Ioulia Karagiannaki
Current Organizations / Institutes
The organizations/institutes from the scholar's latest article
Research Subjects
Ioulia Karagiannaki has a total of 4 co-authors. Here are the top co-authors by number of shared publications:
- Ioannis Tsamardinos
University of Crete
1 shared publication - Krystallia Gourlia
Laboratory of Translational Oncology, School of Medicine, University of Crete, 71500 Heraklion, Greece
1 shared publication - Vincenzo Lagani
Division of Biomedical Sciences, King Abdullah University of Science and Technology, Thuwal, Saudi Arabia
1 shared publication - Yannis Pantazis
Institution of Applied and Computational Mathematics, Foundation of Research and Technology-Hellas, 70013 Heraklion, Greece
1 shared publication
Latest Publications
- Open Access
- 29 April 2022
- Published by Springer Nature in Machine Learning
- Vol. 112 (11) , 4257-4287
Learning biologically-interpretable latent representations for gene expression data
Molecular gene-expression datasets consist of samples with tens of thousands of measured quantities (i.e., high dimensional data). However, lower-dimensional representations that retain the useful biological information do exist. We present a novel algorithm for such dimensionality reduction called Pathway Activity Score Learning (PASL). The major novelty of PASL is that the constructed features directly correspond to known molecular pathways (genesets in general) and can be interpreted aspathway activity scores. Hence, unlike PCA and similar methods, PASL’s latent space has a fairly straightforward biological interpretation. PASL is shown to outperform in predictive performance the state-of-the-art method (PLIER) on two collections of breast cancer and leukemia gene expression datasets. PASL is also trained on a large corpus of 50000 gene expression samples to construct a universal dictionary of features across different tissues and pathologies. The dictionary validated on 35643 held-out samples for reconstruction error. It is then applied on 165 held-out datasets spanning a diverse range of diseases. The AutoML tool JADBio is employed to show that the predictive information in the PASL-created feature space is retained after the transformation. The code is available athttps://github.com/mensxmachina/PASL.
- 29 April 2022
- Published by Springer Nature in Machine Learning
- Vol. 112 (11) , 4257-4287
Open AccessLearning biologically-interpretable latent representations for gene expression data
Molecular gene-expression datasets consist of samples with tens of thousands of measured quantities (i.e., high dimensional data). However, lower-dimensional representations that retain the useful biological information do exist. We present a novel algorithm for such dimensionality reduction called Pathway Activity Score Learning (PASL). The major novelty of PASL is that the constructed features directly correspond to known molecular pathways (genesets in general) and can be interpreted aspathway activity scores. Hence, unlike PCA and similar methods, PASL’s latent space has a fairly straightforward biological interpretation. PASL is shown to outperform in predictive performance the state-of-the-art method (PLIER) on two collections of breast cancer and leukemia gene expression datasets. PASL is also trained on a large corpus of 50000 gene expression samples to construct a universal dictionary of features across different tissues and pathologies. The dictionary validated on 35643 held-out samples for reconstruction error. It is then applied on 165 held-out datasets spanning a diverse range of diseases. The AutoML tool JADBio is employed to show that the predictive information in the PASL-created feature space is retained after the transformation. The code is available athttps://github.com/mensxmachina/PASL.