Biology subjects

Alasoo, K.

Publications and source records attributed to Alasoo, K..

18 recordsLinked to original sources

bioRxiv2026

The Human Pleiotropic Map of GWAS Associations and Therapeutic Implications

Genetic support for drug targets substantially increases clinical success rates, establishing genome-wide association studies (GWAS) as central to therapeutic hypothesis generation. However, the same genetic evidence that reveals causal gene-disease relationships simultaneously exposes organism-level safety liabilities--a dimension requiring principled, genome-wide quantification. Here we systematically analyse 100,526 GWAS to yield 789,453 credible sets and gene prioritisations for 15,641 genes, with discovery showing no saturation as GWAS expand and increase diversity. We find that 64% of GWAS-implicated genes are pleiotropic, associated with traits across multiple diseases and showing a non-linear relationship between the degree of pleiotropy and clinical success. Highly pleiotropic genes--concentrated in immune, inflammatory, and oncogenic signalling programmes--are enriched in safety-terminated clinical programmes, mouse lethal knockouts, and cancer driver genes, establishing gene-level pleiotropy as a potential measure of genetically-informed organism-level safety liability. Protein-altering variant (PAV) support amplifies therapeutic signal (OR = 6.0), yet PAV targets show higher average pleiotropy, introducing a competing safety liability. Combining PAV support with intermediate pleiotropy (2-5 therapeutic areas) resolves this tension, yielding OR = 10.3 and relative success = 4.8--a profile already satisfied by 52 approved therapies. As GWAS continue to expand in scale and resolution, these findings lay the groundwork for increasingly sophisticated target discovery strategies that yield safer and more effective therapeutic hypotheses.

genomics↗

bioRxiv2025

Ultra-fast genetic colocalisation across millions of traits

Colocalisation is a powerful approach to assess if two genetic association signals are likely to share a causal variant. However, association analyses in large biobanks and molecular quantitative trait loci (molQTL) studies now routinely identify millions of association signals across thousands of traits, making it infeasible to test for colocalisation between all pairs of signals. Here we introduce gpu-coloc, a GPU-accelerated re-implementation of the coloc algorithm that combines efficient data storage with parallelisation to achieve a 1000-fold speed increase while maintaining near-identical results. As a result, the run time of gpu-coloc now approaches the colocalisation posterior probability (CLPP) method, a competing method that only uses information from fine mapped credible sets to detect colocalisations. Using summary statistics from UK Biobank, FinnGen, and eQTL Catalogue, we demonstrate that gpu-coloc and CLPP detect highly concordant results, especially when restricting the analysis to confidently fine mapped signals. We introduce the colocalisation collider metric to quantify spurious colocalisations in large-scale colocalisation graphs and use it to choose decision thresholds that provide a reasonable trade-off between sensitivity and specificity. Finally, we demonstrate how gpu-coloc can also be applied to marginal GWAS summary statistics from studies that lack fine mapping, where it is still able to recover molQTL colocalisations for [~]80% of the GWAS loci. Our efficient software and comprehensive analyses provide practical guidelines for future large-scale colocalisation analyses.

genomics↗

bioRxiv2025

Integrated QTL mapping and CRISPR screening in pooled iPSC-derived microglia reveals genetic drivers of neurodegenerative risk

Mounting evidence implicates microglia in neurodegeneration, but linking disease-associated genetic variants to target genes and cellular phenotypes is hindered by the inaccessibility of these cells. We differentiated 261 human iPSC lines into microglia-like cells (iMGL) in pools with phenotypic (differentiation, phagocytosis and migration) and single-cell transcriptomic readouts. Burden analysis of deleterious variants detected 36 genes influencing microglial phenotypes. Expression quantitative trait locus (eQTL) analysis found 7,121 eGenes, and 79 colocalizations across four neurodegenerative disease GWAS, half of which had limited prior evidence of causality. Integration of eQTL and phenotypic associations highlighted the role of disease-relevant variants including LRRK2 and TREM2 acting via microglial phagocytosis. A coupled CRISPR screen identified a role of TREM2 in phagocytosis and highlighted the importance of cellular state in directionality of phenotype. By contextualizing variant effects within disease-relevant microglial states, we provide a comprehensive framework for interpreting the function of risk loci in neurodegenerative disorders.

