Grounding Large Language Models (LLMs) in chemical knowledge graphs (KGs) offers a promising way to support synthesis planning, but reliably retrieving information from these complex structures remains a challenge. Therefore, this work addresses that gap by constructing a bipartite KG and evaluating Text2Cypher query generation across both single- and multi-step retrieval tasks. Different prompting strategies were tested, including zero-shot, one-shot with static, random, or embedding-based example selection, and a checklist-driven self-correction pipeline. Results indicate that one-shot prompting is most effective when the exemplar aligns with the query both structurally and logically. When such an exemplar is provided as context to the Cypher generation prompt, self-correction does not yield significant performance gains. Overall, this study introduces a reproducible setup for Text2Cypher experimentation and evaluation.

Deep Neural Networks (DNNs) are renowned for their high accuracy and versatility, which has led to their application in many fields of research, including biology. However, this accuracy often comes at the expense of interpretability, making it challenging to reason about the inner workings of most DNNs. Particularly in biological research, understanding the mechanisms behind specific outcomes is highly valuable. To elucidate the latent space of DNNs in the context of cancer biology, we introduce GONNECT: a Gene Ontology-derived Neural Network for Explainable Cancer Typing. GONNECT incorporates biological prior knowledge from the Gene Ontology (GO) directly into its network architecture, enabling interpretability through model structure. Using an autoencoder framework, we evaluate GONNECT as both encoder and decoder module and demonstrate its ability to learn which biological processes are distinctive for different cancer types. Furthermore, we show how a variant including soft links (GONNECT-SL) can expand on current knowledge by proposing new interactions between biological processes. GONNECT is flexible both in the amount of prior knowledge it incorporates and the set of input genes, and can potentially be applied in modeling of gene perturbation effects and drug target discovery.

Biological aging clocks estimate age from molecular data and provide insights into age-related functional decline. While aging clocks based on bulk transcriptomic data are well-studied, their single-cell counterparts remain limited and underexplored. In this study, we replicate and enhance a recent single-cell RNA-seq aging clock for human immune cells using ElasticNet, improving its performance through refined preprocessing, feature selection, and regularization. We also explore LightGBM to assess nonlinear modeling potential. Our enhanced models reduce prediction error, generalize better across external datasets, and identify biologically relevant genes through SHAP analysis. These findings support the development of accurate, interpretable, cell-type-specific aging clocks using single-cell data.

The aim of this research is to investigate whether physical gene characteristics can predict age-related changes in gene expression. Specifically, we analyze gene length, GC content, distance to the ends of the chromosome, and similar features to determine their connection with differential expression between young and old individuals. Among these features, gene length consistently shows a strong correlation with age-related expression patterns. However, when combined, the selected features do not provide sufficient predictive power to train a classifier capable of exceeding a modest 66% accuracy. These findings highlight the limitations of the current feature set and point toward the need for more complex feature preprocessing steps or biologically relevant features in future predictive models.

Aging is the biological process that changes the body over time. When we age our bodies become more prone to disease and other health risks. But not everyone experiences these changes at the same age. This is because the age of our cells (biological age) does not always match our chronological age (time since birth). Being able to predict someone’s biological age and comparing it to their chronological age can be used to infer if someone is indeed more prone to diseases or other health risks.

Other studies have been able to predict the age of cells by using gene expressions. They explore the number of expressions in young and old individuals to identify genes that are affected by age. What has not yet been explored is how the correlation of gene pairs are affected by age. How genes cooperate can change with age, this can be captured by looking at how genes correlate and how that correlation changes with age. This paper will explore these correlations and answer the following question. By performing a correlation analysis between features of young individuals, and on the same features for old individuals, can we interpret any differences and use those to improve current age prediction models?

During this study we found a lot of gene pairs that have a significant difference in correlation from younger to older individuals. We also identified hub genes that change correlation with many other genes. Using these genes to train a linear regression model we were able to predict the age of cells with a Mean Absolute Error of 9.7835.

Using the hub genes we were not able to improve the current existing linear regression model. But we did identify genes that have earlier been linked to aging. Like LIMD2, but also a lot of ribosomal genes and mitochondrial genes, both of which lose functionality with aging.

