Mechanism-free repurposing of drugs for C9orf72-related ALS/FTD using large-scale genomic data.

Repeat expansions in the C9orf72 gene are the most common genetic cause of (ALS) and frontotemporal dementia (FTD). Like other genetic forms of neurodegeneration, pinpointing the precise mechanism(s) by which this mutation leads to neuronal death remains elusive, and this lack of knowledge hampers the development of therapy for C9orf72-related disease. We used an agnostic approach based on genomic data (n = 41,273 ALS and healthy samples, and n = 1,516 C9orf72 carriers) to overcome these bottlenecks. Our drug-repurposing screen, based on gene- and expression-pattern matching and information about the genetic variants influencing onset age among C9orf72 carriers, identified acamprosate, a γ-aminobutyric acid analog, as a potentially repurposable treatment for patients carrying C9orf72 repeat expansions. We validated its neuroprotective effect in cell models and showed comparable efficacy to riluzole, the current standard of care. Our work highlights the potential value of genomics in repurposing drugs in situations where the underlying pathomechanisms are inherently complex. VIDEO ABSTRACT.

Investigating the impact of socioeconomic status on amyotrophic lateral sclerosis.

Amyotrophic lateral sclerosis (ALS) is a neurodegenerative disease characterized by the gradual death of motor neurons in the brain and spinal cord, leading to fatal paralysis. Socioeconomic status (SES) is a measure of an individual's shared economic and social status, which has been shown to have an association with health outcomes. Understanding the impact of SES on health conditions is crucial, as it can influence and be influenced by health-related variables. The role of socioeconomic status in influencing the risk and progression of ALS has not been established, and understanding the various factors that impact ALS is important in developing strategies for treatment and prevention. To investigate this relationship, we recruited 413 participants with definite, probable, or possible ALS according to the El Escorial criteria, from three tertiary centers in London, Sheffield, and Birmingham. Logistic regression was used to examine the association between case-control status, socioeconomic criteria, and ALS risk. Linear regression was used to examine the association between age of onset and socioeconomic variables. Two sensitivity analyses were performed, one using an alternative occupational classifier, and the other using Mendelian Randomization analysis to examine association. There was no significant relationship between any variables and ALS risk. We found an inverse relationship between mean lifetime salary and age of ALS onset (Beta = -0.157, p = 0.011), but no effect of education or occupation on the age of onset. The finding was confirmed in both sensitivity analyses and in Mendelian Randomization. We find that a higher salary is associated with a younger age of ALS onset taking into account sex, occupation, years of education, and clinical presentation.

predicTTE: An accessible and optimal tool for time-to-event prediction in neurological diseases.

Time-to-event prediction is a key task for biological discovery, experimental medicine, and clinical care. This is particularly true for neurological diseases where development of reliable biomarkers is often limited by difficulty visualising and sampling relevant cell and molecular pathobiology. To date, much work has relied on Cox regression because of ease-of-use, despite evidence that this model includes incorrect assumptions. We have implemented a set of deep learning and spline models for time-to-event modelling within a fully customizable 'app' and accompanying online portal, both of which can be used for any time-to-event analysis in any disease by a non-expert user. Our online portal includes capacity for end-users including patients, Neurology clinicians, and researchers, to access and perform predictions using a trained model, and to contribute new data for model improvement, all within a data-secure environment. We demonstrate a pipeline for use of our app with three use-cases including imputation of missing data, hyperparameter tuning, model training and independent validation. We show that predictions are optimal for use in downstream applications such as genetic discovery, biomarker interpretation, and personalised choice of medication. We demonstrate the efficiency of an ensemble configuration, including focused training of a deep learning model. We have optimised a pipeline for imputation of missing data in combination with time-to-event prediction models. Overall, we provide a powerful and accessible tool to develop, access and share time-to-event prediction models; all software and tutorials are available at www.predictte.org.

Harnessing transcriptomic signals for amyotrophic lateral sclerosis to identify novel drugs and enhance risk prediction.

