Aleksandar Anžel
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
Research Subjects
Aleksandar Anžel has a total of 29 co-authors. Here are the top co-authors by number of shared publications:
- Georges Hattab
Center for Artificial Intelligence in Public Health Research (ZKI-PH) Robert Koch Institute Berlin Germany
16 shared publications - Akshat Dubey
PH) Center for Artificial Intelligence in Public Health Research (ZKI Robert Koch Institute Nordufer 20 13353 Berlin Germany
6 shared publications - Dominik Heider
Institute of Medical Informatics, University of Münster, Münster, Germany
5 shared publications - Bahar İlGen
Center for Artificial Intelligence in Public Health Research, Robert Koch Institute, Berlin, Germany
3 shared publications Georges Hattab
2 shared publications- Vincent Schilling
Robert-Koch Institute, Center for Artificial Intelligence in Public Health Research (ZKI-PH), Berlin 13353, Germany
2 shared publications
Latest Publications
- Open Access
Network-based characterization of latent co-testing patterns in antimicrobial susceptibility testing
- Open Access
- 9 September 2026
- Published by Elsevier in Machine Learning with Applications
- Vol. 25
UbiQVision: Spatial Dempster–Shafer fusion of XAI attributions for reliable deep vision ensembles
- 9 September 2026
- Published by Elsevier in Machine Learning with Applications
- Vol. 25
Open AccessUbiQVision: Spatial Dempster–Shafer fusion of XAI attributions for reliable deep vision ensembles
- Open Access
- 4 August 2026
- Published by Springer Nature in Discover Computing
- Vol. 29 (1) , 500
Limitations of interaction techniques in modern radial chart implementations
Radial visualization has gained prominence as a versatile approach to exploring complex, multivariate data in information visualization research and practice. However, the interactive potential of radial visualizations remains under-synthesized across existing high-level visualization libraries. In this work, we build upon Kosara et al.’s interaction taxonomy and Draper et al.’s seven radial design patterns to provide a systematic review of peer-reviewed studies and commercial systems, combined with supplementary research to capture significant recent developments beyond prior surveys. We assess the interactive capabilities of these individual tools/systems alongside seven leading high-level visualization libraries and map their support for Focus+Context, Multiple Views, Interaction with Dimensions, and Domain-Specific Methods across a range of radial visual designs. Our analysis reveals that while Focus+Context and Multiple View interactions are increasingly common in modern libraries, Interaction with Dimensions and specialized domain methods remain limited, especially outside standard Star and Concentric layouts. Targeted supplementary research also surfaces numerous contemporary systems that address previously unrecognized gaps. Our work not only reveals these critical gaps, but also identifies practical opportunities and emerging directions for advancing interactivity and methodological rigor in the development of future radial visualization tools and libraries.
- 4 August 2026
- Published by Springer Nature in Discover Computing
- Vol. 29 (1) , 500
Open AccessLimitations of interaction techniques in modern radial chart implementations
Radial visualization has gained prominence as a versatile approach to exploring complex, multivariate data in information visualization research and practice. However, the interactive potential of radial visualizations remains under-synthesized across existing high-level visualization libraries. In this work, we build upon Kosara et al.’s interaction taxonomy and Draper et al.’s seven radial design patterns to provide a systematic review of peer-reviewed studies and commercial systems, combined with supplementary research to capture significant recent developments beyond prior surveys. We assess the interactive capabilities of these individual tools/systems alongside seven leading high-level visualization libraries and map their support for Focus+Context, Multiple Views, Interaction with Dimensions, and Domain-Specific Methods across a range of radial visual designs. Our analysis reveals that while Focus+Context and Multiple View interactions are increasingly common in modern libraries, Interaction with Dimensions and specialized domain methods remain limited, especially outside standard Star and Concentric layouts. Targeted supplementary research also surfaces numerous contemporary systems that address previously unrecognized gaps. Our work not only reveals these critical gaps, but also identifies practical opportunities and emerging directions for advancing interactivity and methodological rigor in the development of future radial visualization tools and libraries.
