Limits...
A hidden two-locus disease association pattern in genome-wide association studies.

Yang C, Wan X, Yang Q, Xue H, Tang NL, Yu W - BMC Bioinformatics (2011)

Bottom Line: The correlation among SNPs can lead to weak marginal effects and the interaction does not play a role in this association pattern.This phenomenon is due to the existence of unfaithfulness: the marginal effects of correlated SNPs do not express their significant joint effects faithfully due to the correlation cancelation.Based on the empirical result of these real data, we show that this type of association masked by unfaithfulness widely exists in GWAS.

View Article: PubMed Central - HTML - PubMed

Affiliation: Department of Electronic and Computer Engineering, Hong Kong University of Science and Technology, Hong Kong. eeyang@ust.hk

ABSTRACT

Background: Recent association analyses in genome-wide association studies (GWAS) mainly focus on single-locus association tests (marginal tests) and two-locus interaction detections. These analysis methods have provided strong evidence of associations between genetics variances and complex diseases. However, there exists a type of association pattern, which often occurs within local regions in the genome and is unlikely to be detected by either marginal tests or interaction tests. This association pattern involves a group of correlated single-nucleotide polymorphisms (SNPs). The correlation among SNPs can lead to weak marginal effects and the interaction does not play a role in this association pattern. This phenomenon is due to the existence of unfaithfulness: the marginal effects of correlated SNPs do not express their significant joint effects faithfully due to the correlation cancelation.

Results: In this paper, we develop a computational method to detect this association pattern masked by unfaithfulness. We have applied our method to analyze seven data sets from the Wellcome Trust Case Control Consortium (WTCCC). The analysis for each data set takes about one week to finish the examination of all pairs of SNPs. Based on the empirical result of these real data, we show that this type of association masked by unfaithfulness widely exists in GWAS.

Conclusions: These newly identified associations enrich the discoveries of GWAS, which may provide new insights both in the analysis of tagSNPs and in the experiment design of GWAS. Since these associations may be easily missed by existing analysis tools, we can only connect some of them to publicly available findings from other association studies. As independent data set is limited at this moment, we also have difficulties to replicate these findings. More biological implications need further investigation.

Availability: The software is freely available at http://bioinformatics.ust.hk/hidden_pattern_finder.zip.

Show MeSH

Related in: MedlinePlus

Illustration of unfaithfulness in association studies. There are three regression models in each scenario: Y ~ β1X1 + β2X2,  and . In this figure, the marginal coefficient  and  are shown as projections (marked with bold red color) of Y on X1 and X2, respectively. (a) X1 is not correlated with X2. (b) X1 is positively correlated with X2. (c) X1 is negatively correlated with X2. (d) X1 is positively correlated with X2 but the sign of β1 is the opposite of the sign of β2. Scenario (c) and Scenario (d) illustrate unfaithfulness.
© Copyright Policy - open-access
Related In: Results  -  Collection

License
getmorefigures.php?uid=PMC3116488&req=5

Figure 1: Illustration of unfaithfulness in association studies. There are three regression models in each scenario: Y ~ β1X1 + β2X2, and . In this figure, the marginal coefficient and are shown as projections (marked with bold red color) of Y on X1 and X2, respectively. (a) X1 is not correlated with X2. (b) X1 is positively correlated with X2. (c) X1 is negatively correlated with X2. (d) X1 is positively correlated with X2 but the sign of β1 is the opposite of the sign of β2. Scenario (c) and Scenario (d) illustrate unfaithfulness.

Mentions: However, if X1 correlates with X2, fitting the model Y ~ β1X1 + β2X2 may identify a new association pattern with β1 and β2 (named as bivariate regression coefficients) being significantly larger than and . This phenomenon is referred to as unfaithfulness. It means that the marginal effects of correlated variables do not express their significant joint effects faithfully due to the correlation cancelation [11]. Figure 1 provides some synthetic examples to show the unfaithfulness involving two variables. There are four scenarios to illustrate the relationship between marginal coefficients (marked using red color) and bivariate regression coefficients. The first scenario (Figure 1:(a)) is a reference case that involves no correlations between X1 and X2. The marginal coefficients and are equal to the bivariate regression coefficients β1 and β2, respectively. In the second scenario (Figure 1:(b)), X1 is positively correlated with X2. The marginal coefficients are bigger than the bivariate regression coefficients. In the third scenario (Figure 1:(c)), X1 is negatively correlated with X2. The marginal coefficient and could be significantly smaller than the bivariate regression coefficients β1 and. β2 In the the fourth scenario (Figure 1:(d)), X1 is positively correlated with X2. But the sign of β1 is the opposite of the sign of β2. The correlation effect in the third scenario and the fourth scenario causes the unfaithfulness. In mathematics, the relationship between the marginal coefficients and the bivariate regression coefficients is formulated as(1)


A hidden two-locus disease association pattern in genome-wide association studies.

