Jurica Levatić
Current Organizations / Institutes
The organizations/institutes from the scholar's latest article
Previous Organizations / Institutes
The organizations / institutes from the scholar's other articles in the past ten years
Jurica Levatić has a total of 46 co-authors. Here are the top co-authors by number of shared publications:
- Dragi Kocev
Jožef Stefan Institute, Ljubljana, Slovenia
13 shared publications - Sašo Džeroski
Jozef Stefan Institute
8 shared publications - Michelangelo Ceci
University of Bari Aldo Moro, Bari, Italy
6 shared publications Sašo Džeroski
6 shared publications- Fran Supek
Institute for Research in Biomedicine (IRB Barcelona), Barcelona, Spain
4 shared publications - Marjan Stoimchev
Department of Knowledge Technologies, Joef Stefan Institute, Ljubljana, Slovenia
3 shared publications
Latest Publications
- 17 October 2026
- Published by Springer Nature
- p. 817-833
Not All Unlabeled Data Are Equal: Unlabeled Data Selection for Semi-supervised QSAR Modeling with Large Chemical Repositories
- 17 October 2026
- Published by Springer Nature
- p. 817-833
Not All Unlabeled Data Are Equal: Unlabeled Data Selection for Semi-supervised QSAR Modeling with Large Chemical Repositories
- 30 July 2026
- Published by Elsevier in The Journal of Molecular Diagnostics
- Vol. 28 (11) , 849-858
Predicting Recurrence of Stage IA Lung Adenocarcinoma Using Circulating Tumor Cell–Free DNA Whole-Genome Mutational Signatures
- 30 July 2026
- Published by Elsevier in The Journal of Molecular Diagnostics
- Vol. 28 (11) , 849-858
Predicting Recurrence of Stage IA Lung Adenocarcinoma Using Circulating Tumor Cell–Free DNA Whole-Genome Mutational Signatures
- 15 May 2026
- Published by Elsevier in Information Sciences
- Vol. 754
Fully- and semi-supervised hierarchical multi-label image classification with graph learning
- 15 May 2026
- Published by Elsevier in Information Sciences
- Vol. 754
Fully- and semi-supervised hierarchical multi-label image classification with graph learning
- Open Access
- 9 June 2025
- Published by Institute of Electrical and Electronics Engineers (IEEE) in IEEE Journal of Selected Topics in Applied Earth Observations and Remote Sensing
- Vol. 18 (19391404) , 14882-14896
SSL-MAE: Adaptive Semisupervised Learning Framework for Multilabel Classification of Remote Sensing Images Using Masked Autoencoders
The increasing volume of remotely sensed imagery (RSI) requires efficient processing and extraction of meaningful information. Modern deep learning architectures excel in various tasks but typically require large labeled datasets, which are often scarce in RSI due to the tedious labeling of complex heterogeneous landscapes containing multiple semantic categories. This can limit the potential of supervised deep learning methods. To address this, we propose SSL-MAE, a novel semi-supervised learning method based on a masked autoencoder. Our approach unifies self-supervision and discriminative learning within a single, end-to-end framework, leveraging both abundant unlabeled data and limited labeled data. Additionally, we introduce an adaptive mechanism to control the level of supervision during learning, crucial for balancing prediction quality with effective use of unlabeled data. To validate SSL-MAE's effectiveness, we conducted comprehensive experiments on 10 publicly available RSI datasets (5 multi-class and 5 multi-label classification). The results show that our method outperforms state-of-the-art semi-supervised methods and performs favorably against self-supervised methods when labeled data are scarce. Finally, our results suggest that the proposed adaptive joint learning strategy is not restricted to a single design choice; instead, when integrated with different state-of-the-art self-supervised approaches, it significantly enhances their predictive performance.
- 9 June 2025
- Published by Institute of Electrical and Electronics Engineers (IEEE) in IEEE Journal of Selected Topics in Applied Earth Observations and Remote Sensing
- Vol. 18 (19391404) , 14882-14896
Open AccessSSL-MAE: Adaptive Semisupervised Learning Framework for Multilabel Classification of Remote Sensing Images Using Masked Autoencoders
The increasing volume of remotely sensed imagery (RSI) requires efficient processing and extraction of meaningful information. Modern deep learning architectures excel in various tasks but typically require large labeled datasets, which are often scarce in RSI due to the tedious labeling of complex heterogeneous landscapes containing multiple semantic categories. This can limit the potential of supervised deep learning methods. To address this, we propose SSL-MAE, a novel semi-supervised learning method based on a masked autoencoder. Our approach unifies self-supervision and discriminative learning within a single, end-to-end framework, leveraging both abundant unlabeled data and limited labeled data. Additionally, we introduce an adaptive mechanism to control the level of supervision during learning, crucial for balancing prediction quality with effective use of unlabeled data. To validate SSL-MAE's effectiveness, we conducted comprehensive experiments on 10 publicly available RSI datasets (5 multi-class and 5 multi-label classification). The results show that our method outperforms state-of-the-art semi-supervised methods and performs favorably against self-supervised methods when labeled data are scarce. Finally, our results suggest that the proposed adaptive joint learning strategy is not restricted to a single design choice; instead, when integrated with different state-of-the-art self-supervised approaches, it significantly enhances their predictive performance.
