Biology subjects

Arnol, D.

Publications and source records attributed to Arnol, D..

5 recordsLinked to original sources

bioRxiv2020

Identifying temporal and spatial patterns of variation from multi-modal data using MEFISTO

Factor analysis is among the most-widely used methods for dimensionality reduction in genome biology, with applications from personalized health to single-cell studies. Existing implementations of factor analysis assume independence of the observed samples, an assumption that fails in emerging spatio-temporal profiling studies. Here, we present MEFISTO, a flexible and versatile toolbox for modelling high-dimensional data when spatial or temporal dependencies between the samples are known. MEFISTO maintains the established benefits of factor analysis for multi-modal data, but enables performing spatio-temporally informed dimensionality reduction, interpolation and separation of smooth from non-smooth patterns of variation. Moreover, MEFISTO can integrate multiple related datasets by simultaneously identifying and aligning the underlying patterns of variation in a data-driven manner. We demonstrate MEFISTO through applications to an evolutionary atlas of mammalian organ development, where the model reveals conserved and evolutionary diverged developmental programs. In applications to a longitudinal microbiome study in infants, birth mode and diet were highlighted as major causes for heterogeneity in the temporally-resolved microbiome over the first years of life. Finally, we demonstrate that the proposed framework can also be applied to spatially resolved transcriptomics.

bioinformatics↗

bioRxiv2019

MOFA+: a probabilistic framework for comprehensive integration of structured single-cell data

Technological advances have enabled the joint analysis of multiple molecular layers at single cell resolution. At the same time, increased experimental throughput has facilitated the study of larger numbers of experimental conditions. While methods for analysing single-cell data that model the resulting structure of either of these dimensions are beginning to emerge, current methods do not account for complex experimental designs that include both multiple views (modalities or assays) and groups (conditions or experiments). Here we present Multi-Omics Factor Analysis v2 (MOFA+), a statistical framework for the comprehensive and scalable integration of structured single cell multi-modal data. MOFA+ builds upon a Bayesian Factor Analysis framework combined with fast GPU-accelerated stochastic variational inference. Similar to existing factor models, MOFA+ allows for interpreting variation in single-cell datasets by pooling information across cells and features to reconstruct a low-dimensional representation of the data. Uniquely, the model supports flexible group-level sparsity constraints that allow joint modelling of variation across multiple groups and views.\n\nTo illustrate MOFA+, we applied it to single-cell data sets of different scales and designs, demonstrating practical advantages when analyzing datasets with complex group and/or view structure. In a multi-omics analysis of mouse gastrulation this joint modelling reveals coordinated changes between gene expression and epigenetic variation associated with cell fate commitment.

bioinformatics↗

bioRxiv2018

Tandem duplications lead to loss of fitness effects in CRISPR-Cas9 data

CRISPR-Cas9 gene-editing is widely used to study gene function and is being advanced for therapeutic applications. Structural rearrangements are a ubiquitous feature of cancers and their impact on CRISPR-Cas9 gene-editing has not yet been systematically assessed. Utilising CRISPR-Cas9 knockout screens for 163 cancer cell lines, we demonstrate that targeting tandem amplified regions is highly detrimental to cellular fitness, in contrast to amplifications caused by chromosomal duplications which have little to no effect. Genomically clustered Cas9 double-strand DNA breaks are associated with a strong gene-independent decrease in cell fitness. We systematically identified collateral vulnerabilities in 25% of cancer cells, introduced by tandem amplifications of tissue non-expressed genes. Our analysis demonstrates the importance of structural rearrangements in mediating the effect of CRISPR-Cas9-induced DNA damage, with implications for the use of CRISPR-Cas9 gene-editing technology, and how resulting collateral vulnerabilities are a generalisable strategy to target cancer cells.

genomics↗

bioRxiv2018

Modelling cell-cell interactions from spatial molecular data with spatial variance component analysis

Technological advances allow for assaying multiplexed spatially resolved RNA and protein expression profiling of individual cells, thereby capturing physiological tissue contexts of single cell variation. While methods for the high-throughput generation of spatial expression profiles are increasingly accessible, computational methods for studying the relevance of the spatial organization of tissues on cell-cell heterogeneity are only beginning to emerge. Here, we present spatial variance component analysis (SVCA), a computational framework for the analysis of spatial molecular data. SVCA enables quantifying the effect of cell-cell interactions, as well as environmental and intrinsic cell features on the expression levels of individual genes or proteins. In application to a breast cancer Imaging Mass Cytometry dataset, our model allows for robustly estimating spatial variance signatures, identifying cell-cell interactions as a major driver of expression heterogeneity. Finally, we apply SVCA to high-dimensional imaging-derived RNA data, where we identify molecular pathways that are linked to cell-cell interactions.

bioinformatics↗

bioRxiv2017

Multi-Omics factor analysis disentangles heterogeneity in blood cancer

Multi-omic studies promise the improved characterization of biological processes across molecular layers. However, methods for the unsupervised integration of the resulting heterogeneous datasets are lacking. We present Multi-Omics Factor Analysis (MOFA), a computational method for discovering the principal sources of variation in multi-omic datasets. MOFA infers a set of (hidden) factors that capture biological and technical sources of variability. It disentangles axes of heterogeneity that are shared across multiple modalities and those specific to individual data modalities. The learnt factors enable a variety of downstream analyses, including identification of sample subgroups, data imputation, and the detection of outlier samples. We applied MOFA to a cohort of 200 patient samples of chronic lymphocytic leukaemia, profiled for somatic mutations, RNA expression, DNA methylation and ex-vivo drug responses. MOFA identified major dimensions of disease heterogeneity, including immunoglobulin heavy chain variable region status, trisomy of chromosome 12 and previously underappreciated drivers, such as response to oxidative stress. In a second application, we used MOFA to analyse single-cell multiomics data, identifying coordinated transcriptional and epigenetic changes along cell differentiation.

bioinformatics↗