A5084663159- Protein Structure and Dynamics
- Advanced Proteomics Techniques and Applications
- Machine Learning in Bioinformatics
- S100 Proteins and Annexins
- Computational Drug Discovery Methods
- Alzheimer's disease research and treatments
- Machine Learning in Materials Science
- Metabolomics and Mass Spectrometry Studies
- Pharmaceutical Economics and Policy
- Health Systems, Economic Evaluations, Quality of Life
- Enzyme Structure and Function
- Click Chemistry and Applications
- Dementia and Cognitive Impairment Research
- Cell Adhesion Molecules Research
- MicroRNA in disease regulation
- Chemistry and Chemical Engineering
- Pharmaceutical studies and practices
- Genetics, Bioinformatics, and Biomedical Research
- Extracellular vesicles in disease
- Protein Degradation and Inhibitors
- Chemical Synthesis and Analysis
- Bioinformatics and Genomic Networks
- Ubiquitin and proteasome pathways
- Peroxisome Proliferator-Activated Receptors
- Mitochondrial Function and Pathology
Vrije Universiteit Amsterdam
2022-2025
Utrecht University
2023-2025
University of Applied Sciences Utrecht
2024
JGC (Japan)
2023
AstraZeneca (Sweden)
2020-2021
Institute of Bioinformatics
2021
University of Groningen
2021
ORCID
2020
Neurodegenerative dementias are progressive diseases that cause neuronal network breakdown in different brain regions often because of accumulation misfolded proteins the extracellular matrix, such as amyloids or inside neurons other cell types brain. Several diagnostic protein biomarkers body fluids being used and implemented, for Alzheimer's disease. However, there is still a lack co-pathologies causes dementia. Such biofluid-based enable precision medicine approaches diagnosis treatment,...
Abstract Extracellular vesicles (EVs) are membranous structures released by cells into the extracellular space and thought to be involved in cell‐to‐cell communication. While EVs their cargo promising biomarker candidates, sorting mechanisms of proteins remain unclear. In this study, we ask if it is possible determine EV association based on protein sequence. Additionally, what most important determinants for association. We answer these questions with explainable AI models, using human...
Neurofilament light chain (NfL) is an early nonspecific biomarker in neurodegenerative diseases and traumatic brain injury, indicating axonal damage. This work describes the detailed structural characterization of a selected primary calibrator with potential to be used future reference measurement procedure (RMP) development for accurate quantification NfL. As part described workflow, sequence, higher-order structure as well solvent accessibility, hydrogen-bonding profile were assessed under...
Providing an accurate prognosis for individual dementia patients remains a challenge since they greatly differ in rates of cognitive decline. In this study, we used machine learning techniques with the aim to identify cerebrospinal fluid (CSF) biomarkers that predict rate decline within patients. First, longitudinal mini-mental state examination scores (MMSE) 210 were create fast and slow progression groups. Second, trained random forest classifiers on CSF proteomic profiles obtained...
Proteins tend to bury hydrophobic residues inside their core during the folding process provide stability protein structure and prevent aggregation. Nevertheless, proteins do expose some 'sticky' solvent. These can play an important functional role, e.g. in protein-protein membrane interactions. Here, we first investigate how surfaces are by providing three measures for surface hydrophobicity: total area, relative area and-using our MolPatch method-the largest patch. Secondly, analyze...
Activity prediction plays an essential role in drug discovery by directing search of candidates the relevant chemical space. Despite being applied successfully to image recognition and semantic similarity, Siamese neural network has rarely been explored where modelling faces challenges such as insufficient data class imbalance. Here, we present a recurrent model (SiameseCHEM) based on bidirectional long short-term memory architecture with self-attention mechanism, which can automatically...
The ubiquitous availability of genome sequencing data explains the popularity machine learning-based methods for prediction protein properties from their amino acid sequences. Over years, while revising our own work, reading submitted manuscripts as well published papers, we have noticed several recurring issues, which make some reported findings hard to understand and replicate. We suspect this may be due biologists being unfamiliar with learning methodology, or conversely, experts miss...
Proteomics studies have shown differential expression of numerous proteins in dementias but rarely led to novel biomarker tests for clinical use. The Marie Curie MIRIADE project is designed experimentally evaluate development strategies accelerate the validation and ultimate implementation biomarkers practice, using proteomics-based main as experimental case studies. We address several knowledge gaps that been identified field. First, there technology-translation gap different technologies...
