Oratie uitgesproken door Prof. Dr. Suzan Verberne bij de aanvaarding van het ambt van hoogleraar Natural Language Processing aan de Universiteit Leiden op maandag 3 juni 2024_______________________... Show moreOratie uitgesproken door Prof. Dr. Suzan Verberne bij de aanvaarding van het ambt van hoogleraar Natural Language Processing aan de Universiteit Leiden op maandag 3 juni 2024____________________________________________________________Text also in English : Is the search engine of the future a chatbot? Show less
Rede uitgesproken door Prof. dr. Anton Jan van Zonneveld ter gelegenheid van zijn afscheid als hoogleraarNierziekten, in het bijzonder de experimentele vasculaire geneeskunde aan de Universiteit... Show moreRede uitgesproken door Prof. dr. Anton Jan van Zonneveld ter gelegenheid van zijn afscheid als hoogleraarNierziekten, in het bijzonder de experimentele vasculaire geneeskunde aan de Universiteit Leiden op vrijdag 31 mei 2024 Show less
Time-series forecasting through modelling sequences of temporally dependent observations has many industrial and scientific applications. While machine learning models have been widely used to... Show moreTime-series forecasting through modelling sequences of temporally dependent observations has many industrial and scientific applications. While machine learning models have been widely used to create time-series forecasting models, creating efficient and performant time-series forecasting models is a complex task for domain users. Automated Machine Learning (AutoML) is a growing field that aims to make the process of creating machine-learning models accessible for non-machine learning experts. This is achieved by optimising machine learning pipelines automatically. Time-series machine-learning pipelines include various specialised pre-processing steps that are not currently supported by existing AutoML systems. This dissertation investigates how AutoML can be extended to time-series data analysis problems such as time-series forecasting. Several challenges arise when developing specialised AutoML systems for time-series forecasting. For instance, advanced machine-learning pipelines that can extract time-series features and select well-suited machine-learning models need to be developed. Also, extra hyperparameters such as the window size, which shows how many historical data points are helpful, need to be optimised by the AutoML system. This dissertation addresses these issues. We provide a comprehensive overview of the AutoML research field, including hyperparameter optimisation techniques, neural architecture search, and existing AutoML systems. Next, we investigate the use of AutoML for short-term forecasting, single-step ahead time-series forecasting, and multi-step time-series forecasting with time-series features. Show less
The decision for immediate breast reconstruction after mastectomy is a preference-sensitive decision. For preference-sensitive decisions, shared decision making is propagated, but not yet properly... Show moreThe decision for immediate breast reconstruction after mastectomy is a preference-sensitive decision. For preference-sensitive decisions, shared decision making is propagated, but not yet properly implemented. Furthermore, prior studies identified unmet decision support needs in women considering breast reconstruction. With this thesis we aimed to optimally support women in making an informed decision about breast reconstruction. For this purpose, we developed an online patient decision aid (pDA) for breast cancer patients considering immediate breast reconstruction after mastectomy. This thesis describes the development and evaluation of this pDA. In our multicenter randomized controlled trial including 250 participants, we found that women who used the pDA felt better prepared for decision making, than women who received usual care. Other outcomes did not differ between both groups. Show less
Nicotinamide adenine dinucleotide (NAD+) is the substrate used for the introduction of the ubiquitous and highly dynamic PTM in which either one or multiple adenosine diphosphate ribose (ADPr)... Show moreNicotinamide adenine dinucleotide (NAD+) is the substrate used for the introduction of the ubiquitous and highly dynamic PTM in which either one or multiple adenosine diphosphate ribose (ADPr) moieties are covalently attached to a nucleophilic side chain of an specific amino acid in the target protein to regulate cellular pathways including adipogenesis, DNA damage repair and gene expression. A significant fraction of the nucleophilic amino acid functionalities, most recently histidine and tyrosine, have been identified as ADPr-acceptor sites. In this thesis, new methodologies have been developed to synthesize peptide fragments carrying an ADPr modification to investigate ADP-ribosylation on histidine. Show less