genomics↗

bioRxiv2025

GAME: Genomic API for Model Evaluation

The rapid expansion of genomics datasets and the application of machine learning has produced sequence-to-activity genomics models with ever-expanding capabilities. However, benchmarking these models on practical applications has been challenging because individual projects evaluate their models in ad hoc ways, and there is substantial heterogeneity of both model architectures and benchmarking tasks. To address this challenge, we have created GAME, a system for large-scale, community-led standardized model benchmarking on user-defined evaluation tasks. We borrow concepts from the Application Programming Interface (API) paradigm to allow for seamless communication between pre-trained models and benchmarking tasks, ensuring consistent evaluation protocols. Because all models and benchmarks are inherently compatible in this framework, the continual addition of new models and new benchmarks is easy. We also developed a Matcher module powered by a large language model (LLM) to automate ambiguous task alignment between benchmarks and models. Containerization of these modules enhances reproducibility and facilitates the deployment of models and benchmarks across computing platforms. By focusing on predicting underlying biochemical phenomena (e.g. gene expression, open chromatin, DNA binding), we ensure that tasks remain technology-independent. We provide examples of benchmarks and models implementing this framework, and anticipate that the community will contribute their own, leading to an ever-expanding and evolving set of models and evaluation tasks. This resource will accelerate genomics research by illuminating the best models for a given task, motivating novel functional genomic benchmarks, and providing a more nuanced understanding of model abilities.

bioinformatics↗

bioRxiv2024

ChromBPNet: bias factorized, base-resolution deep learning models of chromatin accessibility reveal cis-regulatory sequence syntax, transcription factor footprints and regulatory variants

Despite extensive mapping of cis-regulatory elements (cREs) across cellular contexts with chromatin accessibility assays, the sequence syntax and genetic variants that regulate transcription factor (TF) binding and chromatin accessibility at context-specific cREs remain elusive. We introduce ChromBPNet, a deep learning DNA sequence model of base-resolution accessibility profiles that detects, learns and deconvolves assay-specific enzyme biases from regulatory sequence determinants of accessibility, enabling robust discovery of compact TF motif lexicons, cooperative motif syntax and precision footprints across assays and sequencing depths. Extensive benchmarks show that ChromBPNet, despite its lightweight design, is competitive with much larger contemporary models at predicting variant effects on chromatin accessibility, pioneer TF binding and reporter activity across assays, cell contexts and ancestry, while providing interpretation of disrupted regulatory syntax. ChromBPNet also helps prioritize and interpret regulatory variants that influence complex traits and rare diseases, thereby providing a powerful lens to decode regulatory DNA and genetic variation.

genetics↗

bioRxiv2024

Predicted mechanistic impacts of human protein missense variants

Genome sequencing efforts have led to the discovery of tens of millions of protein missense variants found in the human population with the majority of these having no annotated role and some likely contributing to trait variation and disease. Sequence-based artificial intelligence approaches have become highly accurate at predicting variants that are detrimental to the function of proteins but they do not inform on mechanisms of disruption. Here we combined sequence and structure-based methods to perform proteome-wide prediction of deleterious variants with information on their impact on protein stability, protein-protein interactions and small-molecule binding pockets. AlphaFold2 structures were used to predict approximately 100,000 small-molecule binding pockets and stability changes for over 200 million variants. To inform on protein-protein interfaces we used AlphaFold2 to predict structures for nearly 500,000 protein complexes. We illustrate the value of mechanism-aware variant effect predictions to study the relation between protein stability and abundance and the structural properties of interfaces underlying trans protein quantitative trait loci (pQTLs). We characterised the distribution of mechanistic impacts of protein variants found in patients and experimentally studied example disease linked variants in FGFR1.

