Biology subjects

Raetsch, G.

Publications and source records attributed to Raetsch, G..

12 recordsLinked to original sources

bioRxiv2026

Slide-level batch structure limits histology-guided supervision of transcriptomic foundation models

Spatial transcriptomics pairs spatially resolved gene expression with tissue morphology in the same tissue section. Transcriptomic foundation models encode such expression profiles into general-purpose representations, but these representations can retain slide- and cohort-specific variation that obscures biological signal. Here, we test whether matched H\&E histology can improve these representations as a training-time supervisory signal that is discarded at inference. Across three transcriptomic foundation models, histology-guided supervision provides no consistent aggregate improvement in cross-donor annotation transfer, despite substantially stronger transfer from histology alone. We find that gene-expression embeddings from spot-based spatial transcriptomics are low-dimensional and strongly structured by slide identity, limiting the shared gene--morphology signal available for cross-modal transfer. Guidance benefits some morphologically distinctive classes but reduces held-out gene predictivity broadly across the transcriptome. These results suggest that cross-modal supervision can reorganize information already encoded in a frozen representation but cannot recover information the representation does not retain, highlighting the importance of diagnosing slide-specific structure before applying such supervision. Code availability: https://github.com/ratschlab/vision-guided-transcriptomics-fm-2026

bioinformatics↗

bioRxiv2026

Striping artifact removal in VisiumHD data through nuclear counts modeling

Motivation10x Genomics VisiumHD enables spatial transcriptomics at 2 m x 2 m resolution but exhibits slide-specific, non-periodic striping artifacts due to lane-width variability. These multiplicative row/column effects distort bin total counts and can bias downstream analyses. The state-of-the-art destriping approach is the normalization procedure used as a preprocessing step in bin2cell; it applies sequential high-quantile row- then column-wise normalization, which is asymmetric and can introduce edge effects/macro-stripes and distortions of large-scale total-count structure. ResultsWe propose a statistical destriping approach that leverages nuclei segmentation from the coregistered H&E image. Assuming transcript abundance is constant within each nucleus, we model bin counts with a negative binomial distribution whose mean is a product of a nucleus-specific concentration and row- and column-specific stripe-factors reflecting lane-width variation. We fit all parameters in a generalized linear modeling framework with cross-validated regularization on stripe-factors and iterative dispersion estimation, and use the fitted parameters to correct the observed counts into a destriped image. On synthetic data with known ground truth, our method improves stripe-factor estimation accuracy and reduces error in corrected counts relative to bin2cell and bin2cell-derived baselines. Across four public VisiumHD slides, it consistently lowers striping intensity while substantially better preserving biological signal present in the large-scale global count structure and avoiding the artifacts introduced by other methods. Availability and ImplementationAll source code and links to publicly available data used for this study are available at https://github.com/paolamalsot/destriping-GLM. Contactpaola.malsot@inf.ethz.ch, raetsch@inf.ethz.ch NoteThis manuscript extends the version submitted to Intelligent Systems for Molecular Biology (ISMB) 2026 by describing a new optimization algorithm that yields an approximately tenfold speedup. All plots and benchmarks in this manuscript use the updated implementation.

bioinformatics↗

bioRxiv2026

Towards Cross-Sample Alignment for Multi-Modal Representation Learning in Spatial Transcriptomics

AO_SCPLOWBSTRACTC_SCPLOWThe growing number of spatial transcriptomics (ST) datasets enables comprehensive multi-modal characterization of cell types across diverse biological and clinical contexts. However, integration across patient cohorts remains challenging, as local microenvironment, patient-specific variability, and technical batch effects can dominate signals. Here, we hypothesize that combining specialized transcriptomics correction methods with deep representation learning can jointly align morphology, transcriptomics, and spatial information across multiple tissue samples. This approach benefits from recent transcriptomics and pathology foundation models, projecting cells into a shared embedding space where they cluster by cell type rather than dataset-specific conditions. Applying this framework to 18 skin melanoma, 12 human brain, and 4 lung cancer datasets, we demonstrate that it outperforms conventional batch-correction approaches by 58%, 38%, and 2-fold, respectively. Together, this framework enables efficient integration of multi-modal ST data across modalities and samples, facilitating the systematic discovery of conserved cellular programs and spatial niches while remaining robust to cohort-specific batch effects. Code availabilityhttps://github.com/ratschlab/aestetik

