Data-driven causal analysis of observational biological time series
Complex systems are challenging to understand, especially when they defy manipulative experiments for practical or ethical reasons. Several fields have developed parallel approaches to infer causal relations from observational time series. Yet, these methods are easy to misunderstand and often contr...
| Main Authors: | , |
|---|---|
| Format: | Article |
| Language: | English |
| Published: |
eLife Sciences Publications Ltd
2022-08-01
|
| Series: | eLife |
| Subjects: | |
| Online Access: | https://elifesciences.org/articles/72518 |
| _version_ | 1797998205792681984 |
|---|---|
| author | Alex Eric Yuan Wenying Shou |
| author_facet | Alex Eric Yuan Wenying Shou |
| author_sort | Alex Eric Yuan |
| collection | DOAJ |
| description | Complex systems are challenging to understand, especially when they defy manipulative experiments for practical or ethical reasons. Several fields have developed parallel approaches to infer causal relations from observational time series. Yet, these methods are easy to misunderstand and often controversial. Here, we provide an accessible and critical review of three statistical causal discovery approaches (pairwise correlation, Granger causality, and state space reconstruction), using examples inspired by ecological processes. For each approach, we ask what it tests for, what causal statement it might imply, and when it could lead us astray. We devise new ways of visualizing key concepts, describe some novel pathologies of existing methods, and point out how so-called ‘model-free’ causality tests are not assumption-free. We hope that our synthesis will facilitate thoughtful application of methods, promote communication across different fields, and encourage explicit statements of assumptions. A video walkthrough is available (Video 1 or https://youtu.be/AlV0ttQrjK8). |
| first_indexed | 2024-04-11T10:44:56Z |
| format | Article |
| id | doaj.art-d7082464f524427e90b7194cdd961aaf |
| institution | Directory Open Access Journal |
| issn | 2050-084X |
| language | English |
| last_indexed | 2024-04-11T10:44:56Z |
| publishDate | 2022-08-01 |
| publisher | eLife Sciences Publications Ltd |
| record_format | Article |
| series | eLife |
| spelling | doaj.art-d7082464f524427e90b7194cdd961aaf2022-12-22T04:29:05ZengeLife Sciences Publications LtdeLife2050-084X2022-08-011110.7554/eLife.72518Data-driven causal analysis of observational biological time seriesAlex Eric Yuan0https://orcid.org/0000-0002-8972-7497Wenying Shou1https://orcid.org/0000-0001-5693-381XMolecular and Cellular Biology PhD program, University of Washington, Seattle, United States; Basic Sciences Division, Fred Hutchinson Cancer Research Center, Seattle, United StatesCentre for Life’s Origins and Evolution, Department of Genetics, Evolution and Environment, University College London, London, United KingdomComplex systems are challenging to understand, especially when they defy manipulative experiments for practical or ethical reasons. Several fields have developed parallel approaches to infer causal relations from observational time series. Yet, these methods are easy to misunderstand and often controversial. Here, we provide an accessible and critical review of three statistical causal discovery approaches (pairwise correlation, Granger causality, and state space reconstruction), using examples inspired by ecological processes. For each approach, we ask what it tests for, what causal statement it might imply, and when it could lead us astray. We devise new ways of visualizing key concepts, describe some novel pathologies of existing methods, and point out how so-called ‘model-free’ causality tests are not assumption-free. We hope that our synthesis will facilitate thoughtful application of methods, promote communication across different fields, and encourage explicit statements of assumptions. A video walkthrough is available (Video 1 or https://youtu.be/AlV0ttQrjK8).https://elifesciences.org/articles/72518time seriescausalitymodel-freesurrogate dataconvergent cross-mappingGranger causality |
| spellingShingle | Alex Eric Yuan Wenying Shou Data-driven causal analysis of observational biological time series eLife time series causality model-free surrogate data convergent cross-mapping Granger causality |
| title | Data-driven causal analysis of observational biological time series |
| title_full | Data-driven causal analysis of observational biological time series |
| title_fullStr | Data-driven causal analysis of observational biological time series |
| title_full_unstemmed | Data-driven causal analysis of observational biological time series |
| title_short | Data-driven causal analysis of observational biological time series |
| title_sort | data driven causal analysis of observational biological time series |
| topic | time series causality model-free surrogate data convergent cross-mapping Granger causality |
| url | https://elifesciences.org/articles/72518 |
| work_keys_str_mv | AT alexericyuan datadrivencausalanalysisofobservationalbiologicaltimeseries AT wenyingshou datadrivencausalanalysisofobservationalbiologicaltimeseries |