Abstract
Genetic, proteomic, and other biologically derived data sets are often ill-conditioned with many more variables than data records. Furthermore, the variables are often highly correlated as well as coupled. These attributes make such data sets very difficult to analyze with conventional statistical and machine learning techniques. The ParetoGP approach implemented within DataModeler exploring the trade-off between model complexity and accuracy enables attacking such data sets with dual benefits of identifying key variables, associations, and metavariables along with providing concise, explainable, and human-interpretable predictive models. Transparency of key variables, model structures, and response behaviors provide a substantial benefit relative to conventional machine learning and the associated black-box models. In this chapter, we describe the analysis methodology and highlight benefits using available biological data sets.
This is a preview of subscription content, log in via an institution to check access.
Access this chapter
Tax calculation will be finalised at checkout
Purchases are for personal use only
References
Evolved Analytics LLC. DataModeler. Evolved Analytics LLC, Rancho Santa Fe, USA, version 9.5 edition, 9 2021
Dwarshuis, N.J., Song, H.W., Patel, A., Kotanchek, T., Roy, K.: Functionalized microcarriers improve t cell manufacturing by facilitating migratory memory t cell production and increasing cd4/cd8 ratio
Cheng, A., Vantucci, C.E., Krishnan, L., Ruehle, M.A., Kotanchek, T., Wood, L.B., Roy, K., Guldberg, R.E.: Early systemic immune biomarkers predict bone regeneration after trauma. PNAS 118(8) (2021)
Pradhan, P., Chatterjee, P., Stevens, H., Marmon, A., Medrano-Trochez, C., Jimenez, A., Kippner, L., Li, Y., Savage, E., Gaul, D., Fernandez, F., Gibson, G., Kurtzberg, J., Kotanchek, T., Yeago, C., Roy, K.: Multiomic analysis and computational modeling to identify critical quality attributes for immunomodulatory potency of mesenchymal stromal cells. In: International Society Cell and Gene Therapy - Cytotherapy, vol. 23 (2021)
Saplakoglu, U., Kotanchek, T., Marshall, D., Roy, K., Sobecki, S.: Expert roundtable: embracing transformation: how big data, ai and digitization are changing cell and gene therapy manufacture. Bioinsights 503–518 (2021)
Odeh-Couvertier, V.Y., Dwarshuis, N.J., Colonna, M.B., Levine, B.L., Edison, A.S., Kotanchek, T., Roy, K., Torres-Garcia, W.: Predicting t-cell quality during manufacturing through an artificial intelligence-based integrative multiomics analytical platform. Bioeng. Transl, Med (2021)
Maughon, T.S., Shen, X., Shen, X., Huang, D., Adebayo Michael, A.O., Andrew Shockey, W., Andrews, S.H., McRae III, J.M., Platt, M.O., Fernandez, F.M., Edison, A.S., Stice, S.L., Marklein, R.A.: Metabololics amd cytokine profiling of mesenchymal stromal cells idefntify markers predictive of t-cell suppression. ISCT - Cytotherapy 24 (2021)
Fernandez, F.M.: Facundo private communications. Metabolic Data
NSF Engineering Research Center for Cell Manufacturing Technologies. Cell manufacturing technology 9 (2022)
Haldeman-Englert, C., Raymond T. Jr., Novick, T.: University of rochester medical center health encyclopedia cd4-cd8 ratio
Medicine National Academies of Sciences, Engineering. Applying Systems Thinking to Regenerative Medicine: Proceedings of a Workshop. The National Academies Press, Washington, DC (2021)
Hinton, G.E., Salakhutdinov, R.R.: Reducing the dimensionality of reducing the dimensionality of data with neural networks. Science 313, 504–507 (2006)
Keijzer, M.: Improving symbolic regression with interval arithmetic and linear scaling
Kotanchek, M., Smits, G., Kordon, A., Vladislavleva, K., Jordaan, E.: Variable Selection in Industrial Datasets Using Pareto Genetic Programming, volume 9 of Genetic Programming, theory and practice, vo. 6, 1st edn., pp. 79–92. Springer (2006)
Iwadoh, K.: Private communications. Mach Learn Comments
Kotanchek, M., Haut, N.: Back To The Future: Revisiting OrdinalGP and Trustable Models After a Decade, volume 18 of Genetic Programming, Theory and Practice, vol. 7, pp. 129–142. Springer (2022)
Acknowledgements
We would like to thank the leadership and staff of the National Science Foundation Engineering Research Center for Cell Manufacturing Technologies, including Krishnendu Roy, Nathan J. Dwarshuis, Maxwell B. Colonna, Valerie Y. Odeh-Couvertier, Wandaliz Torres-Garcia, and Arthur S. Edison, for their contributions in preparing and providing the CAR-T cell datasets. We thank Facundo M. Fernandez, Alexandria R. Van Group and members of the Marcus Center for Therapeutic Cell Characterization and Manufacturing (MC3M) staff, including Pallab Pradhan, Paramita Chatterjee, Carolyn Yeago, Andrew Marmon, and Annie Boules-Welch, for their contributions and support in providing the MSC datasets. We thank the University of Oregon’s Guldberg Musculoskeletal Research Lab, including Robert E. Guldberg, Albert Cheng, Casey Vantucci, and Kelly Leguineche for their contributions and support in providing the bone regeneration dataset. We would also like to thank Kazuhiro Iwadoh for his insights with issues facing other machine-learning algorithms.
Author information
Authors and Affiliations
Corresponding author
Editor information
Editors and Affiliations
Rights and permissions
Copyright information
© 2023 The Author(s), under exclusive license to Springer Nature Singapore Pte Ltd.
About this chapter
Cite this chapter
Kotanchek, M., Kotanchek, T., Kotanchek, K. (2023). Biological Strategies ParetoGP Enables Analysis of Wide and Ill-Conditioned Data from Nonlinear Systems. In: Trujillo, L., Winkler, S.M., Silva, S., Banzhaf, W. (eds) Genetic Programming Theory and Practice XIX. Genetic and Evolutionary Computation. Springer, Singapore. https://doi.org/10.1007/978-981-19-8460-0_5
Download citation
DOI: https://doi.org/10.1007/978-981-19-8460-0_5
Published:
Publisher Name: Springer, Singapore
Print ISBN: 978-981-19-8459-4
Online ISBN: 978-981-19-8460-0
eBook Packages: Computer ScienceComputer Science (R0)