Introduction: Amyotrophic lateral sclerosis (ALS) is a fatal neurodegenerative disease. This study integrates common genetic association results from the latest ALS genome-wide association study (GWAS) summary statistics with functional genomic annotations with the aim of providing mechanistic insights into ALS risk loci, inferring drug repurposing opportunities, and enhancing prediction of ALS risk and clinical characteristics. Methods: Genes associated with ALS were identified using GWAS summary statistic methodology including SuSiE SNP-based fine-mapping, and transcriptome- and proteome-wide association study (TWAS/PWAS) analyses. Using several approaches, gene associations were integrated with the DrugTargetor drug-gene interaction database to identify drugs that could be repurposed for the treatment of ALS. Furthermore, ALS gene associations from TWAS were combined with observed blood expression in two external ALS case-control datasets to calculate polytranscriptomic scores and evaluate their utility for prediction of ALS risk and clinical characteristics, including site of onset, age at onset, and survival. Results: SNP-based fine-mapping, TWAS and PWAS identified 118 genes associated with ALS, with TWAS and PWAS providing novel mechanistic insights. Drug repurposing analyses identified six drugs significantly enriched for interactions with ALS associated genes, though directionality could not be determined. Additionally, drug class enrichment analysis showed gene signatures linked to calcium channel blockers may reduce ALS risk, whereas antiepileptic drugs may increase ALS risk. Across the two observed expression target samples, ALS polytranscriptomic scores significantly predicted ALS risk (R 2 = 5.1 %; p-value = 3.2 × 10-27) and clinical characteristics. Conclusions: Functionally-informed analyses of ALS GWAS summary statistics identified novel mechanistic insights into ALS aetiology, highlighted several therapeutic research avenues, and enabled statistically significant prediction of ALS risk.

Mutations in the tail and rod domains of the neurofilament heavy-chain gene increase the risk of ALS.

OBJECTIVE: Neurofilament heavy-chain gene (NEFH) variants are associated with multiple neurodegenerative diseases, however, their relationship with ALS has not been robustly explored. Still, NEFH is commonly included in genetic screening panels worldwide. We therefore aimed to determine if NEFH variants modify ALS risk. METHODS: Genetic data of 11,130 people with ALS and 7,416 controls from the literature and Project MinE were analysed. We performed meta-analyses of published case-control studies reporting NEFH variants, and variant analysis of NEFH in Project MinE whole-genome sequencing data. RESULTS: Fixed-effects meta-analysis found that rare (MAF <1%) missense variants in the tail domain of NEFH increase ALS risk (OR 4.55, 95% CI 2.13-9.71, p < 0.0001). In Project MinE, ultrarare NEFH variants increased ALS risk (OR 1.37 95% CI 1.14-1.63, p = 0.0007), with rod domain variants (mostly intronic) appearing to drive the association (OR 1.45 95% CI 1.18-1.77, pMadsen-Browning = 0.0007, pSKAT-O = 0.003). While in the tail domain, ultrarare (MAF <0.1%) pathogenic missense variants were also associated with higher risk of ALS (OR 1.94, 95% CI 0.86-4.37, pMadsen-Browning = 0.039), supporting the meta-analysis results. Finally, several tail in-frame deletions were also found to affect disease risk, however, both protective and pathogenic deletions were found in this domain, highlighting an intricate architecture that requires further investigation. INTERPRETATION: We showed that NEFH tail missense and in-frame deletion variants, and intronic rod variants are risk factors for ALS. However, they are not variants of large effect, and their functional impact needs to be clarified in further studies. Therefore, their inclusion in routine genetic screening panels should be reconsidered.

Rare and common genetic determinants of mitochondrial function determine severity but not risk of amyotrophic lateral sclerosis.

Amyotrophic lateral sclerosis (ALS) is a fatal neurodegenerative disease involving selective vulnerability of energy-intensive motor neurons (MNs). It has been unclear whether mitochondrial function is an upstream driver or a downstream modifier of neurotoxicity. We separated upstream genetic determinants of mitochondrial function, including genetic variation within the mitochondrial genome or autosomes; from downstream changeable factors including mitochondrial DNA copy number (mtCN). Across three cohorts including 6,437 ALS patients, we discovered that a set of mitochondrial haplotypes, chosen because they are linked to measurements of mitochondrial function, are a determinant of ALS survival following disease onset, but do not modify ALS risk. One particular haplotype appeared to be neuroprotective and was significantly over-represented in two cohorts of long-surviving ALS patients. Causal inference for mitochondrial function was achievable using mitochondrial haplotypes, but not autosomal SNPs in traditional Mendelian randomization (MR). Furthermore, rare loss-of-function genetic variants within, and reduced MN expression of, ACADM and DNA2 lead to ∼50 % shorter ALS survival; both proteins are implicated in mitochondrial function. Both mtCN and cellular vulnerability are linked to DNA2 function in ALS patient-derived neurons. Finally, MtCN responds dynamically to the onset of ALS independently of mitochondrial haplotype, and is correlated with disease severity. We conclude that, based on the genetic measures we have employed, mitochondrial function is a therapeutic target for amelioration of disease severity but not prevention of ALS.