bioinformatics↗

bioRxiv2023

An Alzheimer's disease-associated common regulatory variant in PTK2B has causal effects on microglial function

Genome-wide association studies (GWAS) are revealing an ever-growing number of genetic associations with disease, but identifying and functionally validating the causal variants underlying these associations is very challenging and has only been done for a vanishingly small number of variants. Here we validate a single nucleotide polymorphism (SNP) associated with an increased risk of Alzheimers disease (AD) in an intronic enhancer of the PTK2B gene, by engineering it into human induced pluripotent stem cells (hiPSCs). Upon differentiation to macrophages and microglia, this variant shows effects on chromatin accessibility of the enhancer and increased binding of the transcription factor CEBPB but only subtle effects on PTK2B or CLU expression. Nevertheless, this variant results in global changes to the transcriptome and phenotype of these cells. Expression of interferon gamma responsive genes including chemokine transcripts and their protein products are altered, and chemotaxis of the resulting microglial cells is affected. This variant thus causes disease-relevant transcriptomic and phenotypic changes, and we propose that it acts by altering microglia reactivity, consistent with the role of these cells in progression of AD.

neuroscience↗

bioRxiv2023

Extensive co-regulation of neighbouring genes complicates the use of eQTLs in target gene prioritisation

Identifying causal genes underlying genome-wide association studies (GWAS) is a fundamental problem in human genetics. Although colocalisation with gene expression quantitative trait loci (eQTLs) is often used to prioritise GWAS target genes, systematic benchmarking has been limited due to unavailability of large ground truth datasets. Here, we re-analysed plasma protein QTL data from 3,301 individuals of the INTERVAL cohort together with 131 eQTL Catalogue datasets. Focusing on variants located within or close to the affected protein identified 793 proteins with at least one cis-pQTL where we could assume that the most likely causal gene was the gene coding for the protein. We then benchmarked the ability of cis-eQTLs to recover these causal genes by comparing three Bayesian colocalisation methods (coloc.susie, coloc.abf and CLPP) and five Mendelian randomisation (MR) approaches (three varieties of inverse-variance weighted MR, MR-RAPS, and MRLocus). We found that assigning fine-mapped pQTLs to their closest protein coding genes outperformed all colocalisation methods regarding both precision (71.9%) and recall (76.9%). Furthermore, the colocalisation method with the highest recall (coloc.susie - 46.3%) also had the lowest precision (45.1%). Combining evidence from multiple conditionally distinct colocalising QTLs with MR increased precision to 81%, but this was accompanied by a large reduction in recall to 7.1%. Furthermore, the choice of the MR method greatly affected performance, with the standard inverse-variance weighted MR often producing many false positives. Our results highlight that linking GWAS variants to target genes remains challenging with eQTL evidence alone, and prioritising novel targets requires triangulation of evidence from multiple sources.

genetics↗

bioRxiv2023

Genetic determinants of plasma protein levels in the Estonian population

The proteome holds great potential as an intermediate layer between the genome and phenome. Previous protein quantitative trait locus studies have focused mainly on describing the effects of common genetic variations on the proteome. Here, we assessed the impact of the common and rare genetic variations as well as the copy number variants (CNVs) on 326 plasma proteins measured in up to 500 individuals. We identified 184 cis and 94 trans signals for 157 protein traits, which were further fine-mapped to credible sets for 101 cis and 87 trans signals for 151 proteins. Rare genetic variation contributed to the levels of 7 proteins, with 5 cis and 14 trans associations. CNVs were associated with the levels of 11 proteins (7 cis and 5 trans), examples including a 3q12.1 deletion acting as a hub for multiple trans associations; and a CNV overlapping NAIP, a sensor component of the NAIP-NLRC4 inflammasome which is affecting pro-inflammatory cytokine interleukin 18 levels. In summary, this work presents a comprehensive resource of genetic variation affecting the plasma protein levels and provides the interpretation of identified effects.