Single-cell RNA sequencing (scRNAseq) is a measuring technique of gene expressions in single cells that has allowed researchers to tackle Alzheimer’s disease (AD) in many ways. Single-cell data has been joined with machine learning to classify brain cells as affected by AD. However, not much is known regarding the usage of such classification models in a spatial setting. This paper analyzes how models trained on scRNAseq data can be used to find AD properties of single cells when measuring them with spatially resolved transcriptomics. With that we study the hypothesis that cells labeled as affected by the disease should appear closer to amyloid plaques, than those that are unaffected. To find out if this holds, three models are used to classify single cells spatially and their predictions are analyzed. Two single-cell datasets are used for training, each giving a drastically different classification outcome. The models do not come to a consensus on the hypothesis’ validity either, as the analysis finds no significant correlation between the variables.

Online databases contain extensive collections of (bio)chemical reactions serving as valuable resources for a variety of applications. However, these large datasets often suffer from incomplete reaction data missing, for example, co-reactants and by-products. Machine learning can help to predict these missing molecules in partial reactions. In this study, we adapt an existing transformer model to enhance its capability in completing these incomplete reactions. We retrain the model using a more diverse dataset of atom-balanced ground truth reactions and introduce both soft and hard atom-balance constraints to improve the completeness and chemical validity of the predictions. Our findings indicate that models trained with soft constraints in their loss function do not demonstrate improved balancing performance and require further tuning. Conversely, the implementation of hard atom-balance constraints during constrained beam search, where we restrict predicting tokens that violate the atom-balance of the prediction, effectively improves the performance of transformer-based models in reaction completion tasks. However, this approach also presents the risk of inaccurately balancing reactions; a limitation that is difficult to identify without chemical expertise, underscoring the necessity for reliable ground truth data to evaluate the predictions.

Alzheimer’s Disease is a complex neurodegenerative disorder marked by the abnormal build-up of proteins in the brain. As no cure currently exists, understanding the disease’s cellular mechanisms is essential for advancing diagnostics and treatment. To this end, single-cell RNA sequencing (scRNA-seq) is a method that offers detailed information about the gene activity of individual cells but lacks their spatial context. Conversely, spatial transcriptomics technology preserves the localization of the cells but provides more limited transcriptomic information. To resolve this, we provide a model that predicts a cell’s distance to pathology from single-cell RNA-sequencing data. Additionally, we identify APOE, LYVE1, and SLC17A7 as genes potentially associated with AD-related microglial clustering around plaques.

Understanding the mechanisms of aging can help us live longer and healthier lives. Epigenetic age predictors are machine learning models that use methylation levels at CpG sites to predict the biological age of the cell. Horvath’s linear clock uses 353 CpGs with a median absolute error (MedAE) of 3.530, while the deep learning model AltumAge uses 20,318 CpGs to achieve a MedAE of 2.147. This study explores how to improve the accuracy of age predictors through model architecture selection, hyperparameter optimization, and feature selection. ElasticNet regression with recursive feature elimination achieved a MedAE of 2.820 using 341 CpGs, outperforming Horvath’s clock. The two models shared 95 CpG sites, and gene enrichment analysis revealed that several associated genes are involved in stem cell regulation. Feature importance and model interpretation were performed using SHAP analysis, which indicated that age prediction cannot be captured by a small subset of CpG sites. It was concluded that epigenetics has an influence on stem cells, which was found to be a biomarker of aging. Aging remains a complex process that deep learning models may capture better.

