Packaging and containerization of computational methods

Mohammed Alser, Brendan Lawlor, Richard J. Abdill, Sharon Waymost, Ram Ayyala, Neha Rajkumar, Nathan LaPierre, Jaqueline Brito, André M. Ribeiro-dos-Santos, Nour Almadhoun, Varuni Sarwal, Can Firtina, Tomasz Osinski, Eleazar Eskin, Qiyang Hu, Derek Strong, Byoung-Do (B.D) Kim, Malak S. Abedalthagafi, Onur Mutlu, Serghei Mangul

Published: 2024-04-01 DOI: 10.1038/s41596-024-00986-0

omics data analysis

software packaging

containerization

usability

sustainable distribution

AI 解读

Extended

Extended Data Fig. 1 General overview of the installation process for omics software tools.

When biomedical researchers need to use omics software tools and reproduce reported results, they first need to locate the version number and web address of each omics tool using the published research paper and supplementary information. They can then download the omics tools, determine each tool’s dependencies using information provided by the tool’s developers, and try to install the tools on their personal computer or HPC cluster. If the tool is successfully installed, the researchers download the relevant omics data and apply the tools as needed. However, even when the hardware resource requirements (CPU type, memory capacity, and storage) of each tool are met, some omics tools are likely to fail when exact reproduction is attempted because of installation challenges.

Extended Data Fig. 2 Development timeline and brief description of popular (a) package managers and (b) containers.

Each tool is described with key information regarding its functionality, purpose, and supported operating system. In addition to the surveyed package managers and containers, the first package manager, PMS, and the first container, FreeBSD Jail, are shown.

Extended Data Fig. 3 Standard workflow for installing software with a package manager.

The user, usually an administrator, asks the package manager to install a specific piece of software. If the software is not already installed, the package manager fetches the appropriate package from a repository. If any of the dependencies are not already installed, the package manager retrieves the dependency’s package from the repository and starts the installation procedure for that package. Once all the dependencies are installed, the initially requested software is installed. The package manager often goes through several iterations of this process, because every dependency can have its own list of dependencies, in which case each of the dependency’s dependencies must be verified through the same process.

Extended Data Fig. 4 Standard workflow for running software with containerization.

The user asks to install a specific container image. If the container image is available locally, then the user can run it directly through the container engine. Potential dependencies are already handled without any intervention from the user. If the software image is not available locally, the appropriate image must be fetched from a repository.

Supplementary information

Supplementary Information

Supplementary Methods for Tables 1–3.

Reporting Summary

Generation of mouse–human chimeric embryos

Genome-wide quantification of transcription factor binding at single-DNA-molecule resolution using methyl-transferase footprinting

Generation of CRISPR–Cas9-mediated genetic knockout human intestinal tissue–derived enteroid lines by lentivirus transduction and single-cell cloning

A puromycin-dependent activity-based sensing probe for histochemical staining of hydrogen peroxide in cells and animal tissues

References

Disconnect between the user and the solution
Addressing the challenges with packaging and containerization
- Glossary
Software packaging facilitates ease of use of software tools
Centralized package managers
Distributed package managers
- Advantages and limitations of package managers for computational tools
Software containerization offers portable, isolated environments that work the same way everywhere
Packaging and containerization increase the usability of omics tools
Packaging and containerization increase the reproducibility of omics tools
Conclusions
Acknowledgements
Extended Data Fig. 1 General overview of the installation process for omics software tools.
Extended Data Fig. 2 Development timeline and brief description of popular (a) package managers and (b) containers.
Extended Data Fig. 3 Standard workflow for installing software with a package manager.
Extended Data Fig. 4 Standard workflow for running software with containerization.