bioinformatics↗

bioRxiv2025

Perturbation-aware representation learning for in vivo genetic screens

CRISPR-based genetic perturbation screens paired with single-cell transcriptomic readouts (Perturb-seq) offer a powerful tool for interrogating biological systems. Yet the resulting datasets are heterogeneous--particularly in vivo--and currently used cell-level perturbation labels reflect only CRISPR guide RNA exposure rather than perturbation state; further, many perturbations have a minimal effect on gene expression. For perturbations that do alter the transcriptomic state of cells, intracellular guide RNA abundance exhibits a dose-response association with perturbation efficacy. We combine (i) per-perturbation, expression-only classifiers trained with non-negative negative-unlabeled (nnNU) risk to yield calibrated scores reflecting the perturbation state of single cells and (ii) a monotone guide abundance prior to yield a per-cell pseudo-posterior that supports both assignment of perturbation probability and selection of affected gene features. To obtain a low-dimensional representation that allows for the accurate reconstruction of gene-level marginals for counterfactual decoding, we train an autoencoder with a quantile-hurdle reconstruction loss and feature-weighted emphasis on perturbation-affected genes. The result is a perturbation-aware latent embedding amenable to downstream trajectory modeling (e.g., optimal transport or flow matching) and a principled probability of perturbation for each non-control cell derived jointly from its guide counts and transcriptome.

bioinformatics↗

bioRxiv2025

DeepSpot2Cell: Predicting Virtual Single-Cell Spatial Transcriptomics from H&E images using Spot-Level Supervision

Spot-based spatial transcriptomics (ST) technologies like 10x Visium quantify genome-wide gene expression and preserve spatial tissue organization. However, their coarse spot-level resolution aggregates signals from multiple cells, preventing accurate single-cell analysis and detailed cellular characterization. Here, we present DeepSpot2Cell, a novel DeepSet neural network that leverages pretrained pathology foundation models and spatial multi-level context to effectively predict virtual single-cell gene expression from histopathological images using spot-level supervision. DeepSpot2Cell substantially improves gene expression correlations on a newly curated benchmark we specifically designed for single-cell ST deconvolution and prediction from H&E images. The benchmark includes 20 lung, 7 breast, and 2 pancreatic cancer samples, across which DeepSpot2Cell outperformed previous super-resolution methods, achieving respective improvements of 46%, 65%, and 38% in cell expression correlation for the top 100 genes. We hope that DeepSpot2Cell and this benchmark will stimulate further advancements in virtual single-cell ST, enabling more precise delineation of cell-type-specific expression patterns and facilitating enhanced downstream analyses. Code availabilityhttps://github.com/ratschlab/DeepSpot2Cell

bioinformatics↗

bioRxiv2022

Aligning Distant Sequences to Graphs using Long Seed Sketches

Sequence-to-graph alignment is an important step in applications such as variant genotyping, read error correction and genome assembly. When a query sequence requires a substantial number of edits to align, approximate alignment tools that follow the seed-and-extend approach require shorter seeds to get any matches. However, in large graphs with high variation, relying on a shorter seed length leads to an exponential increase in spurious matches. We propose a novel seeding approach relying on long inexact matches instead of short exact matches. We demonstrate experimentally that our approach achieves a better time-accuracy trade-off in settings with up to a 25% mutation rate. We achieve this by sketching a subset of graph nodes and storing them in a K-nearest neighbor index. While sketches are more robust to indels, finding the nearest neighbor of a sketch in a high-dimensional space is more computationally challenging than finding exact seeds. We demonstrate that if we store sketch vectors in a K-nearest neighbor index, we can circumvent the curse of dimensionality. Our long sketch-based seed scheme contrasts existing approaches and highlights the important role that tensor sketching can play in bioinformatics applications. Our proposed seeding method and implementation have several advantages: i) We empirically show that our method is efficient and scales to graphs with 1 billion nodes, with time and memory requirements for preprocessing growing linearly with graph size and query time growing quasi-logarithmically with query length. ii) For queries with an edit distance of 25% relative to their length, on the 1 billion node graph, longer sketch-based seeds yield a 4x increase in recall compared to exact seeds. iii) Conceptually, our seeder can be incorporated into other aligners, proposing a novel direction for sequence-to-graph alignment. The implementation is available at: https://github.com/ratschlab/tensor-sketch-alignment.

