Péter Horváth

This paper introduces a micromanipulation platform to extract viable cell clusters from hydrogel-embedded 3D spheroids without disrupting the parent structure.

The paper identifies gut phages with adhesin proteins that bind epithelial cells and traffic to the endoplasmic reticulum, revealing potential interactions with the human virome.

The paper introduces a subphase-labeled mitotic dataset to improve mitosis detection and subtyping, enhancing domain generalization and enabling biological insights into mitotic subphases.

This study shows that LBR and LAP2 tether heterochromatin to the nuclear periphery, and their loss disrupts genome homeostasis by repositioning heterochromatin and altering gene expression.

The paper demonstrates somatic L1 retrotransposition in adult mouse liver using an intrahepatic reporter assay and identifies distinct regulatory mechanisms in tumor versus normal cells.

This study improves cell type classification in lung cancer tissue sections by incorporating microenvironment information through fisheye transformations, achieving human-level accuracy.

This study identifies VLC fatty acids as mediators of 1-deoxySL toxicity through mitochondrial dysfunction and BAX-dependent apoptosis.

This paper introduces mxDVP, a method combining high-plex imaging and mass spectrometry to profile pancreatic islets, identifying 12 endocrine subtypes and revealing spatial heterogeneity and rare cell states.

This study creates a high-resolution morpho-proteomic atlas of mitosis by decomposing it into 40 subsections and quantifying 4,350 proteins, revealing dynamic changes and links to cancer mutations.

This study uses AI-based image analysis and multi-omics profiling to link morphological features with molecular changes in ccRCC, identifying unique molecular profiles in aggressive rhabdoid tumors.

Spapros selects optimal probe sets for targeted spatial transcriptomics to improve cell type identification and spatial variation detection across multiple datasets.

MorphoNavigator-3D enables automated, annotation-free single-cell analysis of 3D tumor-immune co-cultures, revealing drug-induced phenotypes and their biological relevance.

HCS-3DX is a new AI-driven system for 3D-oid high-content screening that improves resolution and single-cell analysis in 3D tumor models.

Piximi provides a no-programming web tool for biological image analysis using deep learning, enabling rapid insights without installation or coding.

The paper develops a weakly supervised learning method to learn treatment effect representations from imaging data, improving downstream analysis performance.

The paper introduces a semi-automatic annotation tool for 2D and 3D image labelling to reduce expert effort in creating object-based labels.

The paper predicts compound activity using chemical structures, images, and gene-expression profiles, showing that combining these modalities can improve prediction accuracy in drug discovery.

The paper introduces a fisheye transformation for single-cell phenotyping, improving accuracy by weighting nearby cells more heavily than distant ones, and validates it on microscopy and wildlife datasets.

The paper systematically investigates solubility limits of Escherichia coli proteins upon overexpression, revealing that most insoluble proteins aggregate after folding and that high disorder content and low surface stickiness correlate with solubility.

The paper introduces FourierDist, a cell segmentation model using Fourier descriptors, showing high accuracy with fewer coefficients and better performance on non-star-shaped objects.

The paper introduces Regression Plane, a tool for continuous phenotypic analysis using machine learning without requiring class labels.

The paper develops an automated patch clamp system using deep learning and image analysis to measure neurons in human and rodent brain slices.

Pairing AI-driven image analysis with automated laser microdissection and sensitive mass spectrometry yields proteomes of single cells or nuclei chosen by their morphology, with tissue position retained.

This study shows that LGN neurons in awake mice have arousal-dependent activity linked to pupil size and corticothalamic feedback, forming distinct coalitions between thalamocortical neurons and interneurons.

OpSeF provides a Python framework for collaborative instance segmentation of bioimages, integrating pre-trained models and enabling reproducible, modular workflows.

AnnotatorJ is an ImageJ plugin that combines deep learning and manual annotation to improve cell segmentation and annotation efficiency.

The paper introduces a deep learning framework for nucleus segmentation that uses image style transfer to adapt to new data without expert annotations, outperforming existing methods on diverse microscopy images.

This paper applies test-time augmentation to improve cell segmentation accuracy in microscopy images using U-Net and Mask R-CNN models, achieving a better score in the DSB 2018 competition.

3D-Cell-Annotator provides a tool for automated 3D cell segmentation in microscopy images using active surfaces and shape descriptors, achieving expert-level precision.

A deep learning-based algorithm provides an adaptable tool for determining hypocotyl or coleoptile length across different plant species.