genomics↗

bioRxiv2023

Systematic visualisation of molecular QTLs reveals variant mechanisms at GWAS loci

Splicing quantitative trait loci (QTLs) have been implicated as a common mechanism underlying complex trait associations. However, utilising splicing QTLs in target discovery and prioritisation has been challenging due to extensive data normalisation which often renders the direction of the genetic effect as well as its magnitude difficult to interpret. This is further complicated by the fact that strong expression QTLs often manifest as weak splicing QTLs and vice versa, making it difficult to uniquely identify the underlying molecular mechanism at each locus. We find that these ambiguities can be mitigated by visualising the association between the genotype and average RNA sequencing read coverage in the region. Here, we generate these QTL coverage plots for 1.7 million molecular QTL associations in the eQTL Catalogue identified with five quantification methods. We illustrate the utility of these QTL coverage plots by performing colocalisation between vitamin D levels in the UK Biobank and all molecular QTLs in the eQTL Catalogue. We find that while visually confirmed splicing QTLs explain just 6/53 of the colocalising signals, they are significantly less pleiotropic than eQTLs and identify a prioritised causal gene in 4/6 cases. All our association summary statistics and QTL coverage plots are freely available at https://www.ebi.ac.uk/eqtl/.

genomics↗

bioRxiv2022

Optimised whole-genome CRISPR interference screens identify ARID1A-dependent growth regulators in human induced pluripotent stem cells

Perturbation of gene function is a powerful way to understand the role of individual genes in cellular systems. Whole-genome CRISPR/Cas-based screens have parallelized this approach and identified genes that modulate growth in many contexts. However, the DNA break-induced stress upon Cas9 action limits the efficacy of these screens in important models, such as human induced pluripotent stem cells (iPSCs). Silencing with a catalytically inactive Cas9 is a less stressful alternative, but has been considered less effective so far. Here, we first tested the efficiency of several dCas9 fusion proteins for target repression in human iPSCs, and identified dCas9-KRAB-MeCP2 as the most potent. We then produced monoclonal and polyclonal cell lines carrying this construct from multiple iPSC donors, and optimized genome-wide screens with them. We found silencing in a 200bp window around the transcription start site to be as effective as using wild-type Cas9 for identifying essential genes in iPSCs, but with a reduced cost due to better cell survival. Monoclonal lines performed better, but data from polyclonal lines were of sufficient quality for screening for larger effects. Finally, we performed whole-genome screens to identify dosage sensitivities that depend on the functionality of ARID1A, a commonly mutated cancer and developmental disorder gene. We observed differential growth upon depletion of NF2, TAF6L, EZH2, and PSMB2 genes in ARID1A+/-lines compared to wild type, and an enrichment of proteasome genes. Further, we confirmed that the context-specific growth decrease was phenocopied by treating the cells with a proteasome inhibitor, suggesting a pharmacologically targetable synthetic lethal interaction between the proteasome and ARID1A. We propose that many more plausible targets in challenging cell models can be efficiently identified with our approach.

cell biology↗

bioRxiv2022

Improved detection of genetic effects on promoter usage with augmented transcript annotations

Disease-associated non-coding variants can modulate their target genes by disrupting multiple mechanisms, including regulating total gene expression level, splicing, alternative polyadenylation or promoter usage. Quantifying promoter usage from standard RNA sequencing data is challenging due to incomplete reference transcriptome annotations and low read coverage observed at the ends of transcripts. We previously developed the txrevise tool (https://github.com/kauralasoo/txrevise) to quantify promoter usage events from RNA-seq data using reference transcriptome annotations. Here, we augment the txrevise promoter event annotations with experimentally identified Cap Analysis of Gene Expression (CAGE) promoters from the FANTOM5 project. Applying the new annotations to RNA-seq data from 358 individuals, we found that augmented promoter event annotations increased the power to detect promoter usage quantitative trait loci (puQTLs) by ~30%. However, concordance between puQTLs inferred from RNA-seq data and those directly measured using CAGE remained low, suggesting that additional experimental and computational improvements are needed to capture the full range of regulatory effects of non-coding variants.