Alser, M. et al. Technology dictates algorithms: recent developments in read alignment. Genome Biol. 22, 249 (2021).
Mangul, S. et al. Systematic benchmarking of omics computational tools. Nat. Commun. 10, 1393 (2019).
Alser, M., Eudine, J. & Mutlu, O. Genome-on-diet: taming large-scale genomic analyses via sparsified genomics. Preprint at arXiv http://doi.org/10.48550/arXiv.2211.08157 (2022).
Meyer, F. et al. Critical assessment of metagenome interpretation: the second round of challenges. Nat. Methods 19, 429–440 (2022).
Cox, R. Surviving software dependencies. Commun. ACM 62, 36–43 (2019).
Mangul, S., Martin, L. S., Eskin, E. & Blekhman, R. Improving the usability and archival stability of bioinformatics software. Genome Biol. 20, 47 (2019).
Mangul, S. et al. Challenges and recommendations to improve the installability and archival stability of omics computational tools. PLoS Biol. 17, e3000333 (2019).
Begley, C. G., Buchan, A. M. & Dirnagl, U. Robust research: institutions must do their part for reproducibility. Nature 525, 25–27 (2015).
Wratten, L., Wilm, A. & Göke, J. Reproducible, scalable, and shareable analysis pipelines with bioinformatics workflow managers. Nat. Methods http://doi.org/10.1038/s41592-021-01254-9 (2021).
Brito, J. J. et al. Recommendations to enhance rigor and reproducibility in biomedical research. Gigascience 9, giaa056 (2020).
Heil, B. J. et al. Reproducibility standards for machine learning in the life sciences. Nat. Methods 18, 1132–1135 (2021).
Baker, M. 1,500 scientists lift the lid on reproducibility. Nature 533, 452–454 (2016).
Malloy, B. A. & Power, J. F. An empirical analysis of the transition from Python 2 to Python 3. Empir. Softw. Eng. 24, 751–778 (2019).
Gosden, J. A. Software compatibility. In Proc. December 9–11, 1968, Fall Joint Computer Conference, Part I—AFIPS ’68 (Fall, Part I) http://doi.org/10.1145/1476589.1476605 (ACM Press, 1968).
Abate, P., Di Cosmo, R., Treinen, R. & Zacchiroli, S. A modular package manager architecture. Inf. Softw. Technol. 55, 459–474 (2013).
Decan, A., Mens, T. & Grosjean, P. An empirical comparison of dependency network evolution in seven software packaging ecosystems. Empir. Softw. Eng. 24, 381–416 (2018).
Boettiger, C. An introduction to Docker for reproducible research. ACM SIGOPS Oper. Syst. Rev. 49, 71–79 (2015). 49.
Silver, A. Software simplified. Nature 546, 173–174 (2017).
Dunn, M. C. & Bourne, P. E. Building the biomedical data science workforce. PLoS Biol. 15, e2003082 (2017).
Florance, V. in Informatics Education in Healthcare: Lessons Learned (ed. Berner, E. S.) 125–133 (Springer, 2020).
Bush, W. S., Wheeler, N., Darabos, C. & Beaulieu-Jones, B. in Biocomputing 2022 412–416 (World Scientific, 2021).
Wu, J. et al. Virtual meetings promise to eliminate geographical and administrative barriers and increase accessibility, diversity and inclusivity. Nat. Biotechnol. http://doi.org/10.1038/s41587-021-01176-z (2021).
Siepel, A. Challenges in funding and developing genomic software: roots and remedies. Genome Biol. 20, 147 (2019).
Gardner, P. P. et al. Sustained software development, not number of citations or journal choice, is indicative of accurate bioinformatic software. Genome Biol. 23, 56 (2022).
Hoffman, D. et al. The BOGUS Linux Release http://bogus.org/ (2003)
Fernández-Sanguino, J. et al. A Brief History of Debian Ch. 4 http://www.debian.org/doc/manuals/project-history/detailed.en.html (2023).
Gunthorpe, J. APT User’s Guide http://www.debian.org/doc/manuals/apt-guide/index.en.html (1998).
Leonard, T. Introduction. Zero Install Docs http://docs.0install.net/basics/ (CERN Web Services, 2003).
Conda documentation. Conda http://docs.conda.io/en/latest/ (2017).
Bicking, I. pip 24.0. PyPI http://pypi.org/project/pip/ (2024).
Parnas, D. L. Designing software for ease of extension and contraction. IEEE Trans. Softw. Eng. SE-5, 128–138 (1979).
Claes, M., Mens, T., Di Cosmo, R. & Vouillon, J. A historical analysis of Debian package incompatibilities. 2015 IEEE/ACM 12th Working Conference on Mining Software Repositories http://doi.org/10.1109/msr.2015.27 (2015).
Dolstra, E., De Jonge, M., Visser, E. & Others. Nix: a safe and policy-free system for software deployment. In LISA 4, 79–92 (2004).
Grüning, B. et al. Bioconda: sustainable and comprehensive software distribution for the life sciences. Nat. Methods 15, 475–476 (2018).
Mancinelli, F. et al. Managing the complexity of large free and open source package-based software distributions. In 21st IEEE/ACM International Conference on Automated Software Engineering (ASE’06) 199–208 (2006).
Gamblin, T. et al. The Spack package manager. In Proc. International Conference for High Performance Computing, Networking, Storage and Analysis on SC ’15. http://doi.org/10.1145/2807591.2807623 (2015).
Hoste, K., Timmerman, J., Georges, A. & De Weirdt, S. EasyBuild: building software with ease. In 2012 SC Companion.: High. Perform. Comput., Netw. Storage Anal. http://doi.org/10.1109/sc.companion.2012.81 (2012).