- Open Access
- 17 July 2026
- Published by Elsevier in Next Research
- Vol. 12
PepTriX: A framework for explainable peptide analysis through protein language models
- 17 July 2026
- Published by Elsevier in Next Research
- Vol. 12
Open AccessPepTriX: A framework for explainable peptide analysis through protein language models
- Open Access
- 18 June 2026
- Published by JMIR Publications Inc. in Journal of Medical Internet Research
- Vol. 28 (1) , e86249
The Open Syndrome Definition as a Machine-Readable Standard for Public Health: Design and Implementation Study
Background: Case definitions are essential for effectively communicating public health threats. However, the absence of a standardized, machine-readable format poses significant challenges to interoperability, epidemiological research, data sharing, and the application of computational methods, including artificial intelligence. These barriers complicate collaboration across regions and organizations and hinder technological progress in public health. Objective: This study aims to propose and release the first open, machine-readable format for representing case and syndrome definitions, together with tools and resources that enable their standardized and scalable use. Methods: We developed the Open Syndrome Definition, a structured, machine-readable schema for representing case and syndrome definitions. We compiled official public health case definitions from multiple institutions and converted them into standardized, machine-readable representations using open-source tools. These tools, available through GitHub under the Massachusetts Institute of Technology license, automate the translation of narrative definitions into structured data. We also created a platform for browsing, analyzing, and contributing new definitions on our initiative website. Results: The Open Syndrome Definition format enabled consistent, automated representation of case definitions across different diseases and jurisdictions. The conversion tools achieved high semantic fidelity, as assessed by qualitative expert review, between narrative and structured representations, supporting human verification and automated analysis. The dataset and accompanying tools demonstrated structural and semantic interoperability by standardizing definitions from various health systems into a unified format and integrating existing medical ontologies through JSON for Linked Data. To further illustrate practical applicability and downstream usage, we introduced a data filtering prototype that allows users to upload their own datasets and verify the results against the standardized definitions. Conclusions: The Open Syndrome Definition establishes a foundation for consistent and machine-readable public health definitions, facilitating reproducible research and interoperability at scale. By enabling systematic data exchange and artificial intelligence–driven analysis, it strengthens public health preparedness and supports more rapid, coordinated responses to emerging health threats.
- 18 June 2026
- Published by JMIR Publications Inc. in Journal of Medical Internet Research
- Vol. 28 (1) , e86249
Open AccessThe Open Syndrome Definition as a Machine-Readable Standard for Public Health: Design and Implementation Study
Background: Case definitions are essential for effectively communicating public health threats. However, the absence of a standardized, machine-readable format poses significant challenges to interoperability, epidemiological research, data sharing, and the application of computational methods, including artificial intelligence. These barriers complicate collaboration across regions and organizations and hinder technological progress in public health. Objective: This study aims to propose and release the first open, machine-readable format for representing case and syndrome definitions, together with tools and resources that enable their standardized and scalable use. Methods: We developed the Open Syndrome Definition, a structured, machine-readable schema for representing case and syndrome definitions. We compiled official public health case definitions from multiple institutions and converted them into standardized, machine-readable representations using open-source tools. These tools, available through GitHub under the Massachusetts Institute of Technology license, automate the translation of narrative definitions into structured data. We also created a platform for browsing, analyzing, and contributing new definitions on our initiative website. Results: The Open Syndrome Definition format enabled consistent, automated representation of case definitions across different diseases and jurisdictions. The conversion tools achieved high semantic fidelity, as assessed by qualitative expert review, between narrative and structured representations, supporting human verification and automated analysis. The dataset and accompanying tools demonstrated structural and semantic interoperability by standardizing definitions from various health systems into a unified format and integrating existing medical ontologies through JSON for Linked Data. To further illustrate practical applicability and downstream usage, we introduced a data filtering prototype that allows users to upload their own datasets and verify the results against the standardized definitions. Conclusions: The Open Syndrome Definition establishes a foundation for consistent and machine-readable public health definitions, facilitating reproducible research and interoperability at scale. By enabling systematic data exchange and artificial intelligence–driven analysis, it strengthens public health preparedness and supports more rapid, coordinated responses to emerging health threats.
- Open Access
- 1 March 2026
- Published by American Association for the Advancement of Science (AAAS) in Computational and Structural Biotechnology Journal
- Vol. 35 (1)
The State of Peptide Detectability in Computational Proteomics and Guidelines for AI Applications
Artificial intelligence (AI) techniques have transformed proteomics and computational biology over the past decade, particularly in mass spectrometry tasks such as fragment spectra prediction, retention time prediction, and peptide detectability prediction. However, as the volume of proteomics data grows, the need for robust, reproducible, and comparable AI applications has become increasingly urgent. Here, we present a comprehensive scoping review of AI techniques for peptide detectability, an essential and rapidly evolving problem in computational proteomics. By analyzing more than 25 peer-reviewed methods published between 2006 and 2025, we identify a persistent discrepancy between increasing algorithmic sophistication and consistent adherence to established machine learning standards. Current studies frequently exhibit heterogeneity in dataset construction, limited transparency in model design and evaluation, and restricted reproducibility. To address these challenges, we propose actionable guidelines focused on transparent technical reporting, rigorous dataset separation, comprehensive performance evaluation, and reproducibility. Encouragingly, recent tools demonstrate that such standards can be met in practice. Furthermore, we outline a future research agenda that emphasizes the integration of structural biology frameworks, the transition toward data-independent acquisition datasets, and the development of explainable AI to ensure models are both biologically interpretable and transferable across evolving instrument platforms. While centered on peptide detectability, these principles offer valuable insights and can inform future work across a wide range of computational proteomics applications.