Non-motor symptoms in amyotrophic lateral sclerosis.

OBJECTIVE: While motor symptoms are well-known in ALS, non-motor symptoms are often under-reported and may have a significant impact on quality of life. In this study, we aimed to examine the nature and extent of non-motor symptoms in ALS. METHODS: A 20-item questionnaire was developed covering the domains of autonomic function, sleep, pain, gastrointestinal disturbance, and emotional lability, posted online and shared on social media platforms to target people with ALS and controls. RESULTS: A total of 1018 responses were received, of which 927 were complete from 506 people with ALS and 421 unaffected individuals. Cold limbs (p 1.66 × 10-36), painful limbs (p 6.14 × 10-28), and urinary urgency (p 4.70 × 10-23) were associated with ALS. People with ALS were more likely to report autonomic symptoms, pain, and psychiatric symptoms than controls (autonomic symptoms B = 0.043, p 6.10 × 10-5, pain domain B = 0.18, p 3.72 × 10-11 and psychiatric domain B = 0.173, p 1.32 × 10-4). CONCLUSIONS: Non-motor symptoms in ALS are common. The identification and management of non-motor symptoms should be integrated into routine clinical care for people with ALS. Further research is warranted to investigate the relationship between non-motor symptoms and disease progression, as well as to develop targeted interventions to improve the quality of life for people with ALS.

Analysis of non-motor symptoms in amyotrophic lateral sclerosis.

OBJECTIVE: We investigated non-motor symptoms in ALS using sequential questionnaires; here we report the findings of the second questionnaire. METHODS: A social media platform (Twitter, now known as X) was used to publicize the questionnaires. Data were downloaded from SurveyMonkey and analyzed by descriptive statistics, comparison of means, and regression models. RESULTS: There were 182 people with ALS and 57 controls. The most important non-motor symptoms were cold limbs (60.4% cases, 14% controls, p = 9.67 x 10-10) and appetite loss (29.7% cases, 5.3% controls, p = 1.6 x 10-4). The weaker limb was most likely to feel cold (p = 9.67 x 10-10), and symptoms were more apparent in the evening and night. Appetite loss was reported as due to feeling full and the time taken to eat. People with ALS experienced medium-intensity pain, more usually shock-like pain than burning or cold-like pain, although the most prevalent type of pain was non-differentiated. CONCLUSIONS: Non-motor symptoms are an important feature of ALS. Further investigation is needed to understand their physiological basis and whether they represent phenotypic differences useful for subtyping ALS.

C9orf72 Repeat Expansion Discordance in 6 Multigenerational Kindreds.

Background and Objectives: A hexanucleotide repeat expansion in the noncoding region of the C9orf72 gene is the most common genetically identifiable cause of amyotrophic lateral sclerosis (ALS) and frontotemporal dementia in populations of European ancestry. Pedigrees associated with this expansion exhibit phenotypic heterogeneity and incomplete disease penetrance, the basis of which is poorly understood. Relatives of those carrying the C9orf72 repeat expansion exhibit a characteristic cognitive endophenotype independent of carrier status. To examine whether additional shared genetic or environmental risks within kindreds could compel this observation, we have conducted a detailed cross-sectional study of the inheritance within multigenerational Irish kindreds carrying the C9orf72 repeat expansion. Methods: One hundred thirty-one familial ALS pedigrees, 59 of which carried the C9orf72 repeat expansion (45.0% [95% CI 36.7-53.5]), were identified through the Irish population-based ALS register. C9orf72 genotyping was performed using repeat-primed PCR with amplicon fragment length analysis. Pedigrees were further investigated using SNP, targeted sequencing data, whole-exome sequencing, and whole-genome sequencing. Results: We identified 21 kindreds where at least 1 family member with ALS carried the C9orf72 repeat expansion and from whom DNA was available from multiple affected family members. Of these, 6 kindreds (28.6% [95% CI 11.8-48.3]) exhibited discordant segregation. The C9orf72 haplotype was studied in 2 families and was found to segregate with the C9orf72-positive affected relative but not the C9orf72-negative affected relative. No other ALS pathogenic variants were identified within these discordant kindreds. Discussion: Family members of kindreds associated with the C9orf72 repeat expansion may carry an increased risk of developing ALS independent of their observed carrier status. This has implications for assessment and counseling of asymptomatic individuals regarding their genetic risk.

