GWASinspector

Tools for (automated and manual) quality control of the results of Genome Wide Association Studies

When evaluating the results of a genome-wide association study (GWAS), it is important to perform a quality control to ensure that the results are valid, complete, correctly formatted, and, in case of meta-analysis, consistent with other studies that have applied the same analysis. This package was developed to facilitate and streamline this process and provide the user with a comprehensive report. It will also generate cleaned, harmonized GWAS files ready for meta-analysis.

GWASinspector has many features. Detailed instructions on how to use the package can be found in the manual. A sample vignette is also available from here.

Installation

The latest stable version is available from CRAN repository.

License

GWASinspector is licensed under the GNU General Public License, version 3.

Requirements

R 3.2 or higher

Contributors

The following people contributed to the design and development of GWASinspector:

Disclaimer: We hope that GWASinspector will be of significant help in future researches, however, it is published WITHOUT ANY WARRANTY.

Examples

The latest stable version of the package can be installed from CRAN repository using the install.packages function in R, e.g.

install.packages("GWASinspector")

Check if the package is correctly installed and required modules are available on the system

library(GWASinspector)
system.check()

Load the modified configuration file and run the analysis.

## import the QC-configuration file

job <- setup.inspector("/home/user/config.ini")

## check the created instance

job

## run the algorithm

job <- run.inspector(job)

## check the brief report

result.inspector(job)

A demo function is available to explain the package and explore its features.

## report files are saved in the "sample_dir" folder, e.g. "C:/tstFolder"

demo.inspector('/sample_dir')

Download

The prepared reference datasets are available from the below links. SQLite database files should be unzipped.

Allele frequency reference files

Name	File
1000Genome study GRCh37.p13 - 5 populations	1KG-p_3.sqlite.gz¹
1000Genome study GRCh37.p13 - EUR super-population	1KG-p_3_EUR.sqlite.gz²
dbSNP GRCh37.p13	dbSNP_human_9606_b151_GRCh37p13.sqlite.gz³
dbSNP GRCh38.p7	dbSNP_human_9606_b151_GRCh38p7.sqlite.gz⁴
UK10K_20160215 GRCh37	UK10K_20160215.sqlite.gz⁵
TOPMED GRCh37	TOPMED_GRCh37.sqlite.gz⁵
HRC_r1-1 GRCh37	HRC_r1-1_GRCh37.sqlite.gz⁶
HapMap NCBI36	HapMap_CEU_r28_b36_EDIT_v10c_v5.sqlite.gz⁷

¹ Raw files available from here. allele_ref_std_population parameter should be set to one of EAS, AMR, AFR, EUR, SAS in the configuration file.
² Raw files available from here. allele_ref_std_population parameter should be set to EUR in the configuration file.
³ Raw files available from here. allele_ref_std_population parameter should be set to COMMON in the configuration file.
⁴ Raw files available from here. allele_ref_std_population parameter should be set to COMMON in the configuration file.
⁵ Raw files available from here. allele_ref_std_population parameter should be set to COMMON in the configuration file.
⁶ Raw files available from here. allele_ref_std_population parameter should be set to COMMON in the configuration file.
⁷ allele_ref_std_population parameter should be set to COMMON in the configuration file.

Effect-size reference files

Phenotype	RDS file
Heart rate variability traits (SDNN) ^*	beta_SDNN_hg19.rds
Heart rate variability traits (RMSSD) ^*	beta_RMSSD_hg19.rds
Heart rate variability traits (pvRSA/HF) ^*	beta_RSAHF_hg19.rds
Systolic blood pressure^**	beta_ICBP_SBP.rds
Diastolic blood pressure^**	beta_ICBP_DBP.rds
Pulse pressure^**	beta_ICBP_PP.rds

^*Prepared from GWAS catalogue.
^**Prepared from GWAS catalogue.

Frequently asked questions

Q: Can I use this package on Mac OS?
Yes, GWASinspector is a cross platform package and is tested on Windows, Linux and Mac operating systems.

Q: I can't find any HTML report files in the output folder?
First, check the log file for any warnings or errors. Then check if html_report parameter is set as TRUE in the configuraton file. Also, make sure that Pandoc module is installed and available from R by using the system.check() command. Pandoc is a free and open-source document converter which will be installed alongside with RStudio. More information can be found at: https://pandoc.org

Q: Pandoc is installed and available from R, but I get this error in the log file: "Connection can not be made."
This error is due to an access permission problem on your system. Run R or RStudio as administrator.

Q: Why the Excel report file is not generated?
First, check the log file for any warnings or errors. Then check if 'xlsx' is installed by using the system.check() command. 'xlsx' package requires a JAVA environment which might not always be available on a system. For this reason, the user should manually try to install it using this command: install.packages(c('rJava','xlsx'))