Shulou(Shulou.com)06/01 Report--

This article will explain in detail what are the details of case/control association analysis in plink. The content of the article is of high quality, so the editor will share it with you for reference. I hope you will have a certain understanding of the relevant knowledge after reading this article.

As the most commonly used tool for association analysis, plink supports a variety of algorithms for association analysis. For the classical case/control association analysis, the software supports two levels of association analysis: allele and genotype.

1. Allele association test

For each snp locus. Count the frequency of minor and major allele in the two groups of case/control and get the table shown below

AlleleAaCaseabControlcd

For the above 2 X 2 data, chi-square and Fisher exact tests are used for association analysis, which can be easily achieved through the-assoc parameter. The basic usage is as follows

The output is saved in two files with suffixes assoc and assoc.fisher, and the chi-square analysis results are as follows

The results of Fisher's accurate test analysis are as follows

A1 stands for minor allel and A2 stands for major allel. It is important to note that the minor allel here is determined according to the distribution in the actual sample. Fraca represents the frequency of minor allel in case, Fullu represents the frequency of minor allel in control, CHISQ represents chi-square, P stands for pvalue, and OR stands for odd ratio.

2. Genotype association test

For each snp locus. The distribution frequency of each genotype in the two groups of case/control was calculated. When dividing genotypes, the following three models are supported

GENO

DOM

REC

GENO divides genotypes into AA,Aa,aa3, DOM represents dominant genetic model, genotypes are divided into AA/Aa and aa, REC represents recessive genetic model, and genotypes are divided into AA,Aa / aa. The corresponding usage is as follows

The results are saved in a file with the suffix model, and the results of chi-square analysis are as follows

It can be seen that when the frequency is less than 5, the relevant results of Chi-square test are all NA. For each locus, in addition to the three genotypic models mentioned above, two models for allele, TREND and ALLELIC, are also given, in which the ALLELIC model is the same as that in allele association test, and the output results are exactly the same, and TREND calculates the p value by Cochran-Armitage trend test.

The output of Fisher's accurate test is as follows

AFF represents the frequency of allel/genotype in the case group, and UNAFF represents the distribution in the control group. The Fisher exact test in plink is the result of a bilateral test, and the results verified in R language are as follows

If you just want the correlation analysis of allele, use the assoc parameter, and if you need both allele and genotype correlation analysis, use the model parameter.

The above is only the basic usage. For the association analysis of a single site, we also need to correct the multiple hypothesis test, add-- adjust parameter, and sometimes need to add confidence interval, through-- ci parameter to achieve, the corresponding usage is as follows

At present, plink only supports the correction of output confidence interval and multiple hypothesis testing by-assoc, but not by model.

The details of case/control association analysis in plink are shared here. I hope the above content can be helpful to you and learn more knowledge. If you think the article is good, you can share it for more people to see.

Welcome to subscribe "Shulou Technology Information " to get latest news, interesting things and hot topics in the IT industry, and controls the hottest and latest Internet news, technology news and IT industry trends.