 2 were assigned a different zygosity using SNP and STR data as compared to self reported zygosity in a questionnaire. We also performed a sensitivity analysis based on simulated data where we evaluated the effects of genotyping error, shifts in allele frequencies and missing data on the qualitative zygosity assignments. The frequency of false positives was less than 0.01 when assuming a 1% genotyping error, a decrease of 10% of the observed minor allele frequency compared to the actual values and up to 10 missing markers. The SNP markers were also successfully genotyped on both gDNA and WGA DNA from whole blood, saliva and filter paper. In conclusion, we validate a robust panel of 47 highly multiplexed SNPs that provide reliable and high quality data on a range of different DNA templates.

(Received April 12 2007)

(Accepted May 23 2007)

Correspondence:

c1 Address for correspondence: Ulf Hannelius, Department of Biosciences and Nutrition, Karolinska Institutet, Hälsovägen 7-9, 14157 Huddinge, Sweden

Metrics