A deeper understanding of Multiple Sclerosis (MS) symptom progression is required for diagnostic accuracy and patient care. Remote monitoring through smartphones can provide continuous insights in the well-being of MS patients. This research aims to explore differences between MS and healthy typing behavior through collected typing behavior data (keystroke dynamics) on smartphones. The main contribution of this work was finding patterns within typing data that distinguish healthy typing behavior from MS with unlabeled time series data, except the diagnosis. A dataset of three combined studies was provided, which included 182 patients suffering from MS and 335 healthy subjects. An autoencoder was fitted on healthy samples to learn the temporal patterns in healthy typing behavior. Then, an anomaly detection method leveraged both the reconstruction loss and embeddings of these samples to determine whether MS samples deviated from healthy behavior. MS subjects were evaluated against healthy subjects through the ROC-AUC on both sample level and subject level. An initial hold out dataset of 20\% of the subjects was left out for validation. The overall performance was poor (AUC 0.6-0.8), because we aimed to find one general method for all MS subjects. Key limitations were the large variance in healthy typing behavior, relatively short window lengths, and difficulty in quantification of the training step of the autoencoder. Despite the difficulty of evaluating the performance of our suggested method, the described methods provide interesting monitoring possibilities for neurologists to enhance MS diagnosis. These monitoring possibilities could extend towards monitoring the effect of treatment, which is the first step towards personalised healthcare in MS.

Antimicrobial resistance (AMR), termed a "silent pandemic" has caused 4.95 million deaths in 2019, with numbers expected to rise. AMR spans human, animal, and environmental sectors, requiring a One Health approach to address this multifaceted global challenge. This dissertation focuses on the under-represented non-clinical sectors and employs the use of metagenomic data to advance AMR research.

The primary focus in AMR research has been on clinical settings, overlooking animals and the environment and leaving data gaps in resource-limited regions. The world of AMR and metagenomic data is first introduced followed by an in-depth review of AMR in non-clinical sectors and the information metagenomic data can provide. The emphasis is on bioinformatic tools, databases, and workflows to support researchers utilising metagenomic data for AMR studies in these sectors.

Moving forward, the wastewater treatment process, including the neglected upstream and downstream freshwater systems, is examined, to assess the microbiome, resistome and mobilome at each stage. Specific differences within every wastewater treatment plant process sector and their role in AMR transmission are identified. Inspired by the natural baseline of antibiotic resistance in soil, a comparative study of the composition of the microbiome, resistome and mobilome in different soil types, from natural to rural soils, is then further presented. Given the limited information on resistance patterns and the effects of geographical and anthropogenic factors, the influence of antibiotic resistance in different soil types is then further explored.

The swine industry, as the largest consumer of antibiotics, raises concerns about the effects of antibiotic use on the gut microbiome of animals. Antibiotics can impact animal health and promote the transmission of AMR to other non-clinical sectors and humans. How antibiotic use affects the fecal microbiome of pigs raised with and without antibiotics is examined to understand the dynamics of antibiotic resistance in the swine industry.

The burden of AMR, particularly in low- and middle-income countries, where resources for infectious disease surveillance are limited, was the inspiration to propose a method for generating metagenomic data in-field and in resource-limited settings, offering a cost-effective solution for outbreak monitoring and pathogen detection.

The main goal of this dissertation is to highlight the under-represented sectors, their significant role in AMR and to promote global inclusivity.

CRISPR (Clustered Regularly Interspaced Short Palindromic Repeats) technology has transformed molecular biology by enabling a strategy for precise, efficient, and relatively simple genome editing. Guided by a small strand of RNA, CRISPR locates specific DNA sequences within the genome and introduces double-strand breaks (DSBs). A typical cell can detect and fix the damage by invoking one of several DNA repair pathways. However, repair is not error-free and often introduces mutations. The mutagenic nature of repair pathways can be leveraged to disrupt genes or regulatory elements with high specificity, providing a powerful tool for gaining insights into gene function. Researchers can also generate datasets of mutations left behind after DSB induction and repair within different genomic contexts to learn more about the mutagenic effects of DNA repair. In this thesis, we explore challenges and novel approaches for analysing large-scale datasets of mutations and gene essentiality generated via CRISPR technology.

The shift to precision medicine in cancer focuses on providing therapies targeting vulnerabilities of each individual patient tumor. This approach involves identifying cancer subtypes and discovering targets, such as genetic interactions, to treat patients who lack effective therapy. While computational tools, especially machine learning methods, are essential to analyze complex high-dimensional molecular data and suggest new candidate treatment strategies, their effectiveness is often questioned due to data-related challenges. Specifically, limitations in data collection result in sparse or biased biological data, hindering accurate decision-making and the identification of correct patterns. This thesis proposes state of the art solutions to learn improved prediction models for precision medicine and beyond by leveraging relevant data that was previously ignored, and addressing issues of data sparsity and bias.

