Biology subjects

Ratsch, G.

Publications and source records attributed to Ratsch, G..

11 recordsLinked to original sources

bioRxiv2026

Probabilistic uncertainty in virtual spatial transcriptomics from histology is largely redundant with the predicted mean

Deep learning has enabled virtual spatial transcriptomics from histology, offering a scalable approach to molecular profiling of routinely archived tissue without molecular assays. However, reliable uncertainty estimates are essential for identifying potentially inaccurate predictions, and it remains unclear whether existing methods provide useful information about prediction reliability. Here, we present a systematic benchmark of uncertainty estimation for virtual spatial transcriptomics across three cancer cohorts spanning melanoma, lung, and kidney tumors. We evaluate aleatoric and epistemic uncertainty using probabilistic likelihoods and approximate Bayesian methods alongside histology-free and training-free baselines, assessing both calibration and error informativeness. Across cohorts and methods, uncertainty is associated with prediction error, but the predicted mean largely explains this association. After accounting for the mean, uncertainty provides only weak information about larger errors. Moreover, good calibration does not imply informative uncertainty, with simple distributions fitted without histology achieving comparable calibration. We therefore distinguish calibrated from informative uncertainty and argue that uncertainty should provide information about prediction error beyond the predicted mean. This benchmark provides a framework for systematic uncertainty evaluation in virtual spatial transcriptomics. Code availability: https://github.com/ratschlab/uncertainty-vst-2026

bioinformatics↗

bioRxiv2026

Transcriptomics-Conditioned Virtual Tissue Synthesis via Diffusion Transformers

Spatial transcriptomics couples hematoxylin and eosin (H&E) tissue morphology with spatially resolved gene expression (GE). However, generative models that exploit this coupling to synthesize tissue images from transcriptomic profiles remain scarce. We present STMDiT (Spatial Transcriptomics and Morphology Diffusion Transformer), a diffusion transformer that synthesizes H&E histopathology patches conditioned jointly on morphological embeddings and transcriptomic profiles. Building on PixCell (Yellapragada et al., 2025), we integrate gene expression from a frozen CancerFoundation encoder (Theus et al., 2024) through adaptive layer normalization and per-block cross-attention, and we train under dual classifier-free guidance with independent modality dropout. On the 10x TuPro Visium melanoma cohort, GE conditioning improves both image quality over the no-GE PixCell-B baseline (best FID = 252.9 vs 330.7) and transcriptomic fidelity (best AUC = 0.267 vs 0.229, reaching 82% of the real-tile ceiling). Training with DeepSpots predicted-transcriptomics pseudo-labels (PTPL) uniquely transfers zero-shot to TCGA SKCM, an out-of-distribution (OOD) H&E-only melanoma cohort: PTPL-XAttn-PMA-B reaches FID = 690.0, a 57-point improvement over the no-GE baseline (747.1), with a within-model GE-ablation effect of {Delta}OOD = +309.5, enabling virtual tissue synthesis beyond native spatial-transcriptomics coverage. Our results indicate that gene-expression conditioning produces morphologically distinct tissue images and supports virtual tissue simulation for hypothesis testing in computational pathology.

bioinformatics↗

bioRxiv2024

SMAC mimetics overcome apoptotic resistance in ovarian cancer through MSLN-TNF alpha axis

Resistance to chemotherapy and PARPi inhibitors remains a critical challenge in the treatment of epithelial ovarian cancer, mainly due to disabled apoptotic responses in tumor cells. Given mesothelins pivotal role in ovarian cancer and its restricted expression in healthy tissues, we conducted a drug-screening discovery analysis across a range of genetically modified cancer cells to unveil mesothelins therapeutic impact. We observed enhanced cell death in cancer cells with low mesothelin expression when exposed to a second mitochondria-derived activator of caspases (SMAC) mimetics, and demonstrated a compelling synergy when combined with chemotherapy in ex vivo patient-derived cultures and zebrafish tumor xenografts. Mechanistically, the addition of the SMAC mimetics drug birinapant to either carboplatin or paclitaxel triggered the activation of the Caspase 8-dependent apoptotic program facilitated by TNFLJ signaling. Multimodal analysis of neoadjuvant-treated patient samples further revealed an association between tumor-associated macrophages and the activation of TNFLJ-related pathways. Our proposed bimodal treatment shows promise in enhancing the clinical management of patients by harnessing the potential of SMAC mimetics alongside conventional chemotherapy.

