Cross-validation of correlation networks using modular structure
Abstract Correlation networks derived from multivariate data appear in many applications across the sciences. These networks are usually dense and require sparsification to detect meaningful structure. However, current methods for sparsifying correlation networks struggle with balancing overfitting...
| Main Authors: | , , , |
|---|---|
| Format: | Article |
| Language: | English |
| Published: |
SpringerOpen
2022-11-01
|
| Series: | Applied Network Science |
| Subjects: | |
| Online Access: | https://doi.org/10.1007/s41109-022-00516-5 |
| _version_ | 1811319372023595008 |
|---|---|
| author | Magnus Neuman Viktor Jonsson Joaquín Calatayud Martin Rosvall |
| author_facet | Magnus Neuman Viktor Jonsson Joaquín Calatayud Martin Rosvall |
| author_sort | Magnus Neuman |
| collection | DOAJ |
| description | Abstract Correlation networks derived from multivariate data appear in many applications across the sciences. These networks are usually dense and require sparsification to detect meaningful structure. However, current methods for sparsifying correlation networks struggle with balancing overfitting and underfitting. We propose a module-based cross-validation procedure to threshold these networks, making modular structure an integral part of the thresholding. We illustrate our approach using synthetic and real data and find that its ability to recover a planted partition has a step-like dependence on the number of data samples. The reward for sampling more varies non-linearly with the number of samples, with minimal gains after a critical point. A comparison with the well-established WGCNA method shows that our approach allows for revealing more modular structure in the data used here. |
| first_indexed | 2024-04-13T12:41:22Z |
| format | Article |
| id | doaj.art-d1d6e8b7788e422198322b8b9655d246 |
| institution | Directory Open Access Journal |
| issn | 2364-8228 |
| language | English |
| last_indexed | 2024-04-13T12:41:22Z |
| publishDate | 2022-11-01 |
| publisher | SpringerOpen |
| record_format | Article |
| series | Applied Network Science |
| spelling | doaj.art-d1d6e8b7788e422198322b8b9655d2462022-12-22T02:46:31ZengSpringerOpenApplied Network Science2364-82282022-11-017111110.1007/s41109-022-00516-5Cross-validation of correlation networks using modular structureMagnus Neuman0Viktor Jonsson1Joaquín Calatayud2Martin Rosvall3Integrated Science Lab, Department of Physics, Umeå UniversityIntegrated Science Lab, Department of Physics, Umeå UniversityDepartamento de Biología, Geología, Física y Química inorgánica, Universidad Rey Juan CarlosIntegrated Science Lab, Department of Physics, Umeå UniversityAbstract Correlation networks derived from multivariate data appear in many applications across the sciences. These networks are usually dense and require sparsification to detect meaningful structure. However, current methods for sparsifying correlation networks struggle with balancing overfitting and underfitting. We propose a module-based cross-validation procedure to threshold these networks, making modular structure an integral part of the thresholding. We illustrate our approach using synthetic and real data and find that its ability to recover a planted partition has a step-like dependence on the number of data samples. The reward for sampling more varies non-linearly with the number of samples, with minimal gains after a critical point. A comparison with the well-established WGCNA method shows that our approach allows for revealing more modular structure in the data used here.https://doi.org/10.1007/s41109-022-00516-5Correlation networksModular structureCross-validationInformation theoryGene co-expression |
| spellingShingle | Magnus Neuman Viktor Jonsson Joaquín Calatayud Martin Rosvall Cross-validation of correlation networks using modular structure Applied Network Science Correlation networks Modular structure Cross-validation Information theory Gene co-expression |
| title | Cross-validation of correlation networks using modular structure |
| title_full | Cross-validation of correlation networks using modular structure |
| title_fullStr | Cross-validation of correlation networks using modular structure |
| title_full_unstemmed | Cross-validation of correlation networks using modular structure |
| title_short | Cross-validation of correlation networks using modular structure |
| title_sort | cross validation of correlation networks using modular structure |
| topic | Correlation networks Modular structure Cross-validation Information theory Gene co-expression |
| url | https://doi.org/10.1007/s41109-022-00516-5 |
| work_keys_str_mv | AT magnusneuman crossvalidationofcorrelationnetworksusingmodularstructure AT viktorjonsson crossvalidationofcorrelationnetworksusingmodularstructure AT joaquincalatayud crossvalidationofcorrelationnetworksusingmodularstructure AT martinrosvall crossvalidationofcorrelationnetworksusingmodularstructure |