Diacylglycerol lipases (DAGLα and DAGLβ) are responsible for the biosynthesis of the endocannabinoid 2-arachidonoylglycerol (2-AG) in the brain and peripheral tissues. Selective DAGLβ inhibitors... Show moreDiacylglycerol lipases (DAGLα and DAGLβ) are responsible for the biosynthesis of the endocannabinoid 2-arachidonoylglycerol (2-AG) in the brain and peripheral tissues. Selective DAGLβ inhibitors have been proposed as a potential treatment for inflammatory diseases with reduced potential for central nervous system (CNS) mediated side effects, but they are currently lacking. To develop DAGLβ selective inhibitors, a fluorescent biochemical assay was optimized and applied in a high-throughput screening (HTS) for DAGLβ. During the HTS, eight hits classified into four distinct chemotypes were identified. Subsequent structure-activity relationship (SAR) studies, focusing on hit 1 and its modifications, revealed a specific group as the modification hotspot crucial for achieving selectivity towards DAGLβ. Through an extensive SAR investigation, focusing on modifying this group, the first-in-class DAGLβ selective inhibitors, LEI-130 and LEI-131, were discovered. Following their discovery, LEI-130 and LEI-131 underwent comprehensive in vitro and in situ profiling studies. These investigations confirmed that LEI-130 and LEI-131 are selective and noncompetitive inhibitors of DAGLβ, effectively reducing inflammation. Show less
Throughout the history of computer science, a major challenge has been how to assert that software is free of bugs and works as intended. Software bugs can lead to serious negative impacts on any... Show moreThroughout the history of computer science, a major challenge has been how to assert that software is free of bugs and works as intended. Software bugs can lead to serious negative impacts on any software system. Throughout the main body of the thesis, we implemented a series of studies on exploring ways to apply formal methods systematically for the verification of complex object-oriented libraries such as the Java Collection Framework. We start with specifying and verifying methods in the java.util.LinkedList class, but we encounter challenges with methods that take an interface type as a parameter. To address this, we proposed to use histories as method calls and returns to completely determine the concrete state of any implementation and thus can be seen as a way to reason about the interface. The executable history-based (EHB) approach embeds histories and attributes directly as Java objects. This approach could be seamlessly integrated in the KeY theorem prover itself. However, the EHB approach still has its limitations, particularly when it comes to reasoning about the heap and properties of user-defined attributes.To mitigate this, we introduce the logical history-based (LHB) approach, which models histories as an external abstract data type with functions. Building on the LHB approach, we have developed a history-based refinement theory for reasoning about hierarchy in object-oriented programs. Show less
In this dissertation we investigate the emotional and psychiatric effects of glucocorticoids (e.g. cortisol, dexamethasone and prednisone). Glucocorticoids are widely used and their possible... Show moreIn this dissertation we investigate the emotional and psychiatric effects of glucocorticoids (e.g. cortisol, dexamethasone and prednisone). Glucocorticoids are widely used and their possible psychiatric side effects are well known. It is still unclear who is susceptible to side effects and what mechanism is behind this. In this thesis we try to better understand the underlying causes. For example, we use a relatively new static method in this research field, which can analyze changes over time better than the usual methods. In a study, we show that in depressed study participants, changes in certain affect items preceded changes in cortisol levels, while in control participants this was the other way around. Furthermore, we explore a hypothesis that may explain the mechanism of the psychiatric side effects of glucocorticoids. This mechanism is also used for a strategy to prevent the psychiatric side effects. The strategy is based on the fact that synthetic glucocorticoids are very similar to the body's own hormone cortisol, but their binding to the two cortisol receptor proteins and their effects may be different. By achieving a balanced activation of the two cortisol proteins when using glucocorticoids, the psychiatric side effects could possibly be prevented. In summary, this thesis provides further insight into the emotional and psychiatric side effects of glucocorticoids, but the search for a better understanding of the neuropsychiatric side effects remains a pressing concern, which will hopefully benefit patient care in the future. Show less
The lipids in the uppermost layer of the skin, the stratum corneum (SC), play an important role in the skin barrier function. The three main SC lipid classes are ceramides (CER), cholesterol and... Show moreThe lipids in the uppermost layer of the skin, the stratum corneum (SC), play an important role in the skin barrier function. The three main SC lipid classes are ceramides (CER), cholesterol and free fatty acids. In inflammatory skin diseases, such as atopic dermatitis and psoriasis, the SC lipid composition is altered compared to the healthy SC composition. The studies presented in this thesis focused on the effect of the CER subclass composition on the lipid organization, molecular arrangement and barrier function of the skin. First, the CER compositional changes were examined in the SC of seborrheic dermatitis patients, an inflammatory skin disease. A particular emphasis was paid to the ratio of two important CER subclasses, sphingosine CER (CER NS) and phytosphingosine CER (CER NP), and its correlation to the skin barrier. In the subsequent chapters lipid model systems with a simple CER subclass composition were used to study the influence of the CER NS: CER NP molar ratio on the barrier function and lipid organization. Show less