- Open Access
- 11 July 2024
- Published by Institute of Electrical and Electronics Engineers (IEEE) in IEEE Transactions on Geoscience and Remote Sensing
- Vol. 62 (01962892) , 1-16
Semi-Supervised Multi-Label Classification of Land Use/Land Cover in Remote Sensing Images With Predictive Clustering Trees and Ensembles
The task of remote sensing image (RSI) classification has been studied extensively in the geoscience and remote sensing (RS) community. While deep learning methods have shown great success in solving this task, their reliance on large-scale labeled datasets is a serious limitation when dealing with complex labels and multiple semantic categories. The process of annotating such datasets can be time-consuming and tedious, leading to limited availability of labeled data and reduced performance of supervised learning methods. To address this issue, semi-supervised learning (SSL) methods can be applied, as they use both the limited labeled data and the abundant unlabeled data. In this article, we propose an effective SSL framework for RSI classification, which combines two key concepts. First, we employ a deep convolutional feature extractor to learn feature representations that encode the images into a lower dimensional feature space, capturing the rich semantic context present in RSI. Second, we utilize semi-supervised predictive clustering trees (PCTs) and ensembles thereof to learn from both the labeled and unlabeled data. To evaluate the effectiveness of the proposed framework, we compare it against several state-of-the-art self-supervised and semi-supervised methods from the literature. We conduct extensive experiments on ten publicly available land use/land cover RSI classification datasets: five for multiclass classification (MCC) and five for multi-label classification (MLC). The results demonstrate that the proposed framework has superior predictive performance compared to state-of-the-art methods from the literature, highlighting its effectiveness in semi-supervised RSI classification.
- 11 July 2024
- Published by Institute of Electrical and Electronics Engineers (IEEE) in IEEE Transactions on Geoscience and Remote Sensing
- Vol. 62 (01962892) , 1-16
Open AccessSemi-Supervised Multi-Label Classification of Land Use/Land Cover in Remote Sensing Images With Predictive Clustering Trees and Ensembles
The task of remote sensing image (RSI) classification has been studied extensively in the geoscience and remote sensing (RS) community. While deep learning methods have shown great success in solving this task, their reliance on large-scale labeled datasets is a serious limitation when dealing with complex labels and multiple semantic categories. The process of annotating such datasets can be time-consuming and tedious, leading to limited availability of labeled data and reduced performance of supervised learning methods. To address this issue, semi-supervised learning (SSL) methods can be applied, as they use both the limited labeled data and the abundant unlabeled data. In this article, we propose an effective SSL framework for RSI classification, which combines two key concepts. First, we employ a deep convolutional feature extractor to learn feature representations that encode the images into a lower dimensional feature space, capturing the rich semantic context present in RSI. Second, we utilize semi-supervised predictive clustering trees (PCTs) and ensembles thereof to learn from both the labeled and unlabeled data. To evaluate the effectiveness of the proposed framework, we compare it against several state-of-the-art self-supervised and semi-supervised methods from the literature. We conduct extensive experiments on ten publicly available land use/land cover RSI classification datasets: five for multiclass classification (MCC) and five for multi-label classification (MLC). The results demonstrate that the proposed framework has superior predictive performance compared to state-of-the-art methods from the literature, highlighting its effectiveness in semi-supervised RSI classification.
- Open Access
- 13 April 2024
- Published by Wiley in International Journal of Intelligent Systems
- Vol. 2024, 1-21
Semi-Supervised Predictive Clustering Trees for (Hierarchical) Multi-Label Classification
Semi-supervised learning (SSL) is a common approach to learning predictive models using not only labeled, but also unlabeled examples. While SSL for the simple tasks of classification and regression has received much attention from the research community, this is not the case for complex prediction tasks with structurally dependent variables, such as multi-label classification and hierarchical multi-label classification. These tasks may require additional information, possibly coming from the underlying distribution in the descriptive space provided by unlabeled examples, to better face the challenging task of simultaneously predicting multiple class labels. In this paper, we investigate this aspect and propose a (hierarchical) multi-label classification method based on semi-supervised learning of predictive clustering trees, which we also extend towards ensemble learning. Extensive experimental evaluation conducted on 24 datasets shows significant advantages of the proposed method and its extension with respect to their supervised counterparts. Moreover, the method preserves interpretability of classical tree-based models.