Prediction of gene synthetic lethalities to identify novel therapeutic targets has overlooked sequence similarity, which is both a notable indicator of functional relation and available for every gene pair, unlike sparser data sources often used for this prediction task. Existing models also struggle to generalize beyond known synthetic lethalities due to an over reliance on data affected by prominent biases. Similarly, the stratification of cancer cohorts without effective treatments is challenging due to the small sample sizes of cancer (sub)cohorts such as oncogene-driven cohorts. In addition, stratification might not directly uncover an actionable treatment opportunity. The integration of dense protein sequence similarity and comprehensive drug response data each, together with methodological advances, led to significant improvements and revealed promising therapeutic opportunities.

Although these integrations improved the performance of computational methods, selection bias, a nonrandom sampling of training data, remained a significant issue affecting fair evaluation and generalizability. Thus, this thesis also introduces strategies to evaluate and mitigate the impact on model generalizability and fairness when the selected training data is not representative of the underlying population. We first artificially induce multivariate selection bias by favoring the selection of specific clusters of samples to study the fair evaluation of model generalizability. Then, to mitigate selection bias, we advance semi-supervised learning methods that use unlabeled data to gain insight into the distribution of the population beyond the labeled training data and promote sample diversity to counter confirmation bias typical of existing approaches. Our approaches include bias mitigation designed for specific machine learning models, such as forest ensembles and neural networks, and model-agnostic methods that operate under fewer assumptions. We show that diversity-guided semi-supervised learning strategies outperform existing domain adaptation techniques in the presence of various selection biases.

The computational methods proposed in this thesis enhance therapeutic target discovery in cancer and address selection bias in machine learning to advance precision medicine in cancer and improve the generalizability and fairness of bioinformatics models.

Bacteria are everywhere and play essential roles in Earth's diverse ecosystems and human health. For example, humans harbor a complex and essential gut microbial community comprising thousands of bacterial species (in addition to numerous viruses, fungi, and microbial eukaryotes). This community helps break down and synthesize nutrients, trains the immune system, and keeps pathogens at bay. However, imbalances in this community are associated with several diseases, including obesity, inflammatory bowel syndrome, and recurrent urinary tract infections. Moreover, bacteria can cause deadly infections, and many are developing resistance to our most potent antibiotics. Studying bacteria is thus essential for identifying differences between pathogens and harmless commensals, countering antimicrobial resistance, and understanding their impacts on human health.

To study bacteria, we typically characterize and compare their genomes. The genome comprises all of an organism's hereditary information, and its genes encode the molecular machines necessary for cell function, providing an overview of an organism's capabilities. The hereditary information in genomes additionally enables inferring the organism's evolutionary history, which helps in understanding why specific traits evolved or aid in inferring transmission links in case of an outbreak.

A challenge with comparing large sets of bacterial genomes is the extensive variation in genome content among many species. For example, two Escherichia coli strains can share as little as 50% of their genes. Current computational tools offer biased or incomplete views of genetic variation among strains. This hinders the identification of genotype-phenotype associations, prevents tracking mobile genetic elements, and limits our understanding of the microbial communities they are part of.

The central question of this thesis is how to design computational tools that enable accurate characterization of genetic variation among diverse bacterial genomes. This thesis introduces new algorithms to identify and represent genetic variation using graph data structures. It additionally presents a tool that characterizes strain-specific genetic variation in microbial communities, even in the presence of same-species strain mixtures. Finally, this thesis uses the previously mentioned tools to investigate the role of the gut microbiome in women with recurrent urinary tract infections, offering novel insights into the gut and bladder dynamics of  E. coli.

Collectively, we expect this work to contribute to an improved mechanistic understanding of bacteria's role in human health, help track and counter the spread of antimicrobial resistance, and inform on the development of microbiome-mediated therapeutics.

In summary, the contributions within this thesis advance Alzheimer's research by introducing new computational tools and methods to better understand the genetics of the disease and cellular mechanisms. Additionally, showing that single-cell gene expression can be effectively analyzed in a binary format (expressed or not) simplifies genomic data analysis, making it more accessible, efficient, and applicable to a range of diseases and conditions.

