Biology subjects

Streck, A.

Publications and source records attributed to Streck, A..

5 recordsLinked to original sources

bioRxiv2026

Haplotype-resolved genome architecture mapping uncovers pervasive structural heterogeneity between human homologous chromosomes

Resolving the three-dimensional structure of chromatin with haplotype specificity remains a fundamental challenge for understanding genetic and epigenetic contributions to gene regulation in healthy development and in disease. Phasing of chromatin contacts derived by ligation-dependent methods relies on high single-nucleotide variant (SNV) density, which limits its applicability in human genomes where SNVs are sparse. Here, we present CoPhasing, a strategy that leverages the intrinsic haplotype fidelity of Genome Architecture Mapping (GAM), a ligation-free method that inherently captures haplotype-matched genomic neighborhoods in thin nuclear slices. CoPhasing assigns SNV-free reads to their haplotype through proximity to informative reads, enabling efficient and accurate phasing even in SNV-sparse regions. Applying CoPhasing genome-wide reveals extensive, previously inaccessible structural heterogeneity between human homologous chromosomes, demonstrating the power of GAM CoPhasing to investigate allele-specific differences in 3D genome organization and their relevance to gene regulation and disease.

genomics↗

bioRxiv2026

Deciphering selection patterns of somatic copy-number events

Somatic copy-number alterations (SCNAs) are pervasive across cancers, arising through diverse mutational mechanisms and subsequently shaped by selection according to fitness advantages [1-4]. These alterations can drive tumourigenesis via gene-dosage changes in oncogenes and tumour suppressors [5, 6]. However, observed SCNA profiles do not uniquely define the genomic events that led to them, which is a major obstacle to understanding the selective forces shaping cancer genomes. Here we present SPICE, Selection Patterns In somatic Copy-number Events, an event-level framework that infers discrete copy-number events from allele-specific profiles and models focal selection from first principles, incorporating uniform breakpoint formation and locus-specific selective pressures. Applied to 5,966 samples from the TCGA dataset, SPICE quantifies the full spectrum of SCNA events, recapitulates known mutational processes, and tracks systematic shifts in event types before and after whole-genome duplication (WGD). Next, SPICE employs a generative selection model to identify the location of oncogenes and tumour-suppressors. Unlike previous approaches that detected peaks in aggregated copy-number signals, our model operates directly on inferred events and treats genome-wide breakpoint formation as a neutral reference against which locus-specific selection is detected. This analysis reveals 460 loci under selection which are highly enriched for known oncogenes and tumour suppressors, recapitulating most previously reported sites and uncovering many novel regions, and simultaneously showing that most internal copy-number events are not subject to focal selection. These results establish a unified framework that deconvolves copy-number profiles into their underlying evolutionary events and greatly expand the catalogue of loci implicated in cancer development. Graphical abstract O_FIG O_LINKSMALLFIG WIDTH=200 HEIGHT=45 SRC="FIGDIR/small/708809v1_ufig1.gif" ALT="Figure 1"> View larger version (17K): org.highwire.dtl.DTLVardef@7d471corg.highwire.dtl.DTLVardef@c78feeorg.highwire.dtl.DTLVardef@1ba52a8org.highwire.dtl.DTLVardef@188ab0c_HPS_FORMAT_FIGEXP M_FIG C_FIG

cancer biology↗

bioRxiv2025

Path integration impairments reveal early cognitive changes in Subjective Cognitive Decline

Path integration, the ability to track ones position using self-motion cues, is critically dependent on the grid cell network in the entorhinal cortex, a region vulnerable to early Alzheimers disease pathology. In this study, we examined path integration performance in individuals with subjective cognitive decline (SCD), a group at increased risk for Alzheimers disease, and healthy controls using an immersive virtual reality task. We developed a Bayesian computational model to decompose path integration errors into distinct components. SCD participants exhibited significantly higher path integration error, primarily driven by increased memory leak, while other modelling-derived error sources, such as velocity gain, sensory and reporting noise, remained comparable across groups. Our findings suggest that path integration deficits, specifically memory leak, may serve as an early marker of neurodegeneration in SCD and highlight the potential of self-motion-based navigation tasks for detecting pre-symptomatic Alzheimers disease-related cognitive changes. TeaserVirtual reality, modelling, and plasma biomarkers reveal path integration deficits in pre-symptomatic Alzheimers vs normal aging.

neuroscience↗

bioRxiv2024

CNSistent integration and feature extraction from somatic copy number profiles

The vast majority of cancers exhibit Somatic Copy Number Alterations (SCNAs)--gains and losses of variable regions of DNA. SCNAs can shape the phenotype of cancer cells, e.g. by increasing their proliferation rates, removing tumor suppressor genes, or immortalizing cells. While many SCNAs are unique to a patient, certain recurring patterns emerge as a result of shared selectional constraints or common mutational processes. To discover such patterns in a robust way, the size of the dataset is essential, which necessitates combining SCNA profiles from different cohorts, a non-trivial task. To achieve this, we developed CNSistent, a Python package for imputation, filtering, consistent segmentation, feature extraction, and visualization of cancer copy number profiles from heterogeneous datasets. We demonstrate the utility of CNSistent by applying it to the publicly available TCGA, PCAWG, and TRACERx cohorts. We compare different segmentation and aggregation strategies on cancer type and subtype classification tasks using deep convolutional neural networks. We demonstrate an increase in accuracy over training on individual cohorts and efficient transfer learning between cohorts. Using integrated gradients we investigate lung cancer classification results, highlighting SOX2 amplifications as the dominant copy number alteration in lung squamous cell carcinoma.

genomics↗

bioRxiv2022

SMITH: Spatially Constrained Stochastic Model for Simulation of Intra-Tumour Heterogeneity

MotivationSimulations of cancer evolution and cellular growth have proven highly useful to study, in detail, the various aspects of intra-tumour heterogeneity, including the effect of selection, mutation rates, and spatial constraints. However, most methods are computationally expensive lattice-embedded models which cannot simulate tumours with a realistic number of cells and rely on various simplifications. Alternatively, well-mixed stochastic models, while efficient and scalable, do not typically include spatial constraints and cannot reproduce the rich clonal dynamics observed in real-world tumours. ResultsWe present SMITH, a simple, efficient, and explainable model of cancer evolution that combines the advantages of well-mixed stochastic models with a new confinement mechanism which limits the growth of clones based on the overall tumour size. We demonstrate that this confinement mechanism is sufficient to induce the rich clonal dynamics observed in spatial models, while allowing for a clear geometric interpretation and efficient simulation of one billion cells within a few minutes on a desktop PC. We explore the extent of stochasticity and rigorously assess the effects of cell turnover, mutation rate, fitness effects and confinement on the resulting clonal structures. Availability and ImplementationSMITH is implemented in C# and freely available at bitbucket.org/schwarzlab/smith together with binaries for all major platforms. For rich visualisations of the simulated clonal dynamics we provide an accompanying Python package PyFish at bitbucket.org/schwarzlab/pyfish. Supplementary informationAll supplementary figures are in the supplementary document.

bioinformatics↗