Search

Amyotrophic lateral sclerosis (ALS) is a neurodegenerative disease, which primarily affects motor neurons. Years of research have led to the identification of multiple genes involved in ALS. However, the non-coding genome has been largely unexplored. By intersecting open chromatin profiles of healthy human motor cortex with whole genome sequencing data from ALS patients, the neurocalcin delta (NCALD) enhancer region was found to be associated with ALS. An increase in rare variants was found in a 300 base pair region of the NCALD enhancer in healthy individuals compared to ALS patients. This suggests that the rare variants are involved in a protective mechanism towards ALS by increasing motor neuron function and survival. Here, we aim to elucidate the impact of rare variants on NCALD enhancer activity and the role of NCALD enhancer activity on cell morphology and viability in an ALS model. Four rare variants were predicted in silico to impact enhancer activity based on their location in the 300 base pair region and the presence of known transcription factor binding sites. We validated experimentally that three of these rare variants altered NCALD enhancer activity, indicating that they may also affect regulation of gene transcription. To study the effects of gene expression changes in the context of the disease, we set up a simple ALS in vitro model. We found this experimental framework to be robust, replicable, and suitable to use in fundamental ALS research with opportunity of high scalability. Neuroblastoma-derived SH-sy5y cells were differentiated, as they possess neuronal-like properties, are easy to use and to genetically manipulate. We edited the genome of SH-sy5y cells to study effects of gene expression changes in differentiated cells. In basal conditions, neurite length and cell viability were unchanged. In order to mimic ALS, we induced stress in the differentiated cells and aimed to investigate cell morphology and viability. Stress assays were performed t

Understanding the gene transcription rules present in non-coding DNA is essential for unraveling the genetic code that establishes cellular fate. In this study, we aim to narrow down on regulatory regions and motifs within the central nervous system (CNS) that determine cell specificity. While the use of ATAC-seq data has been proven efficient in defining relevant regions of open chromatin, further analysis is required in order to obtain insights into specific regulatory elements. To that end, we propose a strategy involving natural language processing techniques to identify DNA transcription factor (TF) binding sites relevant to each cell type. We employ topic modelling for co-clustering of ATAC-seq peak sequences and cell types; as a result, we can retrieve ‘topics’ consisting of functionally related non-coding DNA regions, that provide a starting point for further analysis and identification of cell-specific feature combinations. Furthermore, we finetune a BigBird language model, pre-trained on the human genome, to distinguish between GABAergic, glutamatergic, and non-neuronal cells. The Byte-Pair Encoding tokenization method allows us to extract the most important DNA motifs for making the class predictions, as well as their corresponding attention scores, which can be mapped back to the peak sequences to identify TF binding sites. We show that this method allows identification of known regulatory elements and propose new strategies to extract more meaningful and specific information from the language models.

ATAC-sequencing is a popular method to measure and sequence the accessible parts of the DNA in a cell. The accessible parts of the DNA are called open chromatin. The closed chromatin is the part of the DNA that is inaccessible. The closed parts of the DNA are bound to nucleosomes, the spool-like proteins that keep DNA organized and compact. These parts are difficult to be accessed by any proteins, and so they can not be reached by the proteins that transcribe DNA and make the cell function as it does. The parts that are not bound to anything, the open parts, play an important role in how the cell behaves. When smaller proteins than nucleosomes are bound to the DNA, this is visible in the ATAC-seq results. These small proteins play a huge part in the behaviour of a cell. A way of analyzing these small bound proteins in the ATAC-seq data is by doing a footprinting analysis. A footprinting analysis could give a lot of information about the behaviour of a cell. When this analysis is done on a diseased cell, information could be obtained about which parts of the DNA and which bound proteins cause a cell to be diseased. This is valuable information for attempts to cure many genetic diseases. Analyzing ATAC-seq data to learn about cell behaviour is still difficult. This paper examines the use of the footprinting analysis on ATAC-seq data by comparing it to other methods used to investigate cell regulation.