Dongarra, J. Report on the Fujitsu Fugaku System. Tech. Report No. ICLUT-20-06 (Univ. Tennessee Knoxville Innovative Computing Laboratory, 2020).
Dagnat, F., Simon, G. & Zhang, X. Toward a distributed package management system. In Lococo 2011: Workshop on Logics for Component Configuration (2011).
Kamp, P.-H. & Watson, R. N. M. Jails: confining the omnipotent root. Proc. 2nd Int. SANE Conf. 43, 116 (2000).
Syed, M. H. & Fernandez, E. B. The software container pattern. In Proc. 22nd Conference on Pattern Languages of Programs 24–26 (The Hillside Group, 2015).
da Veiga Leprevost, F. et al. BioContainers: an open-source and community-driven framework for software standardization. Bioinformatics 33, 2580–2582 (2017).
Adair, R. J., Bayles, R. U., Comeau, L. W. & Creasy, R. J. A Virtual Machine System for the 360/40. Tech. Report (International Business Machines Corporation, 1966).
Smith, J. & Nair, R. Virtual Machines: Versatile Platforms for Systems and Processes (Elsevier, 2005).
Angiuoli, S. V. et al. CloVR: a virtual machine for automated and portable sequence analysis from the desktop using cloud computing. BMC Bioinform. http://doi.org/10.1186/1471-2105-12-356 (2011).
Merkel, D. et al. Docker: lightweight linux containers for consistent development and deployment. Linux J. 2014, 2 (2014).
Cook, J. in Docker for Data Science 103–118 (Apress, 2017).
Kurtzer, G. M., Sochat, V. & Bauer, M. W. Singularity: scientific containers for mobility of compute. PLoS ONE 12, e0177459 (2017).
Huang, D., Cui, H., Wen, S. & Huang, C. Security analysis and threats detection techniques on Docker container. In 2019 IEEE 5th International Conference on Computer and Communications (ICCC) 1214–1220 (2019).
Tomar, A., Jeena, D., Mishra, P. & Bisht, R. Docker security: a threat model, attack taxonomy and real-time attack scenario of DoS. In 2020 10th International Conference on Cloud Computing, Data Science and Engineering (Confluence) 150–155 (2020).
Zahid, F., Kuo, M. M. Y. & Sinha, R. Light-weight active security for detecting DDoS attacks in containerised ICPS. In 2021 18th International Conference on Privacy, Security and Trust (PST) 1–5 (2021).
Martin, A., Raponi, S., Combe, T. & Di Pietro, R. Docker ecosystem—vulnerability analysis. Comput. Commun. 122, 30–43 (2018).
Galaxy Community. The Galaxy platform for accessible, reproducible and collaborative biomedical analyses: 2022 update. Nucleic Acids Res. 50, W345–W351 (2022).
Zhang, Z. et al. Uniform genomic data analysis in the NCI Genomic Data Commons. Nat. Commun. 12, 1226 (2021).
Seven Bridges Genomics—the biomedical data analysis company. Seven Bridges http://www.sevenbridges.com (2016).
Hornik, K. The comprehensive R archive network. Wiley Interdiscip. Rev. Comput. Stat. 4, 394–398 (2012).
Lawlor, B. & Sleator, R. D. The democratization of bioinformatics: a software engineering perspective. Gigascience 9, giaa063 (2020).
Shirinbab, S., Lundberg, L. & Casalicchio, E. Performance evaluation of containers and virtual machines when running Cassandra workload concurrently. Concurr. Comput. Pract. Exp. 32, e5693 (2020).
Felter, W., Ferreira, A., Rajamony, R. & Rubio, J. An updated performance comparison of virtual machines and Linux containers. In 2015 IEEE International Symposium on Performance Analysis of Systems and Software (ISPASS) 171–172 (2015).
BioBuilds home. L7 informatics http://l7informatics.com/resource-center/biobuilds-home/ (2018).
Yuen, D. et al. The Dockstore: enhancing a community platform for sharing reproducible and accessible computational protocols. Nucleic Acids Res. 49, W624–W632 (2021).
Belmann, P. et al. Bioboxes: standardised containers for interchangeable bioinformatics software. Gigascience 4, 47 (2015).
Field, D. et al. Open software for biologists: from famine to feast. Nat. Biotechnol. 24, 801–803 (2006).
Yuen, D. et al. ga4gh/tool-registry-service-schemas: 2.0.1. Zenodo http://zenodo.org/doi/10.5281/zenodo.1193735 (2022).
Dagnat, F. & Simon, G. Toward a distributed package management system. In Lococo 2011: Workshop on Logics for Component Configuration (2011).
Gentleman, R. C. et al. Bioconductor: open software development for computational biology and bioinformatics. Genome Biol. 5, R80 (2004).
Collins, F. S. & Tabak, L. A. Policy: NIH plans to enhance reproducibility. Nature 505, 612–613 (2014).
Turkyilmaz-van der Velden, Y., Dintzner, N. & Teperek, M. Reproducibility starts from you today. Patterns 1, 100099 (2020).
FAIR principles GO FAIR http://www.go-fair.org/fair-principles/ (2017).
Bedő, J., Di Stefano, L. & Papenfuss, A. T. Unifying package managers, workflow engines, and containers: computational reproducibility with BioNix. Gigascience 9, giaa121 (2020).
Courtès, L. Functional package management with Guix. Preprint at arXiv http://doi.org/10.48550/arXiv.1305.4584 (2013).