| Title: | Genetic Association Models for X-Chromosome SNPS on Continuous, Binary and Survival Outcomes |
|---|---|
| Description: | The expression of X-chromosome undergoes three possible biological processes: X-chromosome inactivation (XCI), escape of the X-chromosome inactivation (XCI-E),and skewed X-chromosome inactivation (XCI-S). To analyze the X-linked genetic association for phenotype such as continuous, binary, and time-to-event outcomes with the actual process unknown, we propose a unified approach of maximizing the likelihood or partial likelihood over all of the potential biological processes. The methods are described in Wei Xu, Meiling Hao (2017) <doi:10.1002/gepi.22097>. And also see Dongxiao Han, Meiling Hao, Lianqiang Qu, Wei Xu (2019) <doi:10.1177/0962280219859037>. |
| Authors: | Wei Xu [aut], Meiling Hao [aut], Yi Zhu [cre] |
| Maintainer: | Yi Zhu <[email protected]> |
| License: | GPL-2 |
| Version: | 1.0.1 |
| Built: | 2025-03-08 02:44:49 UTC |
| Source: | https://github.com/qiuanzhu/xlink |
fit_all_models returns model fitting results for each SNP understanding as XCI, XCI-E and XCI-S type respectively. Model comparison results is provided by using AIC as a criterion.
fit_all_models(resp, os, ostime, snp, gender, male, female, covars, model, data)fit_all_models(resp, os, ostime, snp, gender, male, female, covars, model, data)
resp |
Response variable for continuous or binary model fitting. |
os |
Survival indicator, 1 for death, 0 for censoring. |
ostime |
Duration time of survival. |
snp |
Single SNP name. |
gender |
Gender variable. |
male |
Male indicator in gender variable. |
female |
Female indicator in gender variable. |
covars |
Covariates list. |
model |
Fitting model type. For 'linear', fitting linear model. For 'binary', fitting logistic regression model. For 'survival', fitting survival model. |
data |
Data set. |
It returns estimated parameters, confidence interval and P value for each variable. Baseline model and full model maximum likelihood estimation are provided.
lm{stats} for linear model, glm{stats} for logistic regression model, and coxph{survival} for survival model.
fit_XCI_E_model returns model fitting results for each SNP understanding as XCI-E type.
fit_XCI_E_model(resp, os, ostime, snp, gender, male, female, covars, model, data)fit_XCI_E_model(resp, os, ostime, snp, gender, male, female, covars, model, data)
resp |
Response variable for continuous or binary model fitting. |
os |
Survival indicator, 1 for death, 0 for censoring. |
ostime |
Duration time of survival. |
snp |
Single SNP name. |
gender |
Gender variable. |
male |
Male indicator in gender variable. |
female |
Female indicator in gender variable. |
covars |
Covariates list. |
model |
Fitting model type. For 'linear', fitting linear model. For 'binary', fitting logistic regression model. For 'survival', fitting survival model. |
data |
Data set. |
It returns estimated parameters, confidence interval and P value for each variable. Baseline model and full model maximum likelihood estimation are provided.
lm{stats} for linear model, glm{stats} for logistic regression model, and coxph{survival} for survival model.
fit_XCI_model returns model fitting results for each SNP understanding as XCI type.
fit_XCI_model(resp, os, ostime, snp, gender, male, female, covars, model, data)fit_XCI_model(resp, os, ostime, snp, gender, male, female, covars, model, data)
resp |
Response variable for continuous or binary model fitting. |
os |
Survival indicator, 1 for death, 0 for censoring. |
ostime |
Duration time of survival. |
snp |
Single SNP name. |
gender |
Gender variable. |
male |
Male indicator in gender variable. |
female |
Female indicator in gender variable. |
covars |
Covariates list. |
model |
Fitting model type. For 'linear', fitting linear model. For 'binary', fitting logistic regression model. For 'survival', fitting survival model. |
data |
Data set. |
It returns estimated parameters, confidence interval and P value for each variable. Baseline model and full model maximum likelihood estimation are provided.
lm{stats} for linear model, glm{stats} for logistic regression model, and coxph{survival} for survival model.
infor_table returns information table of estimated coefficients/hazard ratio, confidence interval and P value.
infor_table(x, snp, covar_n, MAF_value, model)infor_table(x, snp, covar_n, MAF_value, model)
x |
An output from continuous/binary/survival model. |
snp |
Single SNP name. |
covar_n |
Covariate names. |
MAF_value |
A minimum value of minor allele frequency. |
model |
Model type. |
Information table. If linear or binary model is chosen, it returns estimated coefficients, confidence interval and P value. If survival model is chosen, it returns hazard ratio, confidence interval and P value.
MAF returns SNPs with higher MAF than default value.
MAF(snp, gender, male, MAF_v, data)MAF(snp, gender, male, MAF_v, data)
snp |
SNP name. |
gender |
gender variable. |
male |
male information. |
MAF_v |
minimum MAF value. |
data |
a dataset. |
A simulated dataset containing 400 observations. The variables list as follows:
data(Rdata)data(Rdata)
A data frame with 400 rows and 17 variables.
ID. Identification number.
OS. Survival indicator, 1 for death, 0 for censoring.
OS_time. Duration time of survival.
gender. Binary value 0,1 with P(x=1)=0.5 and hazard ratio is 1.
Age. Uniform distribution in [20,80] and hazard ratio is 1.02.
Smoking. Binary value 0,1 with P(x=1)=0.3 and hazard ratio is 1.2.