This thesis has studied several modalities how to increase the organ utilisation rate. The results in this thesis indicate that the acceptance of kidneys with acute kidney injury stage 1 or 2 will... Show moreThis thesis has studied several modalities how to increase the organ utilisation rate. The results in this thesis indicate that the acceptance of kidneys with acute kidney injury stage 1 or 2 will significantly contribute to the donor pool as AKI kidneys have comparable outcomes and should therefore not be discarded. Clinically relevant biomarkers such as cell-free unmethylated-INS DNA, FMN, GSN, IGFBP3 and IGF2R were identified or explored in the first part of this thesis and may, if analysed and/or validated thoroughly, contribute to a better assessment of organ viability supporting the justified decision whether to accept or decline the donor organ.The second part of this thesis describes different aspects of the organ preservation technique of abdominal normothermic regional perfusion (aNRP). This relatively new machine perfusion technique has been shown to be feasible and safe, however, consensus regarding assessment parameters during perfusion, protocols and outcome measurements is still lacking. Despite of an inspiring surgical enthusiasm and keeninterest to accept this modality as a new standard, a randomised clinical trial is still required and entirely ethically justifiable in order to scientifically demonstrate superiority of this method for each individual abdominal organ comparing it to other successful (ex-situ) preservation and perfusion strategies. If aNRP can be shown to obtain better post transplantation outcomes whilst increasing organ utilisation, it may be the least complex and most cost-effective strategy in organ preservation. On the other hand, aNRP will only be used in DCD donors. As such, uncertainty regarding the quality of higher risk organs from DBD donors will still be evaluated ex-situ during cold and/or warm machine perfusion with the potential to repair or even regenerate injured organs and making them ‘transplantable’ again. Show less
Demand for palliative care is expected to rise due to an ageing population, an increase in chronic diseases, and multimorbidity. Currently, many countries, including the Netherlands, are dealing... Show moreDemand for palliative care is expected to rise due to an ageing population, an increase in chronic diseases, and multimorbidity. Currently, many countries, including the Netherlands, are dealing with a high demand and shortages of professionals in the healthcare sector. Because of the different nature of providing palliative care compared to curative care and the expected increase in demand for palliative care, the focus of this thesis is to gain insight into the emotional impact of providing palliative care on healthcare professionals.This thesis highlights the emotional stressors healthcare professionals encounter in providing palliative care. Building personal relationships with patients bring both rewarding and emotionally experiences, especially when patients are dying. Among palliative care professionals higher burnout rates were found in general settings than in specialised palliative care settings, possibly due to insufficient training and attention to the emotional impact of providing palliative care in the former.Education and group meetings are seen as crucial for raising awareness and reducing stress. This thesis emphasizes the need for team and organisational support to adequately address the wellbeing of healthcare professionals in this growing field. It requires shared responsibility and action from organisations, teams and healthcare professionals to care about the care for healthcare professionals providing palliative care. Show less
Emotions significantly influence our perceptions and interactions with the environment, yet the impact of our emotional states on judging and responding to others remains underexplored. In this... Show moreEmotions significantly influence our perceptions and interactions with the environment, yet the impact of our emotional states on judging and responding to others remains underexplored. In this thesis, I examined how (romantic) attraction influences our attention (Part I), how we judge others’ intentions (Part II), and whether mimicking another facilitates bond formation (Part III). In two empirical chapters, I demonstrate that people have a strong attentional bias towards visual attractiveness and a tendency to indicate interest in establishing a romantic connection to attractive others. In the subsequent empirical chapter, I show that men are more likely to judge women they find attractive as more sexually aroused. Crucially, in a real-life dating study, I show that men who are attracted to another are more likely to assume mutual interest. Further, I introduce a framework suggesting that people align their actions and emotional states with others to foster romantic relationships. The final empirical chapter suggests that mimicking the attraction cues of others enhances bond formation. Overall, this dissertation sheds light on how attractiveness and attraction affect our attention, our social cognition, and the synchronization of actions and emotions, offering insights into the mechanisms underlying romantic connections. Show less