- 13 April 2024
- Published by Wiley in International Journal of Intelligent Systems
- Vol. 2024, 1-21
Open AccessSemi-Supervised Predictive Clustering Trees for (Hierarchical) Multi-Label Classification
Semi-supervised learning (SSL) is a common approach to learning predictive models using not only labeled, but also unlabeled examples. While SSL for the simple tasks of classification and regression has received much attention from the research community, this is not the case for complex prediction tasks with structurally dependent variables, such as multi-label classification and hierarchical multi-label classification. These tasks may require additional information, possibly coming from the underlying distribution in the descriptive space provided by unlabeled examples, to better face the challenging task of simultaneously predicting multiple class labels. In this paper, we investigate this aspect and propose a (hierarchical) multi-label classification method based on semi-supervised learning of predictive clustering trees, which we also extend towards ensemble learning. Extensive experimental evaluation conducted on 24 datasets shows significant advantages of the proposed method and its extension with respect to their supervised counterparts. Moreover, the method preserves interpretability of classical tree-based models.
- Open Access
Semi-supervised Predictive Clustering Trees for (Hierarchical) Multi-label Classification
Semi-supervised learning (SSL) is a common approach to learning predictive models using not only labeled examples, but also unlabeled examples. While SSL for the simple tasks of classification and regression has received a lot of attention from the research community, this is not properly investigated for complex prediction tasks with structurally dependent variables. This is the case of multi-label classification and hierarchical multi-label classification tasks, which may require additional information, possibly coming from the underlying distribution in the descriptive space provided by unlabeled examples, to better face the challenging task of predicting simultaneously multiple class labels. In this paper, we investigate this aspect and propose a (hierarchical) multi-label classification method based on semi-supervised learning of predictive clustering trees. We also extend the method towards ensemble learning and propose a method based on the random forest approach. Extensive experimental evaluation conducted on 23 datasets shows significant advantages of the proposed method and its extension with respect to their supervised counterparts. Moreover, the method preserves interpretability and reduces the time complexity of classical tree-based models.
Open AccessSemi-supervised Predictive Clustering Trees for (Hierarchical) Multi-label Classification
Semi-supervised learning (SSL) is a common approach to learning predictive models using not only labeled examples, but also unlabeled examples. While SSL for the simple tasks of classification and regression has received a lot of attention from the research community, this is not properly investigated for complex prediction tasks with structurally dependent variables. This is the case of multi-label classification and hierarchical multi-label classification tasks, which may require additional information, possibly coming from the underlying distribution in the descriptive space provided by unlabeled examples, to better face the challenging task of predicting simultaneously multiple class labels. In this paper, we investigate this aspect and propose a (hierarchical) multi-label classification method based on semi-supervised learning of predictive clustering trees. We also extend the method towards ensemble learning and propose a method based on the random forest approach. Extensive experimental evaluation conducted on 23 datasets shows significant advantages of the proposed method and its extension with respect to their supervised counterparts. Moreover, the method preserves interpretability and reduces the time complexity of classical tree-based models.
- Open Access
- 15 December 2022
- Published by Slovenian Association Informatika in Informatica
- Vol. 46 (4)
Semi-supervised learning for structured output prediction
This article presents a summary of the doctoral dissertation of the author on the topic of semi-supervised learning for predicting structured outputs.
- 15 December 2022
- Published by Slovenian Association Informatika in Informatica
- Vol. 46 (4)
Open AccessSemi-supervised learning for structured output prediction
This article presents a summary of the doctoral dissertation of the author on the topic of semi-supervised learning for predicting structured outputs.
- Open Access
- 11 January 2022
- Published by Public Library of Science (PLoS) in PLOS ONE
- Vol. 17 (1) , e0262495
A framework for mutational signature analysis based on DNA shape parameters
The mutation risk of a DNA locus depends on its oligonucleotide context. In turn, mutability of oligonucleotides varies across individuals, due to exposure to mutagenic agents or due to variable efficiency and/or accuracy of DNA repair. Such variability is captured by mutational signatures, a mathematical construct obtained by a deconvolution of mutation frequency spectra across individuals. There is a need to enhance methods for inferring mutational signatures to make better use of sparse mutation data (e.g., resulting from exome sequencing of cancers), to facilitate insight into underlying biological mechanisms, and to provide more accurate mutation rate baselines for inferring positive and negative selection. We propose a conceptualization of mutational signatures that represents oligonucleotides via descriptors of DNA conformation: base pair, base pair step, and minor groove width parameters. We demonstrate how such DNA structural parameters can accurately predict mutation occurrence due to DNA repair failures or due to exposure to diverse mutagens such as radiation, chemical exposure, and the APOBEC cytosine deaminase enzymes. Furthermore, the mutation frequency of DNA oligomers classed by structural features can accurately capture systematic variability in mutagenesis of >1,000 tumors originating from diverse human tissues. A nonnegative matrix factorization was applied to mutation spectra stratified by DNA structural features, thereby extracting novel mutational signatures. Moreover, many of the known trinucleotide signatures were associated with an additional spectrum in the DNA structural descriptor space, which may aid interpretation and provide mechanistic insight. Overall, we suggest that the power of DNA sequence motif-based mutational signature analysis can be enhanced by drawing on DNA shape features.