bioinformatics↗

bioRxiv2020

Co-expression analysis reveals interpretable gene modules controlled by trans-acting genetic variants

BackgroundDeveloping novel therapies for complex disease requires better understanding of the causal processes that contribute to disease onset and progression. Although trans-acting gene expression quantitative trait loci (trans-eQTLs) can be a powerful approach to directly reveal cellular processes modulated by disease variants, detecting trans-eQTLs remains challenging due to their small effect sizes and large number of genes tested. However, if a single trans-eQTL controls a group of co-regulated genes, then multiple testing burden can be greatly reduced by summarising gene expression at the level of co-expression modules prior to trans-eQTL analysis. ResultsWe analysed gene expression and genotype data from six blood cell types from 226 to 710 individuals. We inferred gene co-expression modules with five methods on the full dataset, as well as in each cell type separately. We detected a number of established co-expression module trans-eQTLs, such as the monocyte-specific associations at the IFNB1 and LYZ loci, as well as a platelet-specific ARHGEF3 locus associated with mean platelet volume. We also discovered a novel trans association near the SLC39A8 gene in LPS-stimulated monocytes. Here, we linked an early-response cis-eQTL of the SLC39A8 gene to a module of co-expressed metallothionein genes upregulated more than 20 hours later and used motif analysis to identify zinc-induced activation of the MTF1 transcription factor as a likely mediator of this effect. ConclusionsOur analysis provides a rare detailed characterisation of a trans-eQTL effect cascade from a proximal cis effect to the affected signalling pathway, transcription factor, and target genes. This highlights how co-expression analysis combined with functional enrichment analysis can greatly improve the identification and prioritisation of trans-eQTLs when applied to emerging cell-type specific datasets.

genomics↗

bioRxiv2020

eQTL Catalogue: a compendium of uniformly processed human gene expression and splicing QTLs

An increasing number of gene expression quantitative trait locus (eQTL) studies have made summary statistics publicly available, which can be used to gain insight into complex human traits by downstream analyses, such as fine mapping and colocalisation. However, differences between these datasets, in their variants tested, allele codings, and in the transcriptional features quantified, are a barrier to their widespread use. Consequently, target genes for most GWAS signals have still not been identified. Here, we present the eQTL Catalogue (https://www.ebi.ac.uk/eqtl/), a resource which contains quality controlled, uniformly recomputed QTLs from 21 eQTL studies. We find that for matching cell types and tissues, the eQTL effect sizes are highly reproducible between studies, enabling the integrative analysis of these data. Although most cis-eQTLs were shared between most bulk tissues, the analysis of purified cell types identified a greater diversity of cell-type-specific eQTLs, a subset of which also manifested as novel disease colocalisations. Our summary statistics can be downloaded by FTP, accessed via a REST API, and visualised on the Ensembl genome browser. New datasets will continuously be added to the eQTL Catalogue, enabling the systematic interpretation of human GWAS associations across many cell types and tissues.

genomics↗

bioRxiv2019

Immune disease variants modulate gene expression in regulatory CD4+ T cells and inform drug targets

Identifying cellular functions dysregulated by disease associated variants could implicate novel pathways for drug targeting or modulation in cell therapies. Variants associated with immune diseases point towards the role of CD4+ regulatory T cells (Tregs), cell type critical for immune homeostasis. To translate the effects of immune disease alleles on Treg function we mapped genetic regulation (QTL) of gene expression and chromatin activity in Tregs. We identified 123 loci where Treg QTLs colocalized with immune disease variants. These colocalizations indicated that dysregulation of key Treg pathways, including cell activation via CD28 and IL-2 signalling mediated by STAT5A contribute to the shared pathobiology of immune diseases. Finally, using disease GWAS signals colocalizing with Treg QTLs, we identified 34 known drug targets and 270 targets with drug tractability evidence. Our study is the first in-depth characterization of immune disease variant effects on Treg gene expression modulation and dysregulation of Treg function.

