Skip to main page content
U.S. flag

An official website of the United States government

Dot gov

The .gov means it’s official.
Federal government websites often end in .gov or .mil. Before sharing sensitive information, make sure you’re on a federal government site.

Https

The site is secure.
The https:// ensures that you are connecting to the official website and that any information you provide is encrypted and transmitted securely.

Access keys NCBI Homepage MyNCBI Homepage Main Content Main Navigation
. 2024 May 9;4(1):vbae067.
doi: 10.1093/bioadv/vbae067. eCollection 2024.

COSGAP: COntainerized Statistical Genetics Analysis Pipelines

Bayram Cevdet Akdeniz  1   2 , Oleksandr Frei  1   2 , Espen Hagen  2 , Tahir Tekin Filiz  2 , Sandeep Karthikeyan  2 , Joëlle Pasman  3 , Andreas Jangmo  4 , Jacob Bergstedt  5 , John R Shorter  6   7 , Richard Zetterberg  6   7 , Joeri Meijsen  6   7 , Ida Elken Sønderby  2   8 , Alfonso Buil  6   7 , Martin Tesli  4 , Yi Lu  3 , Patrick Sullivan  3 , Ole A Andreassen  2   9 , Eivind Hovig  1   10
Affiliations

COSGAP: COntainerized Statistical Genetics Analysis Pipelines

Bayram Cevdet Akdeniz et al. Bioinform Adv. .

Abstract

Summary: The collection and analysis of sensitive data in large-scale consortia for statistical genetics is hampered by multiple challenges, due to their non-shareable nature. Time-consuming issues in installing software frequently arise due to different operating systems, software dependencies, and limited internet access. For federated analysis across sites, it can be challenging to resolve different problems, including format requirements, data wrangling, setting up analysis on high-performance computing (HPC) facilities, etc. Easier, more standardized, automated protocols and pipelines can be solutions to overcome these issues. We have developed one such solution for statistical genetic data analysis using software container technologies. This solution, named COSGAP: "COntainerized Statistical Genetics Analysis Pipelines," consists of already established software tools placed into Singularity containers, alongside corresponding code and instructions on how to perform statistical genetic analyses, such as genome-wide association studies, polygenic scoring, LD score regression, Gaussian Mixture Models, and gene-set analysis. Using provided helper scripts written in Python, users can obtain auto-generated scripts to conduct the desired analysis either on HPC facilities or on a personal computer. COSGAP is actively being applied by users from different countries and projects to conduct genetic data analyses without spending much effort on software installation, converting data formats, and other technical requirements.

Availability and implementation: COSGAP is freely available on GitHub (https://github.com/comorment/containers) under the GPLv3 license.

PubMed Disclaimer

Conflict of interest statement

Dr. Andreassen has received speaker fees from Lundbeck, Janssen, Otsuka, and Sunovion and is a consultant to Cortechs.ai. and Precision Health. Dr. Frei is a consultant to Precision Health.

Figures

Figure 1.
Figure 1.
The diagram for distributed data analysis using COSGAP. COSGAP can be uploaded to each HPC system, allowing users to conduct distributed analysis.
Figure 2.
Figure 2.
An illustrative example of the COSGAP pipeline for conducting GWAS analysis using different tools, such as PLINK and REGENIE. Initially, the data are arranged according to the specified format (https://cosgap.readthedocs.io/en/latest/specifications/README.html) and then run using gwas.py with the specified analysis (currently, it supports PLINK and REGENIE, for others you can use our documentation: https://cosgap.readthedocs.io/en/latest/usecases/README.html). Once gwas.py is run with desired options (in these examples, we run both PLINK and REGENIE with the figures option), scripts to run for PC and HPC are generated and the user can run these scripts without any modification and get the corresponding outputs with quantile–quantile and Manhattan plots.
See this image and copyright information in PMC

References

    1. Alles GR, Carissimi A, Schnorr LM. Assessing the computation and communication overhead of Linux containers for HPC applications. In: 2018 Symposium on High Performance Computing Systems (WSCAD), pp. 116–23. IEEE, 2018.
    1. Corfield EC, Shadrid, AA, Frei O et al. The Norwegian Mother, Father, and Child cohort study (MoBa) genotyping data resource: MoBaPsychGen pipeline v.4. bioRxiv https://www.biorxiv.org/content/10.1101/2022.06.23.496289v4,2024, preprint: not peer reviewed. - DOI
    1. Dagasso G, Yan Y, Wang L et al. Comprehensive-GWAS: a pipeline for genome-wide association studies utilizing cross-validation to assess the predictivity of genetic variations. In: 2020 IEEE International Conference on Bioinformatics and Biomedicine (BIBM).2020, 1361–7.
    1. Frei O, Jangmo A, Hagen E. et al. comorment/containers: Comorment-Containers-v1.8.1 (v1.8.1). Zenodo, 2024. 10.5281/zenodo.10782180. - DOI
    1. Lam M, Awasthi S, Watson HJ. et al. RICOPILI: rapid imputation for COnsortias PIpeLIne. Bioinformatics 2020;36:930–3. - PMC - PubMed