- 11 January 2022
- Published by Public Library of Science (PLoS) in PLOS ONE
- Vol. 17 (1) , e0262495
Open AccessA framework for mutational signature analysis based on DNA shape parameters
The mutation risk of a DNA locus depends on its oligonucleotide context. In turn, mutability of oligonucleotides varies across individuals, due to exposure to mutagenic agents or due to variable efficiency and/or accuracy of DNA repair. Such variability is captured by mutational signatures, a mathematical construct obtained by a deconvolution of mutation frequency spectra across individuals. There is a need to enhance methods for inferring mutational signatures to make better use of sparse mutation data (e.g., resulting from exome sequencing of cancers), to facilitate insight into underlying biological mechanisms, and to provide more accurate mutation rate baselines for inferring positive and negative selection. We propose a conceptualization of mutational signatures that represents oligonucleotides via descriptors of DNA conformation: base pair, base pair step, and minor groove width parameters. We demonstrate how such DNA structural parameters can accurately predict mutation occurrence due to DNA repair failures or due to exposure to diverse mutagens such as radiation, chemical exposure, and the APOBEC cytosine deaminase enzymes. Furthermore, the mutation frequency of DNA oligomers classed by structural features can accurately capture systematic variability in mutagenesis of >1,000 tumors originating from diverse human tissues. A nonnegative matrix factorization was applied to mutation spectra stratified by DNA structural features, thereby extracting novel mutational signatures. Moreover, many of the known trinucleotide signatures were associated with an additional spectrum in the DNA structural descriptor space, which may aid interpretation and provide mechanistic insight. Overall, we suggest that the power of DNA sequence motif-based mutational signature analysis can be enhanced by drawing on DNA shape features.
- Open Access
Mutational signatures are markers of drug sensitivity of cancer cells
Genomic analyses have revealed mutational signatures that are associated with DNA maintenance gone awry, a common occurrence in tumors. Because cancer therapeutics often target synthesis of DNA building blocks, DNA replication or DNA repair, we hypothesized that mutational signatures would make useful markers of drug sensitivity. We rigorously tested this hypothesis by a global analysis of various drug screening and genetic screening data sets, derived from cancer cell line panels. We introduce a novel computational method that detects mutational signatures in cell lines by stringently adjusting for the confounding germline mutational processes, which are difficult to remove when healthy samples from the same individuals are not available. This revealed many associations between diverse mutational signatures and drug activity in cancer cell lines, which are comparably or more numerous than associations with classical genetic features such as cancer driver mutations or copy number alterations. Validation across independent drug screening data and across genetic screens involving drug target genes revealed hundreds of robustly supported associations, which are provided as a resource for drug repurposing guided by mutational signature markers. We suggest that cancer cells bearing genomic signatures of deficiencies in certain DNA repair pathways may be vulnerable to particular types of therapeutics, such as epigenetic drugs.
Open AccessMutational signatures are markers of drug sensitivity of cancer cells
Genomic analyses have revealed mutational signatures that are associated with DNA maintenance gone awry, a common occurrence in tumors. Because cancer therapeutics often target synthesis of DNA building blocks, DNA replication or DNA repair, we hypothesized that mutational signatures would make useful markers of drug sensitivity. We rigorously tested this hypothesis by a global analysis of various drug screening and genetic screening data sets, derived from cancer cell line panels. We introduce a novel computational method that detects mutational signatures in cell lines by stringently adjusting for the confounding germline mutational processes, which are difficult to remove when healthy samples from the same individuals are not available. This revealed many associations between diverse mutational signatures and drug activity in cancer cell lines, which are comparably or more numerous than associations with classical genetic features such as cancer driver mutations or copy number alterations. Validation across independent drug screening data and across genetic screens involving drug target genes revealed hundreds of robustly supported associations, which are provided as a resource for drug repurposing guided by mutational signature markers. We suggest that cancer cells bearing genomic signatures of deficiencies in certain DNA repair pathways may be vulnerable to particular types of therapeutics, such as epigenetic drugs.