cancer biology↗

bioRxiv2023

SimReadUntil for Benchmarking Selective Sequencing Algorithms on ONT Devices

MotivationThe Oxford Nanopore Technologies (ONT) ReadUntil API enables selective sequencing, which aims to reduce time spent on sequencing uninteresting reads in favor of more interesting reads, e.g., to deplete or enrich certain genomic regions. The performance gain depends on the selective sequencing decision-making algorithm (SSDA) which decides whether to reject a read, stop receiving a read or wait for more data. Since real runs are time-consuming and costly (at scale), simulating the ONT device with support for the ReadUntil API is highly beneficial to compare and optimize the parameters of SSDAs. Existing software like MinKNOW and UNCALLED only return raw signal data, are memory-intensive, require huge and often unavailable multi-fast5 files ([≥] 100GB) and are not clearly documented. ResultsWe present the ONT device simulator SimReadUntil that takes a set of full (real or simulated) reads as input, distributes them to channels and plays them back in real time including mux scans, channel gaps and blockages, and allows to unblock (reject) reads as well as stop receiving data from them (imitating the ReadUntil API). Our modified ReadUntil API provides the basecalled reads rather than the raw signal to reduce computational load and focus on the SSDA rather than basecalling. Tuning the parameters of tools like ReadFish and ReadBouncer becomes easier because no GPU is required anymore for basecalling. We offer various methods to extract simulation parameters from a sequencing summary file and compare them. SimReadUntil s gRPC interface allows standardized interaction with a wide range of programming languages. AvailabilityThe code is freely available on GitHub (https://github.com/ratschlab/sim_read_until) along with a fully worked use case that combines the simulator with ReadFish (and optionally NanoSim). Supplementary informationSupplementary data are available at Bioinformatics online.

bioinformatics↗

bioRxiv2022

PipeIT2: A tumor-only somatic variant calling workflow for Molecular Diagnostic Ion Torrent sequencing data

Precision oncology relies on the accurate identification of somatic mutations in cancer patients. While the sequencing of the tumoral tissue is frequently part of routine clinical care, the healthy counterparts are rarely sequenced. We previously published PipeIT, a somatic variant calling workflow specific for Ion Torrent sequencing data enclosed in a Singularity container. PipeIT combines user-friendly execution, reproducibility and reliable mutation identification, but relies on matched germline sequencing data to exclude germline variants. Expanding on the original PipeIT, here we describe PipeIT2 to address the clinical need to define somatic mutations in the absence of germline control. We show that PipeIT2 achieves a >95% recall for variants with variant allele fraction >10%, reliably detects driver and actionable mutations and filters out most of the germline mutations and sequencing artifacts. With its performance, reproducibility and ease of execution, PipeIT2 is a valuable addition to molecular diagnostics laboratories.

bioinformatics↗

bioRxiv2021

Learning Single-Cell Perturbation Responses using Neural Optimal Transport

The ability to understand and predict molecular responses towards external perturbations is a core question in molecular biology. Technological advancements in the recent past have enabled the generation of high-resolution single-cell data, making it possible to profile individual cells under different experimentally controlled perturbations. However, cells are typically destroyed during measurement, resulting in unpaired distributions over either perturbed or non-perturbed cells. Leveraging the theory of optimal transport and the recent advents of convex neural architectures, we learn a coupling describing the response of cell populations upon perturbation, enabling us to predict state trajectories on a single-cell level. We apply our approach, CO_SCPLOWELLC_SCPLOWOT, to predict treatment responses of 21,650 cells subject to four different drug perturbations. CO_SCPLOWELLC_SCPLOWOT outperforms current state-of-the-art methods both qualitatively and quantitatively, accurately capturing cellular behavior shifts across all different drugs.

bioinformatics↗

bioRxiv2020

Using Genome Graph Topology to Guide Annotation Matrix Sparsification

Since the amount of published biological sequencing data is growing exponentially, efficient methods for storing and indexing this data are more needed than ever to truly benefit from this invaluable resource for biomedical research. Labeled de Bruijn graphs are a frequently-used approach for representing large sets of sequencing data. While significant progress has been made to succinctly represent the graph itself, efficient methods for storing labels on such graphs are still rapidly evolving. In this paper, we present RowDiff, a new technique for compacting graph labels by leveraging expected similarities in annotations of vertices adjacent in the graph. RowDiff can be constructed in linear time relative to the number of vertices and labels in the graph, and in space proportional to the graph size. In addition, construction can be efficiently parallelized and distributed, making the technique applicable to graphs with trillions of nodes. RowDiff can be viewed as an intermediary sparsification step of the original annotation matrix and can thus naturally be combined with existing generic schemes for compressed binary matrices. Experiments on 10,000 RNA-seq datasets show that RowDiff combined with Multi-BRWT results in a 30% reduction in annotation footprint over Mantis-MST, the previously known most compact annotation representation. Experiments on the sparser Fungi subset of the RefSeq collection show that applying RowDiff sparsification reduces the size of individual annotation columns stored as compressed bit vectors by an average factor of 42. When combining RowDiff with a Multi-BRWT representation, the resulting annotation is 26 times smaller than Mantis-MST.