bioinformatics↗

bioRxiv2022

ResMiCo: increasing the quality of metagenome-assembled genomes with deep learning

The number of published metagenome assemblies is rapidly growing due to advances in sequencing technologies. However, sequencing errors, variable coverage, repetitive genomic regions, and other factors can produce misassemblies, which are challenging to detect for taxonomically novel genomic data. Assembly errors can affect all downstream analyses of the assemblies. Accuracy for the state of the art in reference-free misassembly prediction does not exceed an AUPRC of 0.57, and it is not clear how well these models generalize to real-world data. Here, we present the Residual neural network for Misassembled Contig identification (ResMiCo), a deep learning approach for reference-free identification of misassembled contigs. To develop ResMiCo, we first generated a training dataset of unprecedented size and complexity that can be used for further benchmarking and developments in the field. Through rigorous validation, we show that ResMiCo is substantially more accurate than the state of the art, and the model is robust to novel taxonomic diversity and varying assembly methods. ResMiCo estimated 4.7% misassembled contigs per metagenome across multiple real-world datasets. We demonstrate how ResMiCo can be used to optimize metagenome assembly hyperparameters to improve accuracy, instead of optimizing solely for contiguity. The accuracy, robustness, and ease-of-use of ResMiCo make the tool suitable for general quality control of metagenome assemblies and assembly methodology optimization. Author summaryMetagenome assembly quality is fundamental to all downstream analyses of such data. The number of metagenome assemblies, especially metagenome-assembled genomes (MAGs), is rapidly increasing, but tools to assess the quality of these assemblies lack the accuracy needed for robust quality control. Moreover, existing models have been trained on datasets lacking complexity and realism, which may limit their generalization to novel data. Due to the limitations of existing models, most studies forgo such approaches and instead rely on CheckM to assess assembly quality, an approach that only utilizes a small portion of all genomic information and does not identify specific misassemblies. We harnessed existing large genomic datasets and high-performance computing to produce a training dataset of unprecedented size and complexity and thereby trained a deep learning model for predicting misassemblies that can robustly generalize to novel taxonomy and varying assembly methodologies.

bioinformatics↗

bioRxiv2021

pmVAE: Learning Interpretable Single-Cell Representations with Pathway Modules

MotivationDeep learning techniques have yielded tremendous progress in the field of computational biology over the last decade, however many of these techniques are opaque to the user. To provide interpretable results, methods have incorporated biological priors directly into the learning task; one such biological prior is pathway structure. While pathways represent most biological processes in the cell, the high level of correlation and hierarchical structure make it complicated to determine an appropriate computational representation. ResultsHere, we present pathway module Variational Autoencoder (pmVAE). Our method encodes pathway information by restricting the structure of our VAE to mirror gene-pathway memberships. Its architecture is composed of a set of subnetworks, which we refer to as pathway modules. The subnetworks learn interpretable latent representations by factorizing the latent space according to pathway gene sets. We directly address correlation between pathways by balancing a module-specific local loss and a global reconstruction loss. Furthermore, since many pathways are by nature hierarchical and therefore the product of multiple downstream signals, we model each pathway as a multidimensional vector. Due to their factorization over pathways, the representations allow for easy and interpretable analysis of multiple downstream effects, such as cell type and biological stimulus, within the contexts of each pathway. We compare pmVAE against two other state-of-the-art methods on two single-cell RNA-seq case-control data sets, demonstrating that our pathway representations are both more discriminative and consistent in detecting pathways targeted by a perturbation. Availability and implementationhttps://github.com/ratschlab/pmvae

bioinformatics↗

bioRxiv2020

SF3B1 promotes tumor malignancy through splicing-independent co-activation of HIF1α