In this thesis, I focused on studying the above- and belowground interactions of J. vulgaris from a plant-soil feedback (hereafter, PSF) perspective. I investigated the temporal variation of... Show moreIn this thesis, I focused on studying the above- and belowground interactions of J. vulgaris from a plant-soil feedback (hereafter, PSF) perspective. I investigated the temporal variation of negative PSF and examined the effects of root-associated bacteria on plant performance and aboveground herbivores. Additionally, I tested the role of PSF in relation to plant population structure and the significance of soil legacy effects in natural conditions. The findings reveal that temporal dynamics in PSF are driven by changes in plant sensitivity and in the soil microbiome. Although bacteria isolated from J. vulgaris roots can negatively affect plant performance, they can also affect aboveground herbivores and other plant species. Consequently, these bacteria may not be suitable for biological control of J. vulgaris. Moreover, I discovered that soil nematodes can mediate plant-plant interactions, but often favoring J. vulgaris. In my field work, I detected soil legacy effects, but seedling recruitment spatial patterns of J. vulgaris were not soil-mediated. The insights gained from studying PSF and above- and belowground interactions have the potential to reshape traditional approaches employed in controlling invasive plants. This thesis emphasizes the importance of transitioning PSF experiments from indoor to outdoor settings considering various influencing factors simultaneously. Show less
De Wet open overheid (Woo) is op 1 mei 2022 in werking getreden. Dit is de eerste uitspraak over de vraag of er op grond van de Woo een verplichting bestaat om documenten te vervaardigen. Het korte... Show moreDe Wet open overheid (Woo) is op 1 mei 2022 in werking getreden. Dit is de eerste uitspraak over de vraag of er op grond van de Woo een verplichting bestaat om documenten te vervaardigen. Het korte antwoord daarop is nee. In deze annotatie wordt eerst de casus en het oordeel van de Afdeling weergegeven. Vervolgens wordt op het oordeel van de Afdeling gereflecteerd in het licht van de roep om een responsiever bestuursrecht en de vraag of verzoeker een andere laagdrempelige wijze had om het gevraagde register te krijgen. Show less
This thesis focuses on data found in the field of computational drug discovery. New insight can be obtained by applying machine learning in various ways and in a variety of domains. Two studies... Show moreThis thesis focuses on data found in the field of computational drug discovery. New insight can be obtained by applying machine learning in various ways and in a variety of domains. Two studies delved into the application of proteochemometrics (PCM), a machine learning technique that can be used to find relations in protein-ligand bioactivity data and then predict using a virtual screen whether compounds that had never been tested on a particular protein, or set of proteins. With this, sets of compounds were suggested for experimental validation that were significant in a myriad of ways. Another study investigated the mutational patterns in cancer, applying a large dataset of mutation data and identifying several motifs in G protein-coupled receptors. The thesis also contains the work done on the Papyrus dataset, a large scale bioactivity dataset that focuses on standardising data for computational drug discovery and providing an out-of-the-box set that can be used in a variety of settings. Show less
The sequence-dependence of biomolecular interactions involving nucleic acids and proteins is essential for numerous processes inside the cell. Insights into the underlying molecular mechanisms have... Show moreThe sequence-dependence of biomolecular interactions involving nucleic acids and proteins is essential for numerous processes inside the cell. Insights into the underlying molecular mechanisms have been obtained using various biochemical and biophysical methods on two different levels — bulk and single-molecule. These have complemented each other as single-molecule studies excel in observing multi-state dynamic interactions, but perform only at low throughput; while bulk studies can probe many different sequences simultaneously, but providing limited kinetic information. To unite the strengths of both levels, we developed high-throughput Single-molecule Parallel Analysis for Rapid eXploration of Sequence space (SPARXS), that allows the study of molecular structure, kinetics and interactions for thousands of different sequences simultaneously at the single-molecule level. We, for the first time, combine single-molecule fluorescence with next-generation Illumina sequencing. As a proof of principle we apply SPARXS to study the sequence-dependent kinetics of the four-way DNA Holliday junction, occurring among others during homologous recombination. Using SPARXS we observe the dynamic behavior of 120,000 Holliday junction molecules covering 3750 distinct core sequences, a result unattainable with previous techniques. Overall, the mechanistic insights obtained using SPARXS will give an entirely new view on the relationship between sequence, structure and function. Show less
A strong correlation exists between abdominal aortopathy and atherosclerosis. In thoracic aortopathy however, the prevalence of atherosclerosis and its role in the etiology of thoracic aortopathy... Show moreA strong correlation exists between abdominal aortopathy and atherosclerosis. In thoracic aortopathy however, the prevalence of atherosclerosis and its role in the etiology of thoracic aortopathy remained unknown. This thesis therefore studied the cardiovascular disease burden within this patient group. These results showed that the prevalence of atherosclerosis (i.e. cardiovascular disease burden), in contrast to abdominal aortopathy, is not increased within the thoracic aortopathy population. Show less