genetics↗

bioRxiv2018

Genetic effects on promoter usage are highly context-specific and contribute to complex traits

Genetic variants regulating RNA splicing and transcript usage have been implicated in both common and rare diseases. Although transcript usage quantitative trait loci (tuQTLs) have now been mapped in multiple cell types and conditions, the molecular mechanisms through which these variants exert their effect have remained elusive. Specifically, changes in transcript usage could arise from promoter choice, alternative splicing or 3' end choice, but current tuQTL studies have not been able to distinguish between them. Here, we performed comprehensive analysis of RNA-seq data from human macrophages exposed to a range of inflammatory stimuli (IFN{gamma}, Salmonella, IFN{gamma} + Salmonella) and a metabolic stimulus (acetylated LDL), obtained from up to 84 individuals. In addition to conventional gene-level and transcript-level analyses, we also developed an analytical approach to directly quantify promoter, internal exon and 3' end usage. We found that although naive transcript-level analysis often links single genetic variants to multiple coupled changes on the transcriptome, this appears to be an artefact of incomplete transcript annotations. Most of this coupling disappears when promoters, splicing and 3' end usage are quantified directly. Furthermore, promoter, splicing and 3' end QTLs are each enriched in distinct genomic features, suggesting that they are predominantly controlled by independent regulatory mechanisms. We also find that promoter usage QTLs are 50% more likely to be context-specific than canonical splicing QTLs and constitute 25% of the transcript-level colocalisations with complex traits. Thus, promoter usage might be a previously underappreciated molecular mechanism mediating complex trait associations in a context-specific manner.

genomics↗

bioRxiv2017

Genetic effects on chromatin accessibility foreshadow gene expression changes in macrophage immune response

Noncoding regulatory variants play an important role in the genetics of complex traits. Although quantitative trait locus (QTL) mapping is a powerful approach to identify these variants, many genetic effects may remain unobserved when cells are sampled in only one of a large number of possible environments. Using a novel induced pluripotent stem cell-derived system, we mapped QTLs regulating chromatin accessibility and gene expression in macrophages in four conditions mimicking the interplay between interferon-gamma response and Salmonella infection. We found that approximately 50% of condition-specific effects on gene expression altered chromatin accessibility prior to stimulation. Furthermore, 6% of the chromatin accessibility QTLs regulated multiple neighbouring regions and these interactions were modulated by stimulation, occasionally producing condition-specific changes in gene expression. Profiling additional states also doubled the number of expression QTLs that could be confidently colocalised with disease associations. Thus, a substantial fraction of disease-associated variants may affect primed regulatory elements in naive cells.

genomics↗

bioRxiv2017

Molecular and functional variation in iPSC-derived sensory neurons

Induced pluripotent stem cells (iPSCs), and cells derived from them, have become key tools to model biological processes and disease mechanisms, particularly in cell types such as neurons that are difficult to access from living donors. Here, we present the first map of regulatory variants in an iPSC-derived cell type. To investigate genetic contributions to human sensory function, we performed 123 differentiations of iPSCs from 103 unique donors to a sensory neuronal fate, and measured gene expression, chromatin accessibility, and neuronal excitability. Compared with primary dorsal root ganglion, where sensory nerves collect near the spinal cord, gene expression was more variable across iPSC-derived neuronal cultures, particularly in genes related to differentiation and nervous system development. Single cell RNA-sequencing revealed that although the majority of cells are neuronal and express the expected marker genes, a substantial fraction have a fibroblast-like expression profile. By applying an allele-specific method we identify 3,778 quantitative trait loci influencing gene expression, 6,318 for chromatin accessibility, and 2,097 for RNA splicing at FDR 10%. A number of these overlap with common disease associations, and suggest candidate causal variants and target genes. These include known causal variants at SNCA for Parkinsons disease and TNFRSF1A for multiple sclerosis, as well as new candidates for migraine, Parkinsons disease, and schizophrenia.

genomics↗