Molecular dynamics analysis of superoxide dismutase 1 mutations suggests decoupling between mechanisms underlying ALS onset and progression.

Mutations in the superoxide dismutase 1 (SOD1) gene are the second most common known cause of ALS. SOD1 variants express high phenotypic variability and over 200 have been reported in people with ALS. It was previously proposed that variants can be broadly classified in two groups, 'wild-type like' (WTL) and 'metal binding region' (MBR) variants, based on their structural location and biophysical properties. MBR variants, but not WTL variants, were associated with a reduction of SOD1 enzymatic activity. In this study we used molecular dynamics and large clinical datasets to characterise the differences in the structural and dynamic behaviour of WTL and MBR variants with respect to the wild-type SOD1, and how such differences influence the ALS clinical phenotype. Our study identified marked structural differences, some of which are observed in both variant groups, while others are group specific. Moreover, collecting clinical data of approximately 500 SOD1 ALS patients carrying variants, we showed that the survival time of patients carrying an MBR variant is generally longer (∼6 years median difference, p < 0.001) with respect to patients with a WTL variant. In conclusion, our study highlighted key differences in the dynamic behaviour between WTL and MBR SOD1 variants, and between variants and wild-type SOD1 at an atomic and molecular level, that could be further investigated to explain the associated phenotypic variability. Our results support the hypothesis of a decoupling between mechanisms of onset and progression of SOD1 ALS, and an involvement of loss-of-function of SOD1 with the disease progression.

VariantSurvival: a tool to identify genotype-treatment response.

Motivation: For a number of neurological diseases, such as Alzheimer's disease, amyotrophic lateral sclerosis, and many others, certain genes are known to be involved in the disease mechanism. A common question is whether a structural variant in any such gene may be related to drug response in clinical trials and how this relationship can contribute to the lifecycle of drug development. Results: To this end, we introduce VariantSurvival, a tool that identifies changes in survival relative to structural variants within target genes. VariantSurvival matches annotated structural variants with genes that are clinically relevant to neurological diseases. A Cox regression model determines the change in survival between the placebo and clinical trial groups with respect to the number of structural variants in the drug target genes. We demonstrate the functionality of our approach with the exemplary case of the SETX gene. VariantSurvival has a user-friendly and lightweight graphical user interface built on the shiny web application package.

SOD1-ALS-Browser: a web-utility for investigating the clinical phenotype in SOD1 amyotrophic lateral sclerosis.

Objective: Variants in the superoxide dismutase (SOD1) gene are among the most common genetic causes of amyotrophic lateral sclerosis. Reflecting the wide spectrum of putatively deleterious variants that have been reported to date, it has become clear that SOD1-linked ALS presents a highly variable age at symptom onset and disease duration.Methods: Here we describe an open access web tool for comparative phenotype analysis in ALS: https://sod1-als-browser.rosalind.kcl.ac.uk/. The tool contains a built-in dataset of clinical information from 1383 people with ALS harboring a SOD1 variant resulting in one of 162 unique amino acid sequence alterations and from a non-SOD1 comparator ALS cohort of 13,469 individuals. We present two examples of analyses possible with this tool, testing how the ALS phenotype relates to SOD1 variants that alter amino acid residue hydrophobicity and to distinct variants at the 94th residue of SOD1, where six are sampled.Results and conclusions: The tool provides immediate access to the datasets and enables bespoke analysis of phenotypic trends associated with different protein variants, including the option for users to upload their own datasets for integration with the server data. The tool can be used to study SOD1-ALS and provides an analytical framework to study the differences between other user-uploaded ALS groups and our large reference database of SOD1 and non-SOD1 ALS. The tool is designed to be useful for clinicians and researchers, including those without programming expertise, and is highly flexible in the analyses that can be conducted.

Epidemiological and clinical profile of amyotrophic lateral sclerosis in Ethiopia: a 5-year multicenter retrospective study.