Abstract Numerous ligand-based drug discovery projects are based on structure-activity relationship (SAR) analysis, such as Free-Wilson (FW) or matched molecular pair (MMP) analysis. Intrinsically they assume linearity and additivity of substituent contributions. These techniques challenged by nonadditivity (NA) in protein–ligand binding where the change two functional groups one molecule results much higher lower activity than expected from respective single changes. Identifying nonlinear...
Hydrophobic patches on protein surfaces play important functional roles in protein-protein and protein-ligand interactions. Large hydrophobic are also involved the progression of aggregation diseases. Predicting exposed from a sequence has shown to be difficult task. Fine-tuning foundation models allows for adapting model specific nuances new task using much smaller dataset. Additionally, multitask deep learning offers promising solution addressing data gaps, simultaneously outperforming...
Lipoxygenase (LOX) activity provides oxidative lipid metabolites, which are involved in inflammatory disorders and tumorigenesis. Activity-based probes to detect the of LOX enzymes their cellular context provide opportunities explore biology inhibition. Here, we developed Labelox B as a potent covalent inhibitor for one-step activity-based labeling proteins with activity. was used establish an ELISA-based assay affinity capture antibody-based detection specific isoenzymes. Moreover, enabled...
Abstract Extracellular vesicles (EVs) are membranous structures released by cells into the extracellular space and thought to be involved in cell-to-cell communication. While EVs their cargo promising biomarker candidates, protein sorting mechanisms of proteins remain unclear. In this study, we ask if it is possible determine EV association based on sequence. Additionally, what most important determinants for association. We answer these questions with explainable AI models, using human...
Abstract Glial fibrillary acidic protein (GFAP) is a promising biomarker for brain and spinal cord disorders. Recent studies have highlighted the differences in reliability of GFAP measurements different biological matrices. The reason these discrepancies poorly understood as our knowledge protein's 3‐dimensional conformation, proteoforms, aggregation remains limited. Here, we investigate structural properties under conditions. For this, characterized recombinant proteins from various...
Abstract Numerous ligand-based drug discovery projects are based on structure-activity relationship (SAR) analysis, such as Free-Wilson (FW) or matched molecular pair (MMP) analysis. Intrinsically they assume linearity and additivity of substituent contributions. These techniques challenged by nonadditivity (NA) in protein-ligand binding where the change two functional groups one molecule results much higher lower activity than expected from respective single changes. Identifying nonlinear...
Abstract Background Glial fibrillary acidic protein (GFAP) is a promising biomarker for brain and spinal cord disorders. Recent studies have highlighted the differences in reliability of GFAP measurements different biological matrices. The reason these discrepancies poorly understood as our knowledge protein's 3‐dimensional conformation, proteoforms, aggregation remains limited. Method Here, we investigate structural properties under conditions. For this, characterised recombinant proteins...
Hydrophobic patches on protein surfaces play important functional roles in protein-protein and protein-ligand interactions. Large hydrophobic are also involved the progression of aggregation diseases. Predicting exposed from a sequence has been shown to be difficult task. Fine-tuning foundation models allows for adapting model specific nuances new task using much smaller dataset. Additionally, multi-task deep learning offers promising solution addressing data gaps, simultaneously...
While many good textbooks are available on Protein Structure, Molecular Simulations, Thermodynamics and Bioinformatics methods in general, there is no introductory level book for the field of Structural Bioinformatics. This aims to give an introduction into Bioinformatics, which where previous topics meet explore three dimensional protein structures through computational analysis. We provide overview existing techniques, validate, simulate, predict analyse structures. More importantly, it...
Glial fibrillary acidic protein (GFAP) is a promising biomarker for brain and spinal cord disorders. Recent studies have highlighted the differences in reliability of GFAP measurements different biological matrices. The reason these discrepancies poorly understood as our knowledge protein’s 3-dimensional conformation, proteoforms, aggregation remains limited. Here, we investigate structural properties under conditions. For this, characterised recombinant proteins from various suppliers...
Abstract Providing an accurate prognosis for individual dementia patients remains a challenge since they greatly differ in rates of cognitive decline. In this study, we used machine learning techniques to identify cerebrospinal fluid (CSF) biomarkers that predict the rate First, longitudinal follow-up data 210 were create fast and slow progression groups. Secondly, trained random forest classifiers on CSF proteomic profiles obtained well-performing prediction model group (ROC-AUC = 0.82). As...