Volume 1 Supplement 1
A new transmission test for affected sib-pair families
© Xu and George; licensee BioMed Central Ltd. 2007
Published: 18 December 2007
Family-based association approaches such as the transmission-disequilibrium test (TDT) are used extensively in the study of genetic traits because they are generally robust to the presence of population structure. However, these approaches necessarily involve recruitment of families, which is more costly and time-consuming than sampling unrelated individuals in the population-based approaches. Therefore, a family-based approach, which has high power, would be appealing because of the gain in time and cost due to the reduced sample size that is required to attain adequate power. Here we introduce a new family-based transmission test using the joint transmission status from affected sib pairs. We show that by including the transmission status of both siblings, our method gives higher power than the TDT design, while maintaining the correct type I error rate. We use the simulated data from affected sib-pair families with rheumatoid arthritis provided by Genetic Analysis Workshop 15 to illustrate our approach.
Genetic association studies have contributed significantly in recent years to our understanding of the genetic basis of complex diseases. Association studies are roughly categorized into either population-based or family-based association approaches. Population-based approaches have the advantage that samples are easy to ascertain. However, it has been shown that population-based approaches, such as case-control studies, can produce spurious associations in the presence of population substructure, especially in large-scale studies at the genomic level [1, 2]. In the presence of population substructure, family-based approaches, such as the transmission-disequilibrium test (TDT) originally proposed by Spielman et al. , have the advantage that they are robust against population substructure. Over the past decade or so, the original TDT has been extended and expanded to cover many practical scenarios, as alternative approaches to population-based association studies. Some of these extensions include the sibling-TDT , the homozygote parent-TDT , the pedigree disequilibrium test (PDT) [6, 7], the quantitative TDT [8, 9], the Bayesian TDT , an entropy-based method , and the more general family-based association test (FBAT) . The motivation for these alternatives is that they are robust against population substructure and other cryptic relatedness in the samples [3, 13]. However, these methods necessarily involve the recruitment of families, which may be more costly and time-consuming than sampling of unrelated individuals in population-based approaches. Therefore, a family-based approach, which has high power (consequently requiring a smaller sample size to achieve the same power) will be preferable. In this study we introduce a new family-based transmission test that is more powerful than the standard TDT, incorporating pair-wise transmission status of siblings. We demonstrate our approach using the simulated data from affected sib-pair (ASP) families with rheumatoid arthritis (RA) provided by Genetic Analysis Workshop 15 (GAW15). The simulated ASP families contain the genotypes of the parents, homozygous by heterozygous for the marker locus, and two affected siblings, from which the transmission status of both siblings from the heterozygous parent can be inferred. We show that by considering the transmission status of both siblings, our method gives the correct type I error rate while yielding higher power than the standard TDT design.
Each of the 1500 simulated ASP families consists of four members: the father, the mother, and the ASP. The genotypes of all four members in each family are available. We used the simulated genotype data at both the genome-wide single-nucleotide polymorphism (SNP) markers and the chromosome 6 dense SNP markers from simulation Replicate 1.
We applied our method to both the simulated data set of 9187 genome-wide SNPs from Replicate 1 and the data set of 17,820 dense SNPs on chromosome 6. For comparison, we also applied the standard TDT to the same set of families with homozygote by heterozygote matings. In performing the TDT analysis, we treated the transmission/non-transmission of the alleles to the two siblings in an ASP as two independent observations. This assumption may not necessarily be valid, and could result in inflated power.
Type I error rate
The type I error rate is estimated empirically by performing the association tests on markers that are not associated with RA status and calculating the proportion of times the null hypothesis is rejected. We excluded the markers on the chromosomes that have trait loci associated with RA status and combined the test results of all available markers from all other chromosomes. Specifically, trait loci DR, C, and D on chromosome 6, locus F on chromosome 11, and locus E on chromosome 18 are associated with RA status, and, therefore, we used markers from all chromosomes excluding 6, 11 and 18, resulting in 7718 SNP markers for type I error analysis. Out of the 7718 tests performed using our approach and the standard TDT, our test rejected the null hypothesis 328 times, and the standard TDT rejected the null hypothesis 367 times. The corresponding estimated type I error rates of our method and TDT are 0.043 and 0.048, respectively, both of which are well under the nominal level of 0.05.
Number of signals detected at various significance levels in the analysis of genome wide SNPs on chromosome 6
Number of signals
Number of signals detected at various significance levels in the analysis of dense SNPs on chromosome 6
Number of signals
Because of the extra cost of recruiting family members, it is desirable to develop a family-based association method with relatively high power so that a smaller sample size is needed to achieve the same power. In this paper, we developed a new family-based transmission test using the joint transmission status of ASPs instead of the transmission status of individual offspring. The method gives the correct type I error rate and is more powerful than the TDT. In order to compare our method with the standard TDT, we treated the transmissions to the two siblings as two independent observations, which may not be a valid assumption for testing association . The TDT is generally applicable to only parent-offspring trio data. Using more than one offspring may result in inflation of power because the sibship correlation is not taken into account, and the effective sample size is inflated. If we used only one offspring per ASP family for the TDT, then the number of positive signals detected using the standard TDT would have been even smaller because of the drop in sample size and resulting reduction in power. By considering the joint transmission statuses of two siblings, we, in fact, gained more power compared with the standard TDT, even in the case in which the two transmissions are erroneously treated as independent. The increased power is critical for the study of complex diseases, because it could reduce the necessary sample size, which is especially important for late-onset diseases in which recruiting families could be difficult. A disadvantage of our approach compared with the standard TDT is that it requires sib-pair data instead of singletons. However, the proposed method is a complementary approach to the traditional TDT when sib-pair family data are available. Further, it should be easy to combine the two approaches when both singletons and sib pairs are available.
