PhySeqNet: physics-informed fluorescent foci detection for BARseq spatial transcriptomics
Abstract
BARseq enables spatially resolved gene-expression profiling across intact tissue, but its sensitivity depends on accurate detection and decoding of dense fluorescent rolonies. Existing spot-calling approaches can be conservative where signals overlap, while supervised deep learning is limited by the lack of reliable manual ground truth in real tissue images. Here we present PhySeqNet, a physics-informed deep learning framework for annotation-free fluorescent foci detection in BARseq. PhySeqNet trains U-Net detectors entirely on semi-synthetic images generated from an instrument-matched optical model combined with real sample-free experimental noise, and is integrated into pyBARseq, a modular Python pipeline for BARseq analysis. PhySeqNet achieved high detection performance on held-out simulated data across wavelength bands. On real tissue images, where ground truth is unavailable, we evaluated detections by forward-model reconstruction: PhySeqNet with conservative barcode-aware non-maxima suppression (NMSB) achieved higher reconstruction fidelity (NCC and MS-SSIM) than the BarDensr single-shot estimator, tuned classical spot detectors, and a pretrained deep-learning detector. In the analyzed brain section, PhySeqNet with NMSB yielded a median of 51 counts per cell versus 47 for BarDensr, with a similar number of genes detected per cell. The recovered transcripts produced spatial expression patterns qualitatively consistent with Allen MERFISH reference data and preserved cell-type composition after scANVI label transfer. These results show that physics-informed simulation can train practical, annotation-free detectors that improve BARseq sensitivity without disrupting downstream biological interpretation.
Related
- Cutting-edge microfluidic technology allows detection and sorting of single fluorescent extracellular vesicles
- Rapid detection of major Gram-positive pathogens in ocular specimens using a novel fluorescent vancomycin-based probe
- AI-Generated Peptide Vectors: Advancing Fluorescent Detection of Bacteria
- Spatially Informed and Direction Sensitive Representation Learning for Spatial Transcriptomics
- A Physics-Informed Neural Network Surrogate for Patient-Specific Hepatic Arterial Hemodynamics in Yttrium-90 Radioembolization: Network Architecture, Boundary-Condition Enforcement, and Data Efficiency
BibTeX
@misc{goswami2026physeqnet,
title = {{PhySeqNet: physics-informed fluorescent foci detection for BARseq spatial transcriptomics}},
author = {Neha Goswami and John Hover and Renae Galluccio and Yichen Wu and Huiqing Zhan and Xiaoyin Chen and Anthony M. Zador},
howpublished = {arXiv},
year = {2026},
month = oct,
doi = {10.48550/arxiv.2610.13779},
eprint = {2610.13779},
archivePrefix = {arXiv},
url = {https://arxiv.org/abs/2610.13779},
note = {CC BY 4.0},
}
RIS
TY - UNPB
TI - PhySeqNet: physics-informed fluorescent foci detection for BARseq spatial transcriptomics
AU - Neha Goswami
AU - John Hover
AU - Renae Galluccio
AU - Yichen Wu
AU - Huiqing Zhan
AU - Xiaoyin Chen
AU - Anthony M. Zador
PY - 2026
DA - 2026/10/09
PB - arXiv
DO - 10.48550/arxiv.2610.13779
UR - https://arxiv.org/abs/2610.13779
AB - BARseq enables spatially resolved gene-expression profiling across intact tissue, but its sensitivity depends on accurate detection and decoding of dense fluorescent rolonies. Existing spot-calling approaches can be conservative where signals overlap, while supervised deep learning is limited by the lack of reliable manual ground truth in real tissue images. Here we present PhySeqNet, a physics-informed deep learning framework for annotation-free fluorescent foci detection in BARseq. PhySeqNet trains U-Net detectors entirely on semi-synthetic images generated from an instrument-matched optical model combined with real sample-free experimental noise, and is integrated into pyBARseq, a modular Python pipeline for BARseq analysis. PhySeqNet achieved high detection performance on held-out simulated data across wavelength bands. On real tissue images, where ground truth is unavailable, we evaluated detections by forward-model reconstruction: PhySeqNet with conservative barcode-aware non-maxima suppression (NMSB) achieved higher reconstruction fidelity (NCC and MS-SSIM) than the BarDensr single-shot estimator, tuned classical spot detectors, and a pretrained deep-learning detector. In the analyzed brain section, PhySeqNet with NMSB yielded a median of 51 counts per cell versus 47 for BarDensr, with a similar number of genes detected per cell. The recovered transcripts produced spatial expression patterns qualitatively consistent with Allen MERFISH reference data and preserved cell-type composition after scANVI label transfer. These results show that physics-informed simulation can train practical, annotation-free detectors that improve BARseq sensitivity without disrupting downstream biological interpretation.
ER -