Yang C, Wan X, Yang Q, Xue H, Tang NL, Yu W - BMC Bioinformatics (2011)

Illustration of unfaithfulness in association studies. There are three regression models in each scenario: Y ~ β1X1 + β2X2,  and . In this figure, the marginal coefficient  and  are shown as projections (marked with bold red color) of Y on X1 and X2, respectively. (a) X1 is not correlated with X2. (b) X1 is positively correlated with X2. (c) X1 is negatively correlated with X2. (d) X1 is positively correlated with X2 but the sign of β1 is the opposite of the sign of β2. Scenario (c) and Scenario (d) illustrate unfaithfulness.
© Copyright Policy - open-access
Related In: Results  -  Collection

License
Show All Figures
getmorefigures.php?uid=PMC3116488&req=5

Figure 1: Illustration of unfaithfulness in association studies. There are three regression models in each scenario: Y ~ β1X1 + β2X2, and . In this figure, the marginal coefficient and are shown as projections (marked with bold red color) of Y on X1 and X2, respectively. (a) X1 is not correlated with X2. (b) X1 is positively correlated with X2. (c) X1 is negatively correlated with X2. (d) X1 is positively correlated with X2 but the sign of β1 is the opposite of the sign of β2. Scenario (c) and Scenario (d) illustrate unfaithfulness.
Mentions: However, if X1 correlates with X2, fitting the model Y ~ β1X1 + β2X2 may identify a new association pattern with β1 and β2 (named as bivariate regression coefficients) being significantly larger than and . This phenomenon is referred to as unfaithfulness. It means that the marginal effects of correlated variables do not express their significant joint effects faithfully due to the correlation cancelation [11]. Figure 1 provides some synthetic examples to show the unfaithfulness involving two variables. There are four scenarios to illustrate the relationship between marginal coefficients (marked using red color) and bivariate regression coefficients. The first scenario (Figure 1:(a)) is a reference case that involves no correlations between X1 and X2. The marginal coefficients and are equal to the bivariate regression coefficients β1 and β2, respectively. In the second scenario (Figure 1:(b)), X1 is positively correlated with X2. The marginal coefficients are bigger than the bivariate regression coefficients. In the third scenario (Figure 1:(c)), X1 is negatively correlated with X2. The marginal coefficient and could be significantly smaller than the bivariate regression coefficients β1 and. β2 In the the fourth scenario (Figure 1:(d)), X1 is positively correlated with X2. But the sign of β1 is the opposite of the sign of β2. The correlation effect in the third scenario and the fourth scenario causes the unfaithfulness. In mathematics, the relationship between the marginal coefficients and the bivariate regression coefficients is formulated as(1)

Bottom Line: The correlation among SNPs can lead to weak marginal effects and the interaction does not play a role in this association pattern.This phenomenon is due to the existence of unfaithfulness: the marginal effects of correlated SNPs do not express their significant joint effects faithfully due to the correlation cancelation.Based on the empirical result of these real data, we show that this type of association masked by unfaithfulness widely exists in GWAS.

View Article: PubMed Central - HTML - PubMed

Affiliation: Department of Electronic and Computer Engineering, Hong Kong University of Science and Technology, Hong Kong. eeyang@ust.hk

ABSTRACT

Background: Recent association analyses in genome-wide association studies (GWAS) mainly focus on single-locus association tests (marginal tests) and two-locus interaction detections. These analysis methods have provided strong evidence of associations between genetics variances and complex diseases. However, there exists a type of association pattern, which often occurs within local regions in the genome and is unlikely to be detected by either marginal tests or interaction tests. This association pattern involves a group of correlated single-nucleotide polymorphisms (SNPs). The correlation among SNPs can lead to weak marginal effects and the interaction does not play a role in this association pattern. This phenomenon is due to the existence of unfaithfulness: the marginal effects of correlated SNPs do not express their significant joint effects faithfully due to the correlation cancelation.

Results: In this paper, we develop a computational method to detect this association pattern masked by unfaithfulness. We have applied our method to analyze seven data sets from the Wellcome Trust Case Control Consortium (WTCCC). The analysis for each data set takes about one week to finish the examination of all pairs of SNPs. Based on the empirical result of these real data, we show that this type of association masked by unfaithfulness widely exists in GWAS.

Conclusions: These newly identified associations enrich the discoveries of GWAS, which may provide new insights both in the analysis of tagSNPs and in the experiment design of GWAS. Since these associations may be easily missed by existing analysis tools, we can only connect some of them to publicly available findings from other association studies. As independent data set is limited at this moment, we also have difficulties to replicate these findings. More biological implications need further investigation.

Availability: The software is freely available at http://bioinformatics.ust.hk/hidden_pattern_finder.zip.

Show MeSH
Related in: MedlinePlus