The integration of wearable technology into healthcare is revolutionizing health monitoring by enabling continuous tracking of vital metrics like heart rate and blood sugar. Devices such as smartwatches and glucose monitors empower proactive interventions, reducing hospital visits and personalizing care. For instance, wearables can detect irregular heart rhythms for early cardiovascular disease detection or assist individuals with diabetes in managing glucose levels. These advancements are enabled by technologies like photoplethysmography (PPG), a non-invasive method for real-time monitoring of physiological signals. Continuous monitoring generates time series data that captures dynamic health fluctuations over time. This data allows for identifying irregularities and deviations that isolated measurements might miss. Detecting anomalies, such as abrupt changes in heart rate or prolonged abnormal patterns, is essential for timely interventions in managing chronic conditions like hypertension and cardiovascular diseases.

However, the analysis of time series data introduces challenges. For instance, label scarcity arises because labeling health anomalies requires expert input, which is often infeasible for large datasets. Inter-subject variability becomes a concern as physiological patterns differ significantly across individuals, complicating model generalization. Furthermore, temporal dependencies in time series data add complexity, as observations are sequentially related and anomalies may not manifest as isolated points but as patterns or sequences deviating from normal behavior. Detecting subtle anomalies, minor deviations that accumulate over time but may signal early-stage conditions, becomes particularly challenging due to their resemblance to normal temporal variations and noise in the data. For example, a gradual change in heart rate variability might indicate the onset of an irregular rhythm but could easily blend into inherent variability if not carefully analyzed. Moreover, evaluation metrics for time series data are insufficient, failing to capture the temporal complexities of real-world applications. Consequently, conventional metrics can misrepresent model performance, leading to unreliable or misleading assessments.

This thesis addresses these challenges by advancing time series anomaly detection through innovative methodologies. A key focus is on addressing the limitations of existing evaluation metrics by introducing new evaluation metrics to better capture temporal complexities, ensuring reliable and meaningful performance assessments. Beyond evaluation, this thesis is guided by several core principles to address the challenges inherent in time series data analysis. Central to this is the use of unsupervised representation learning to tackle label scarcity and variability, enabling robust feature extraction from unlabeled data while maintaining generalizability. Finally, the thesis develops strategies for increasing sensitivity to subtle anomalies, providing effective solutions for identifying small yet significant deviations in complex datasets. Together, these contributions present a comprehensive framework for improving anomaly detection systems across diverse applications, bridging theoretical advancements with practical, real-world needs.

The intensive care unit (ICU) is a hospital department where critically ill patients requiring organ support or intensive monitoring are admitted. Nowadays, the care provided in an intensive care unit has advanced so that more patients are being discharged alive. Advances in ICU care have increased patient survival rates, yet ICU admission is still associated with high morbidity and mortality both during and after the stay.

Identifying patients at high risk of complications during ICU admission is crucial. Even though there has been an increasing focus on predictive models, making accurate predictions with the data currently available remains challenging. A non-conventional parameter that appears to have promising value is heart rate variability (HRV), which reflects the fluctuations in time intervals between consecutive heartbeats and can be derived from the electrocardiogram. We investigated whether heart rate variability was able to predict ICU mortality and ICU length of stay. We employed a machine learning approach, assessing three models for each outcome. We used nested cross validation to estimate the performance and optimize and select the final models. Data from 468 adult patients admitted to the ICU for 48 hours or longer were analyzed and nine HRV measures were calculated. Two HRV measures, the power in the high frequency band and the standard deviation of 5-minute average RR intervals, showed a significant difference between the ICU survivors (n=398) and ICU non-survivors (n=71). While individual HRV measures had limited predictive power for ICU mortality, combining HRV with clinical features improved performance. The best performing model was an eXtreme Boosting Gradient classifier that used clinical features in combination with three HRV measures (power in the high frequency band, power in the low frequency band and the ratio between the power in the low frequency and high frequency band) achieving an AUC of 0.76. The models predicting ICU length of stay performed poorly, with the best model achieving a mean absolute error of 5.07 days. These findings suggest that HRV, when combined with clinical features, has potential in predicting ICU mortality, though further research is necessary before clinical implementation.