It should be noted that the proposed approach, as well as the standard TDT, can easily be extended to include homozygous offspring from double heterozygous parents . Also, both methods can be extended to include unaffected offspring and sib pairs, when available .
We proposed a new family based transmission test using the simulated ASP family data from GAW15. Our method gives the correct type I error rate. By considering the transmission status of the two siblings simultaneously, our method has higher power than the standard TDT.
This work was supported in part by the Medical College of Georgia Scientist Training Grant.
This article has been published as part of BMC Proceedings Volume 1 Supplement 1, 2007: Genetic Analysis Workshop 15: Gene Expression Analysis and Approaches to Detecting Multiple Functional Loci. The full contents of the supplement are available online at http://www.biomedcentral.com/1753-6561/1?issue=S1.
- Marchini J, Cardon LR, Phillips MS, Donnelly P: The effects of human population structure on large genetic association studies. Nat Genet. 2004, 36: 512-517. 10.1038/ng1337.View ArticlePubMedGoogle Scholar
- Xu H, Shete S: Effects of population structure on genetic association studies. BMC Genet. 2005, 6 (Suppl 1): S109-10.1186/1471-2156-6-S1-S109.View ArticlePubMed CentralPubMedGoogle Scholar
- Spielman RS, McGinnis RE, Ewens WJ: Transmission test for linkage disequilibrium: the insulin gene region and insulin-dependent diabetes mellitus (IDDM). Am J Hum Genet. 1993, 52: 506-516.PubMed CentralPubMedGoogle Scholar
- Spielman RS, Ewens WJ: A sibship test for linkage in the presence of association: the sib transmission/disequilibrium test. Am J Hum Genet. 1998, 62: 450-458. 10.1086/301714.View ArticlePubMed CentralPubMedGoogle Scholar
- Lie BA, Todd JA, Pociot F, Nerup J, Akselsen HE, Joner G, Dahl-Jorgensen K, Ronningen KS, Thorsby E, Undlien DE: The predisposition to type 1 diabetes linked to the human leukocyte antigen complex includes at least one non-class II gene. Am J Hum Genet. 1999, 64: 793-800. 10.1086/302283.View ArticlePubMed CentralPubMedGoogle Scholar
- Martin ER, Monks SA, Warren LL, Kaplan NL: A test for linkage and association in general pedigrees: the pedigree disequilibrium test. Am J Hum Genet. 2000, 67: 146-154. 10.1086/302957.View ArticlePubMed CentralPubMedGoogle Scholar
- Abecasis GR, Cookson WO, Cardon LR: Pedigree tests of transmission disequilibrium. Eur J Hum Genet. 2000, 8: 545-551. 10.1038/sj.ejhg.5200494.View ArticlePubMedGoogle Scholar
- Allison DB: Transmission-disequilibrium tests for quantitative traits. Am J Hum Genet. 1997, 60: 676-690.PubMed CentralPubMedGoogle Scholar
- George V, Tiwari HK, Zhu X, Elston RC: A test of transmission/disequilibrium for quantitative traits in pedigree data, by multiple regression. Am J Hum Genet. 1999, 65: 236-245. 10.1086/302444.View ArticlePubMed CentralPubMedGoogle Scholar
- George V, Laud PW: A Bayesian approach to the transmission/disequilibrium test for binary traits. Genet Epidemiol. 2002, 22: 41-51. 10.1002/gepi.1042.View ArticlePubMedGoogle Scholar
- Zhao J, Boerwinkle E, Xiong M: An entropy-based genome-wide transmission/disequilibrium test. Hum Genet. 2007, 121: 357-367. 10.1007/s00439-007-0322-6.View ArticlePubMedGoogle Scholar
- Rabinowitz D, Laird N: A unified approach to adjusting association tests for population admixture with arbitrary pedigree structure and arbitrary missing marker information. Hum Hered. 2000, 50: 211-223. 10.1159/000022918.View ArticlePubMedGoogle Scholar
- Ewens WJ, Spielman RS: The transmission/disequilibrium test: history, subdivision, and admixture. Am J Hum Genet. 1995, 57: 455-464.View ArticlePubMed CentralPubMedGoogle Scholar
- Martin ER, Kaplan NL, Weir BS: Tests for linkage and association in nuclear families. Am J Hum Genet. 1997, 61: 439-448. 10.1017/S0003480097006362.View ArticlePubMed CentralPubMedGoogle Scholar
This article is published under license to BioMed Central Ltd. This is an open access article distributed under the terms of the Creative Commons Attribution License (http://creativecommons.org/licenses/by/2.0), which permits unrestricted use, distribution, and reproduction in any medium, provided the original work is properly cited.