Background: Amyotrophic lateral sclerosis (ALS) is a fatal neurodegenerative disease that results in progressive weakness of skeletal muscles including respiratory muscles. Epidemiological and clinical aspects of ALS are derived from a few world regions with very little representation of low- and middle-income countries. We therefore set out to determine the epidemiological and clinical phenotype of individuals with ALS in Ethiopia. Methods: Multicenter retrospective analysis was conducted using clinical records from ALS patients seen in Ethiopia at Tikur Anbessa Specialized Hospital and Yehuleshet specialty clinic between January 2016 and August 2021. The data collected included clinical characteristics, disease-related symptoms, a revised ALS functional rating scale, and medications. Results: Patients in Ethiopia had a younger age of onset with a mean age of disease onset of 51.9 years. 2.9% of patients had juvenile ALS, and the male-to-female ratio was almost 2:1. 4.9% had a positive family history of the disease. 68% of patients had spinal region involvement at onset, while 32% had bulbar region involvement at onset. Riluzole was used by 31% of ALS patients. 20.6% of patients had some respiratory symptoms, but none received a standard respiratory function assessment. 33.3% of patients were wheelchair-bound. Conclusion: In this retrospective study spanning 5 years, we examined the clinical phenotype of ALS in Ethiopian patients. Our findings suggest that most patients had clinically definite ALS with spinal region involvement. Further research, including genetic and epigenetic information, is necessary to understand the early onset of the disease in Ethiopia.

Non-motor symptoms in amyotrophic lateral sclerosis: lessons from Parkinson's disease.

Amyotrophic lateral sclerosis and Parkinson's disease are neurodegenerative diseases of the motor system which are now recognized also to affect non-motor pathways. Non-motor symptoms have been acknowledged as important determinants of quality of life in Parkinson's disease, and there is increasing interest in understanding the extent and role of non-motor symptoms in amyotrophic lateral sclerosis. We therefore reviewed what is known about non-motor symptoms in amyotrophic lateral sclerosis, using lessons from Parkinson's disease.

DNAscan2: a versatile, scalable, and user-friendly analysis pipeline for human next-generation sequencing data.

SUMMARY: The current widespread adoption of next-generation sequencing (NGS) in all branches of basic research and clinical genetics fields means that users with highly variable informatics skills, computing facilities and application purposes need to process, analyse, and interpret NGS data. In this landscape, versatility, scalability, and user-friendliness are key characteristics for an NGS analysis software. We developed DNAscan2, a highly flexible, end-to-end pipeline for the analysis of NGS data, which (i) can be used for the detection of multiple variant types, including SNVs, small indels, transposable elements, short tandem repeats, and other large structural variants; (ii) covers all standard steps of NGS analysis, from quality control of raw data and genome alignment to variant calling, annotation, and generation of reports for the interpretation and prioritization of results; (iii) is highly adaptable as it can be deployed and run via either a graphic user interface for non-bioinformaticians and a command line tool for personal computer usage; (iv) is scalable as it can be executed in parallel as a Snakemake workflow, and; (v) is computationally efficient by minimizing RAM and CPU time requirements. AVAILABILITY AND IMPLEMENTATION: DNAscan2 is implemented in Python3 and is available at https://github.com/KHP-Informatics/DNAscanv2.

Artificial Intelligence for Dementia Research Methods Optimization.

INTRODUCTION: Machine learning (ML) has been extremely successful in identifying key features from high-dimensional datasets and executing complicated tasks with human expert levels of accuracy or greater. METHODS: We summarize and critically evaluate current applications of ML in dementia research and highlight directions for future research. RESULTS: We present an overview of ML algorithms most frequently used in dementia research and highlight future opportunities for the use of ML in clinical practice, experimental medicine, and clinical trials. We discuss issues of reproducibility, replicability and interpretability and how these impact the clinical applicability of dementia research. Finally, we give examples of how state-of-the-art methods, such as transfer learning, multi-task learning, and reinforcement learning, may be applied to overcome these issues and aid the translation of research to clinical practice in the future. DISCUSSION: ML-based models hold great promise to advance our understanding of the underlying causes and pathological mechanisms of dementia.

Large-scale analyses of CAV1 and CAV2 suggest their expression is higher in post-mortem ALS brain tissue and affects survival.