- 1 March 2026
- Published by American Association for the Advancement of Science (AAAS) in Computational and Structural Biotechnology Journal
- Vol. 35 (1)
Open AccessThe State of Peptide Detectability in Computational Proteomics and Guidelines for AI Applications
Artificial intelligence (AI) techniques have transformed proteomics and computational biology over the past decade, particularly in mass spectrometry tasks such as fragment spectra prediction, retention time prediction, and peptide detectability prediction. However, as the volume of proteomics data grows, the need for robust, reproducible, and comparable AI applications has become increasingly urgent. Here, we present a comprehensive scoping review of AI techniques for peptide detectability, an essential and rapidly evolving problem in computational proteomics. By analyzing more than 25 peer-reviewed methods published between 2006 and 2025, we identify a persistent discrepancy between increasing algorithmic sophistication and consistent adherence to established machine learning standards. Current studies frequently exhibit heterogeneity in dataset construction, limited transparency in model design and evaluation, and restricted reproducibility. To address these challenges, we propose actionable guidelines focused on transparent technical reporting, rigorous dataset separation, comprehensive performance evaluation, and reproducibility. Encouragingly, recent tools demonstrate that such standards can be met in practice. Furthermore, we outline a future research agenda that emphasizes the integration of structural biology frameworks, the transition toward data-independent acquisition datasets, and the development of explainable AI to ensure models are both biologically interpretable and transferable across evolving instrument platforms. While centered on peptide detectability, these principles offer valuable insights and can inform future work across a wide range of computational proteomics applications.
- Open Access
- 28 November 2025
- Published by F1000 Research Ltd
- Vol. 14
Enhancing biomedical time series analysis and visualization through interactive summary polar diagrams
This presentation introduces interactive summary polar diagrams as a novel visual-analytics technique for biomedical time series, transforming complex signals such as EEG into compact polar-coordinate summaries that highlight temporal structure, patterns, and anomalies to support tasks like distinguishing epileptic from non-epileptic patients, enable intuitive model comparison, and enhance exploratory analysis for researchers and clinicians.
- 28 November 2025
- Published by F1000 Research Ltd
- Vol. 14
Open AccessEnhancing biomedical time series analysis and visualization through interactive summary polar diagrams
This presentation introduces interactive summary polar diagrams as a novel visual-analytics technique for biomedical time series, transforming complex signals such as EEG into compact polar-coordinate summaries that highlight temporal structure, patterns, and anomalies to support tasks like distinguishing epileptic from non-epileptic patients, enable intuitive model comparison, and enhance exploratory analysis for researchers and clinicians.
- Open Access
- 7 October 2025
- Published by Oxford University Press (OUP) in NAR Genomics and Bioinformatics
- Vol. 7 (4)
Genomic data representations for horizontal gene transfer detection
Horizontal gene transfer (HGT) accelerates the spread of antimicrobial resistance (AMR) via mobile genetic elements allowing pathogens to acquire resistance genes across species. This process drives the evolution of multidrug-resistant “superbugs” in clinical settings. Detection of HGT is critical to mitigating AMR, but traditional methods based on sequence assembly or comparative genomics lack resolution for complex transfer events. While machine learning (ML) promises improved detection, several studies in other domains have demonstrated that data representations will strongly influence its performance. There is, however, no clear recommendation on the best data representation for HGT detection. Here, we evaluated 44 genomic data representations using five ML models across four data sets. We demonstrate that ML performance is highly dependent on the genomic data representation. The RCKmer-based representation (k = 7) paired with a support vector machine is found to be optimal (F1: 0.959; MCC: 0.908), outperforming other approaches. Moreover, models trained on multi-species data sets are shown to generalize better. Our findings suggest that genomic surveillance benefits from task-specific genome data representations. This work provides state-of-the-art, fine-tuned models for identifying and annotating genomic islands that will enable proper detection of transfer of AMR-related genes between species.
- 7 October 2025
- Published by Oxford University Press (OUP) in NAR Genomics and Bioinformatics
- Vol. 7 (4)
Open AccessGenomic data representations for horizontal gene transfer detection
Horizontal gene transfer (HGT) accelerates the spread of antimicrobial resistance (AMR) via mobile genetic elements allowing pathogens to acquire resistance genes across species. This process drives the evolution of multidrug-resistant “superbugs” in clinical settings. Detection of HGT is critical to mitigating AMR, but traditional methods based on sequence assembly or comparative genomics lack resolution for complex transfer events. While machine learning (ML) promises improved detection, several studies in other domains have demonstrated that data representations will strongly influence its performance. There is, however, no clear recommendation on the best data representation for HGT detection. Here, we evaluated 44 genomic data representations using five ML models across four data sets. We demonstrate that ML performance is highly dependent on the genomic data representation. The RCKmer-based representation (k = 7) paired with a support vector machine is found to be optimal (F1: 0.959; MCC: 0.908), outperforming other approaches. Moreover, models trained on multi-species data sets are shown to generalize better. Our findings suggest that genomic surveillance benefits from task-specific genome data representations. This work provides state-of-the-art, fine-tuned models for identifying and annotating genomic islands that will enable proper detection of transfer of AMR-related genes between species.