Monitoring ICU survivors after discharge is equally important, as half of these survivors suffer from Post Intensive Care Syndrome, which negatively impacts their quality of life and increases their healthcare needs. A team of researchers from the ICU conducted a pilot study establishing the feasibility of monitoring fifteen ICU survivors using eHealth, but a larger clinical trial was needed to assess feasibility on a larger scale. Therefore, we aimed to prepare for the ICU Recover Box 2.0 study. We evaluated the results and challenges of the first pilot. The study protocol was revised to include more participants and the use of new smart technology, specifically the Corsano CardioWatch, replacing the non-CE marked devices from the first pilot, which required 24/7 researcher availability. Preparations included finalizing the application to the Medical Ethical Review Committee and thinking out the study logistics, resulting in a comprehensive plan for the conduct of the study. Key takeaways from this process include the recognition that research is an iterative process of
continuous learning, that the purpose of the study must be carefully considered with ethical concerns in mind, and that the increasing importance of data requires careful planning for its security, processing and storage. The METC application has been submitted. Once approved, the study can proceed on a well-prepared basis.

Alzheimer’s disease (AD) is a neurodegenerative disorder prevalent in older adults, leading to loss in memory, cognitive, and executive function. A characteristic feature of AD is the accumulation of amyloid-beta (Aβ) plaques, which are extracellular deposits of Aβ protein primarily found in grey matter. These Aβ deposits can appear in different forms. The six primary Aβ deposits covered in this work are: diffuse plaques, cored plaques, compact plaques, coarse grained plaques, cerebral amyloid angiopathy (CAA), and subpial deposits. It is speculated that some of the plaques types may be more harmful than others. Given that AD primarily affects an older population, the question arises of how individuals with AD differentiate from cognitively healthy centenarians (people over 100 years old). Consequently, this work attempts to classify and analyse different Aβ types present in donated brain tissue of cognitively healthy centenarians who escaped dementia and individuals diagnosed with AD. The main goal is to identify differences between these two cohorts. However, a challenge in this process is that the brain tissues contain many Aβ plaques, making manual identification difficult in terms of time and labor. To address this, a fine-tuned ResNet50 Aβ plaque classifier was developed in this research that was integrated into an Aβ detection pipeline capable of localising plaques. The model was initially pre-trained on the ImageNet dataset through contrastive learning, and subsequently fine-tuned using few-shot learning with a small number of annotated samples (315). The annotations included the six primary Aβ types whose structure are known and well-defined, and three other anomaly Aβ types that
served to filter out irregular plaques in the unlabeled Aβ data. After performing 5-fold cross-validation, the fine-tuned models demonstrated an average accuracy of 85.71% and precision of 89.47% on the primary types. The final classifier used on the unlabeled Aβ dataset was an ensemble model that incorporated majority voting, combining the predictions of the five models trained during cross-validation. Aβ loads were calculated for each primary Aβ type based on the classifier’s predictions. It was observed that across all considered primary Aβ types, the centenarians’ Aβ loads were statistically significantly lower compared to the AD cohort. The lower Aβ load in centenarians also held true across the frontal, temporal, parietal, and occipital cerebral regions for each primary Aβ type. To partially validate the
model’s performance, correlations were computed between the predicted Aβ loads of the primary Aβ types and neuropathological assessments collected from 75 centenarians. These assessments are common in related works and serve as a benchmark for the ensemble model. They include the Thal Aβ phase, which categorizes the distribution of general Aβ in the brain; the Thal CAA stage, which measures the severity of CAA; and CERAD NP scores, which evaluates the spread of neuritic plaques in the brain, which are a subset of cored plaques. Consequently, statistically significant positive correlations were revealed between: the Thal Aβ phase and the Aβ load of all primary types (r ranging 0.59-0.73); the Thal CAA stage and Aβ load of predicted CAA deposits (r = 0.66); and the CERAD NP scores and Aβ load of cored plaques (r = 0.68). Since the correlated types coincide with what the benchmark staging schemes measure, the model’s predictions seems to align with existing literature.