Presentation Open Access
Interpretable classification of molecular measurements via pathway-induced multiple kernel learning
Cadow Joris
Roadmap
Molecular data classification: use network topology as a regulariser, define meta-features using pathway information Pathway-Induced. Multiple Kernel Learning (PIMKL): concept, pathway induction, multiple kernel learning PIMKL benchmarking: benchmark against other prior knowledge informed methods on multiple breast cancer cohorts.
PIMKL application: detects tumor samples accurately, integrates multiple omics seamlessly, identifies relevant pathways for each data type.
Files
(3.1 MB)
| Name | Size | |
|---|---|---|
|
pimkl_slides_ismbeccb2019.pdf
md5:e56578d909431e88de335e8d50d21489 |
3.1 MB | Download |
44
42
views
downloads
| All versions | This version | |
|---|---|---|
| Views | 44 | 44 |
| Downloads | 42 | 42 |
| Data volume | 129.2 MB | 129.2 MB |
| Unique views | 42 | 42 |
| Unique downloads | 41 | 41 |
Indexed in
- Publication date:
- August 22, 2019
- DOI:
-
- Keyword(s):
- molecular measurements
- Grants:
-
European Commission:
- PrECISE - PERSONALIZED ENGINE FOR CANCER INTEGRATIVE STUDY AND EVALUATION (668858)
- Meeting:
- 27th Conference on Intelligent Systems for Molecular Biology and the 18th European Conference on Computational Biology (ISMB/ECCB 2019), Basel, Switzerland, 21-25 July 2019
- Communities:
- License (for files):
- Creative Commons Attribution 4.0 International