bioinformatics↗

bioRxiv2020

Fast Alignment-Free Similarity Estimation By Tensor Sketching

The sharp increase in next-generation sequencing technologies capacity has created a demand for algorithms capable of quickly searching a large corpus of biological sequences. The complexity of biological variability and the magnitude of existing data sets have impeded finding algorithms with guaranteed accuracy that efficiently run in practice. Our main contribution is the Tensor Sketch method that efficiently and accurately estimates edit distances. In our experiments, Tensor Sketch had 0.956 Spearmans rank correlation with the exact edit distance, improving its best competitor Ordered MinHash by 23%, while running almost 5 times faster. Finally, all sketches can be updated dynamically if the input is a sequence stream, making it appealing for large-scale applications where data cannot fit into memory. Conceptually, our approach has three steps: 1) represent sequences as tensors over their sub-sequences, 2) apply tensor sketching that preserves tensor inner products, 3) implicitly compute the sketch. The sub-sequences, which are not necessarily contiguous pieces of the sequence, allow us to outperform k-mer-based methods, such as min-hash sketching over a set of k-mers. Typically, the number of sub-sequences grows exponentially with the sub-sequence length, introducing both memory and time overheads. We directly address this problem in steps 2 and 3 of our method. While the sketching of rank-1 or super-symmetric tensors is known to admit efficient sketching, the sub-sequence tensor does not satisfy either of these properties. Hence, we propose a new sketching scheme that completely avoids the need for constructing the ambient space. Our tensor-sketching techniques main advantages are three-fold: 1) Tensor Sketch has higher accuracy than any of the other assessed sketching methods used in practice. 2) All sketches can be computed in a streaming fashion, leading to significant time and memory savings when there is overlap between input sequences. 3) It is straightforward to extend tensor sketching to different settings leading to efficient methods for related sequence analysis tasks. We view tensor sketching as a framework to tackle a wide range of relevant bioinformatics problems, and we are confident that it can bring significant improvements for applications based on edit distance estimation.

bioinformatics↗

bioRxiv2020

MetaGraph: Indexing and Analysing Nucleotide Archives at Petabase-scale

The amount of biological sequencing data available in public repositories is growing exponentially, forming an invaluable biomedical research resource. Yet, making it full-text searchable and easily accessible to researchers in life and data science is an unsolved problem. In this work, we take advantage of recently developed, very efficient data structures and algorithms for representing sequence sets. We make Petabases of DNA sequences across all clades of life, including viruses, bacteria, fungi, plants, animals, and humans, fully searchable. Our indexes are freely available to the research community. This highly compressed representation of the input sequences (up to 5800x) fits on a single consumer hard drive ({approx}100 USD), making this valuable resource cost-effective to use and easily transportable. We present the underlying methodological framework, called MetaGraph, that allows us to scalably index very large sets of DNA or protein sequences using annotated De Bruijn graphs. We demonstrate the feasibility of indexing the full extent of existing sequencing data and present new approaches for efficient and cost-effective full-text search at an on-demand cost of $0.10 per queried Mpb. We explore several practical use cases to mine existing archives for interesting associations and demonstrate the utility of our indexes for integrative analyses.

bioinformatics↗

bioRxiv2019

A Bayesian Nonparametric Approach to Discover Clinico-Genetic Associations across Cancer Types

MotivationPersonalized medicine aims at combining genetic, clinical, and environmental data to improve medical diagnosis and disease treatment, tailored to each patient. This paper presents a Bayesian nonparametric (BNP) approach to identify genetic associations with clinical/environmental features in cancer. We propose an unsupervised approach to generate data-driven hypotheses and bring potentially novel insights about cancer biology. Our model combines somatic mutation information at gene-level with features extracted from the Electronic Health Record. We propose a hierarchical approach, the hierarchical Poisson factor analysis (H-PFA) model, to share information across patients having different types of cancer. To discover statistically significant associations, we combine Bayesian modeling with bootstrapping techniques and correct for multiple hypothesis testing.\n\nResultsUsing our approach, we empirically demonstrate that we can recover well-known associations in cancer literature. We compare the results of H-PFA with two other classical methods in the field: case-control (CC) setups, and linear mixed models (LMMs).

bioinformatics↗

bioRxiv2016

Genomic Rearrangements Considered as Quantitative Traits

To understand the population genetics of structural variants (SVs), and their effects on phenotypes, we developed an approach to mapping SVs, particularly transpositions, segregating in a sequenced population, and which avoids calling SVs directly. The evidence for a potential SV at a locus is indicated by variation in the counts of short-reads that map anomalously to the locus. These SV traits are treated as quantitative traits and mapped genetically, analogously to a gene expression study. Association between an SV trait at one locus and genotypes at a distant locus indicate the origin and target of a transposition. Using ultra-low-coverage (0.3x) population sequence data from 488 recombinant inbred Arabidopsis genomes, we identified 6,502 segregating SVs. Remarkably, 25% of these were transpositions. Whilst many SVs cannot be delineated precisely, PCR validated 83% of 44 predicted transposition breakpoints. We show that specific SVs may be causative for quantitative trait loci for germination, fungal disease resistance and other phenotypes. Further we show that the phenotypic heritability attributable to sequence anomalies differs from, and in the case of time to germination and bolting, exceeds that due to standard genetic variation. Gene expression within SVs is also more likely to be silenced or dysregulated. This approach is generally applicable to large populations sequenced at low-coverage, and complements the prevalent strategy of SV discovery in fewer individuals sequenced at high coverage.

genetics↗