Amyotrophic lateral sclerosis (ALS) is a progressive neurodegenerative disease with a large genetic component. Many of the variants seen in ALS patients and controls are so rare that a potential association between the variant and ALS cannot be detected with genome-wide association studies (GWAS). Instead, rare-variant burden (RVB) tests can be used, which combine the signal of all variants in a gene into one signal. A potential limitation of this technique is that the signal of a group of damaging variants can be weakened by the presence of neutral or protective variants in the same gene. Because damaging mutations might occur in different densities across the gene, this research aims to reduce the limitation of testing neutral and damaging mutations together by using two different methods of grouping variants: a functional domain-based method and a spatial clustering method based on the distances between variants. Both methods were tested on three ALS-associated genes: SOD1, FUS and NEK1. The patterns of damaging mutations found in previous studies of SOD1 and FUS were replicated, i.e. no hotspots were seen in SOD1, while robust hotspots of rare and ultra-rare variants were seen in the C-terminus of FUS. In NEK1, two clusters dependent on single intermediate-frequency variants were seen. Additionally, an enrichment of damaging rare variants was found at the N-terminus of NEK1. While the spatial clustering method resulted in more consistent hotspots of ALS variants than the functional-domain based methods, combining both methods strengthens the evidence for hotspots and facilitates the interpretation of the significant results. All in all, this method has the potential to find new hotspots in known ALS genes or new genes that are associated with ALS.

Studies of the pathological features of Amyotrophic Lateral Sclerosis (ALS) implicate anomalous RNA misprocessing with the disease. Here, we investigate motor cortex RNA stability and its genetic underpinning in a cohort of ALS patients. RNA stability captures information about RNA misprocessing and is estimated using total RNA sequencing data. Analyses of RNA stability estimates show that outliers occur disproportionally in neuronal pathways relevant to ALS such as synaptic vesicle recycling and neuron projection regeneration. The genetics underlying RNA stability are studied firstly by relating the most common mutation underlying ALS, C9orf72 expansions, to RNA stability. We find that C9orf72 positive samples generally have lower RNA stabilities. Next, evolutionary scores as well as scores for impact on RNA-binding affinity are calculated for genetic variants. However, relating these scores with RNA stabilities did not yield any significant results. Overall, we demonstrate the importance of using RNA stability for studying ALS and recommend several improvements to the methodology, including the incorporation of micro-RNAs and transcript features into the statistical models, to capitalize on its potential for further discoveries in ALS and other phenotypes.

Studies of the pathological features of Amyotrophic Lateral Sclerosis (ALS) implicate anomalous RNA misprocessing with the disease. Here, we investigate motor cortex RNA stability and its genetic underpinning in a cohort of ALS patients. RNA stability captures information about RNA misprocessing and is estimated using total RNA sequencing data. Analyses of RNA stability estimates show that outliers occur disproportionally in neuronal pathways relevant to ALS such as synaptic vesicle recycling and neuron projection regeneration. The genetics underlying RNA stability are studied firstly by relating the most common mutation underlying ALS, C9orf72 expansions, to RNA stability. We find that C9orf72 positive samples generally have lower RNA stabilities. Next, evolutionary scores as well as scores for impact on RNA-binding affinity are calculated for genetic variants. However, relating these scores with RNA stabilities did not yield any significant results. Overall, we demonstrate the importance of using RNA stability for studying ALS and recommend several improvements to the methodology, including the incorporation of micro-RNAs and transcript features into the statistical models, to capitalize on its potential for further discoveries in ALS and other phenotypes.

A rare disease is an illness that affects less than one in every 2,000 individuals. There are more than 6,000 recognized rare diseases in the European Union. Collectively, rare diseases affect thirty million people in the European Union. Many doctors do not have sufficient experience and knowledge to diagnose such a diverse and rare group of diseases. As a result, rare disease patients often wait for years before receiving a definite diagnosis. Electronic health records of diagnosed patients can guide the diagnosis process of current and future rare disease patients. However, extracting relevant clinical information from millions of EHRs is challenging, especially when most diagnosis information is recorded in unstructured texts. Different clinicians may use different terms to describe the same disease and symptoms. Additionally, contexts, such as negation and cues for familial history, may affect diagnosis interpretation. BERT is one of the current state-of-the-art natural language processing (NLP) models that have been shown to understand linguistic contexts and perform NLP tasks well. This review aims to explore how BERT can improve rare disease diagnosis by processing clinical notes in EHRs. The ability of BERT to learn contextualized embeddings from data helps it to identify important words for rare disease diagnoses, such as symptoms and clinical signs, reliably. Additionally, BERT can also predict the most probable diagnosis given all information recorded in clinical notes. This information can help doctors restrict their diagnosis search space and expedite the diagnosis process of rare diseases. The use of contextualized embedding also allows BERT to be trained with imperfect labels in the fine-tuning phase. This skips the need to use labeled rare disease datasets for BERT fine-tuning process. BERT shows potential to be used in diagnosis support. However, class imbalance and limited training data for certain diseases must be sorted to improve BERT performanc