Introduction: Caveolin-1 and Caveolin-2 (CAV1 and CAV2) are proteins associated with intercellular neurotrophic signalling. There is converging evidence that CAV1 and CAV2 (CAV1/2) genes have a role in amyotrophic lateral sclerosis (ALS). Disease-associated variants have been identified within CAV1/2 enhancers, which reduce gene expression and lead to disruption of membrane lipid rafts. Methods: Using large ALS whole-genome sequencing and post-mortem RNA sequencing datasets (5,987 and 365 tissue samples, respectively), and iPSC-derived motor neurons from 55 individuals, we investigated the role of CAV1/2 expression and enhancer variants in the ALS phenotype. Results: We report a differential expression analysis between ALS cases and controls for CAV1 and CAV2 genes across various post-mortem brain tissues and three independent datasets. CAV1 and CAV2 expression was consistently higher in ALS patients compared to controls, with significant results across the primary motor cortex, lateral motor cortex, and cerebellum. We also identify increased survival among carriers of CAV1/2 enhancer mutations compared to non-carriers within Project MinE and slower progression as measured by the ALSFRS. Carriers showed a median increase in survival of 345 days. Discussion: These results add to an increasing body of evidence linking CAV1 and CAV2 genes to ALS. We propose that carriers of CAV1/2 enhancer mutations may be conceptualised as an ALS subtype who present a less severe ALS phenotype with a longer survival duration and slower progression. Upregulation of CAV1/2 genes in ALS cases may indicate a causal pathway or a compensatory mechanism. Given prior research supporting the beneficial role of CAV1/2 expression in ALS patients, we consider a compensatory mechanism to better fit the available evidence, although further investigation into the biological pathways associated with CAV1/2 is needed to support this conclusion.

Harnessing Transcriptomic Signals for Amyotrophic Lateral Sclerosis to Identify Novel Drugs and Enhance Risk Prediction.

Introduction: Amyotrophic lateral sclerosis (ALS) is a fatal neurodegenerative disease. This study integrates the latest ALS genome-wide association study (GWAS) summary statistics with functional genomic annotations with the aim of providing mechanistic insights into ALS risk loci, inferring drug repurposing opportunities, and enhancing prediction of ALS risk and clinical characteristics. Methods: Genes associated with ALS were identified using GWAS summary statistic methodology including SuSiE SNP-based fine-mapping, and transcriptome- and proteome-wide association study (TWAS/PWAS) analyses. Using several approaches, gene associations were integrated with the DrugTargetor drug-gene interaction database to identify drugs that could be repurposed for the treatment of ALS. Furthermore, ALS gene associations from TWAS were combined with observed blood expression in two external ALS case-control datasets to calculate polytranscriptomic scores and evaluate their utility for prediction of ALS risk and clinical characteristics, including site of onset, age at onset, and survival. Results: SNP-based fine-mapping, TWAS and PWAS identified 117 genes associated with ALS, with TWAS and PWAS providing novel mechanistic insights. Drug repurposing analyses identified five drugs significantly enriched for interactions with ALS associated genes, with directional analyses highlighting α-glucosidase inhibitors may exacerbate ALS pathology. Additionally, drug class enrichment analysis showed calcium channel blockers may reduce ALS risk. Across the two observed expression target samples, ALS polytranscriptomic scores significantly predicted ALS risk (R2 = 4%; p-value = 2.1×10-21). Conclusions: Functionally-informed analyses of ALS GWAS summary statistics identified novel mechanistic insights into ALS aetiology, highlighted several therapeutic research avenues, and enabled statistically significant prediction of ALS risk.

Harnessing the potential of machine learning and artificial intelligence for dementia research.

Progress in dementia research has been limited, with substantial gaps in our knowledge of targets for prevention, mechanisms for disease progression, and disease-modifying treatments. The growing availability of multimodal data sets opens possibilities for the application of machine learning and artificial intelligence (AI) to help answer key questions in the field. We provide an overview of the state of the science, highlighting current challenges and opportunities for utilisation of AI approaches to move the field forward in the areas of genetics, experimental medicine, drug discovery and trials optimisation, imaging, and prevention. Machine learning methods can enhance results of genetic studies, help determine biological effects and facilitate the identification of drug targets based on genetic and transcriptomic information. The use of unsupervised learning for understanding disease mechanisms for drug discovery is promising, while analysis of multimodal data sets to characterise and quantify disease severity and subtype are also beginning to contribute to optimisation of clinical trial recruitment. Data-driven experimental medicine is needed to analyse data across modalities and develop novel algorithms to translate insights from animal models to human disease biology. AI methods in neuroimaging outperform traditional approaches for diagnostic classification, and although challenges around validation and translation remain, there is optimism for their meaningful integration to clinical practice in the near future. AI-based models can also clarify our understanding of the causality and commonality of dementia risk factors, informing and improving risk prediction models along with the development of preventative interventions. The complexity and heterogeneity of dementia requires an alternative approach beyond traditional design and analytical approaches. Although not yet widely used in dementia research, machine learning and AI have the potential to unlock current challenges and advance precision dementia medicine.