Treatment. Binary value 0,1 with P(X=1)=0.3 and hazard ratio is 1.2.
snp_1. True type in coxph model is 'XCI', minor allele frequency is 0.2, hazard ratio is 1.5.
snp_2. True type in coxph model is 'XCI-E', minor allele frequency is 0.3, hazard ratio is 1.5.
snp_3. True type in coxph model is 'XCI-S', minor allele frequency is 0.4, hazard ratio is 1.5.
snp_4. True type in coxph model is 'XCI', minor allele frequency is 0.3, hazard ratio is 1.
snp_5. True type in coxph model is 'XCI-E', minor allele frequency is 0.1, hazard ratio is 1.
snp_6. True type in coxph model is 'XCI', minor allele frequency is 0.2, hazard ratio is 1.
snp_7. True type in coxph model is 'XCI', minor allele frequency is 0.15, hazard ratio is 1.
snp_8. True type in coxph model is 'XCI-E', minor allele frequency is 0.1, hazard ratio is 1.
snp_9. True type in coxph model is 'XCI', minor allele frequency is 0.2, hazard ratio is 1.
snp_10. True type in coxph model is 'XCI', minor allele frequency is 0.15, hazard ratio is 1.
select_output returns selected SNP information by P value.
select_output(input, pv_thold = 1)select_output(input, pv_thold = 1)
input |
Input results from xlink fit. |
pv_thold |
P value threshold for output. |
It returns estimated parameters, confidence interval, P value, MAF and Best model information.
xlink_fit{xlink} for input results.
Covars<-c("Age","Smoking","Treatment") SNPs<-c("snp_1","snp_2","snp_3") result<-xlink_fit(os="OS",ostime ="OS_time",snps=SNPs,gender ="gender",covars=Covars, option =list(type="all",MAF_v=0.05), model="survival", data = Rdata) select_output(input=result,pv_thold=10^-5)Covars<-c("Age","Smoking","Treatment") SNPs<-c("snp_1","snp_2","snp_3") result<-xlink_fit(os="OS",ostime ="OS_time",snps=SNPs,gender ="gender",covars=Covars, option =list(type="all",MAF_v=0.05), model="survival", data = Rdata) select_output(input=result,pv_thold=10^-5)
The expression of X-chromosome undergoes three possible biological processes: X-chromosome inactivation (XCI), escape of the X-chromosome inactivation (XCI-E),and skewed X-chromosome inactivation (XCI-S). To analyze the X-linked genetic association for phenotype such as continuous, binary, and time-to-event outcomes with the actual process unknown, we propose a unified approach of maximizing the likelihood or partial likelihood over all of the potential biological processes.
xlink_fit, select_output.
Xu, Wei, and Meiling Hao. 'A unified partial likelihood approach for X-chromosome association on time-to-event outcomes.' Genetic epidemiology 42.1 (2018): 80-94.
xlink_fit returns model fitting results for each SNP with the covariates.
xlink_fit(resp = c(), os = c(), ostime = c(), snps = c(), gender = c(), covars = c(), option = c(type = c(), male = c(), female = c(), MAF_v = 0), model = c(), data)xlink_fit(resp = c(), os = c(), ostime = c(), snps = c(), gender = c(), covars = c(), option = c(type = c(), male = c(), female = c(), MAF_v = 0), model = c(), data)
resp |
Response variable for continuous or binary model fitting. |
os |
Survival indicator, 1 for death, 0 for censoring. |
ostime |
Duration time of survival. |
snps |
SNP name list for model fitting. |
gender |
Gender information must be included in the data. Default setting is male=1 and female=0. If not as default setting, please provide male and female information in the option. |
covars |
Covariates list if needed. |
option |
There are three options. First, type has default 'all', which provides model fitting results for each SNP understanding as 'XCI', 'XCI-E' and 'XCI-S' type respectively. If type is chosen as 'XCI' or 'XCI-E', all the SNPS consider as 'XCI' or 'XCI-E' type in corresponding model. Secondly, if gender is not as default gender setting (male=1,female=0), male and female information should be provided here.The third one, MAF_v is the low bound of the minimum allele frequency, the SNP MAF below this value will not be used in xlink_fit. |
model |
Fitting model. For 'linear', fitting linear model. For 'binary', fitting logistic regression model. For 'survival', fitting survival model. |
data |
Data set. |
It returns estimated parameters, confidence interval and P value for each variable in the chosen model. The baseline and full model maximum likelihood estimation are provided. If type is 'all', best model choice is provided by using AIC as an benchmark.
Xu, Wei, and Meiling Hao. 'A unified partial likelihood approach for X-chromosome association on time-to-event outcomes.' Genetic epidemiology 42.1 (2018): 80-94.
Han, D., Hao, M., Qu, L., & Xu, W. (2019). A novel model for the X-chromosome inactivation association on survival data. Statistical Methods in Medical Research.
lm{stats} for linear model, glm{stats} for logistic regression model, and coxph{survival} for survival model.
Covars<-c("Age","Smoking","Treatment") SNPs<-c("snp_1","snp_2","snp_3") xlink_fit(os="OS",ostime="OS_time",snps=SNPs,gender="gender",covars=Covars, option =list(MAF_v=0.05),model="survival",data = Rdata) xlink_fit(resp="OS_time",snps=SNPs,gender="gender",option =list(type="XCI",MAF_v=0.05), model="linear",data = Rdata)Covars<-c("Age","Smoking","Treatment") SNPs<-c("snp_1","snp_2","snp_3") xlink_fit(os="OS",ostime="OS_time",snps=SNPs,gender="gender",covars=Covars, option =list(MAF_v=0.05),model="survival",data = Rdata) xlink_fit(resp="OS_time",snps=SNPs,gender="gender",option =list(type="XCI",MAF_v=0.05), model="linear",data = Rdata)