Mutations in the splicing factor SF3B1 are frequently occurring in various cancers and drive tumor progression through the activation of cryptic splice sites in multiple genes. Recent studies also demonstrate a positive correlation between expression levels of wildtype SF3B1 and tumor malignancy, but underlying mechanisms remain elusive. Here, we report that SF3B1 acts as an activator of HIF signaling through a splicing-independent mechanism. We demonstrate that SF3B1 forms a heterotrimer with HIF1 and HIF1{beta}, facilitating binding of the HIF1 complex to hypoxia response elements (HREs) to activate target gene expression. We further validate the relevance of this mechanism for tumor progression. Monoallelic deletion of Sf3b1 impedes formation and progression of hypoxic pancreatic cancer via impaired HIF signaling, but is well tolerated in normoxic chromophobe renal cell carcinoma. Our work uncovers an essential role of SF3B1 in HIF1 signaling, providing a causal link between high SF3B1 expression and aggressiveness of solid tumors.

cell biology↗

bioRxiv2019

Characterization of HIF-dependent alternative isoforms in pancreatic cancer

Intra-tumor hypoxia is a common feature in many solid cancers. Although transcriptional targets of hypoxia-inducible factors (HIFs) have been well characterized, alternative splicing or processing of pre-mRNA transcripts which occurs during hypoxia and subsequent HIF stabilization is much less understood. Here, we identify HIF-dependent alternative splicing events after whole transcriptome sequencing in pancreatic cancer cells exposed to hypoxia with and without downregulation of the aryl hydrocarbon receptor nuclear translocator (ARNT), a protein required for HIFs to form a transcriptionally active dimer. We correlate the discovered hypoxia-driven events with available sequencing data from pan-cancer TCGA patient cohorts to select a narrow set of putative biologically relevant splice events for experimental validation. We validate a small set of candidate HIF-dependent alternative splicing events in multiple human cancer cell lines as well as patient-derived human pancreatic cancer organoids. Lastly, we report the discovery of a HIF-dependent mechanism to produce a hypoxia-dependent, long and coding isoform of the UDP-N-acetylglucosamine transporter SLC35A3.

cell biology↗

bioRxiv2017

Metannot: A succinct data structure for compression of colors in dynamic de Bruijn graphs

Much of the DNA and RNA sequencing data available is in the form of high-throughput sequencing (HTS) reads and is currently unindexed by established sequence search databases. Recent succinct data structures for indexing both reference sequences and HTS data, along with associated metadata, have been based on either hashing or graph models, but many of these structures are static in nature, and thus, not well-suited as backends for dynamic databases.\n\nWe propose a parallel construction method for and novel application of the wavelet trie as a dynamic data structure for compressing and indexing graph metadata. By developing an algorithm for merging wavelet tries, we are able to construct large tries in parallel by merging smaller tries constructed concurrently from batches of data.\n\nWhen compared against general compression algorithms and those developed specifically for graph colors (VARI and Rainbowfish), our method achieves compression ratios superior to gzip and VARI, converging to compression ratios of 6.5% to 2% on data sets constructed from over 600 virus genomes.\n\nWhile marginally worse than compression by bzip2 or Rainbowfish, this structure allows for both fast extension and query. We also found that additionally encoding graph topology metadata improved compression ratios, particularly on data sets consisting of several mutually-exclusive reference genomes.\n\nIt was also observed that the compression ratio of wavelet tries grew sublinearly with the density of the annotation matrices.\n\nThis work is a significant step towards implementing a dynamic data structure for indexing large annotated sequence data sets that supports fast query and update operations. At the time of writing, no established standard tool has filled this niche.

bioinformatics↗

bioRxiv2017

Assessing the Gene Regulatory Landscape in 1,188 Human Tumors

Cancer is characterised by somatic genetic variation, but the effect of the majority of non-coding somatic variants and the interface with the germline genome are still unknown. We analysed the whole genome and RNA-Seq data from 1,188 human cancer patients as provided by the Pan-cancer Analysis of Whole Genomes (PCAWG) project to map cis expression quantitative trait loci of somatic and germline variation and to uncover the causes of allele-specific expression patterns in human cancers. The availability of the first large-scale dataset with both whole genome and gene expression data enabled us to uncover the effects of the non-coding variation on cancer. In addition to confirming known regulatory effects, we identified novel associations between somatic variation and expression dysregulation, in particular in distal regulatory elements. Finally, we uncovered links between somatic mutational signatures and gene expression changes, including TERT and LMO2, and we explained the inherited risk factors in APOBEC-related mutational processes. This work represents the first large-scale assessment of the effects of both germline and somatic genetic variation on gene expression in cancer and creates a valuable resource cataloguing these effects.

cancer biology↗