# INST-Align: Implicit Neural Alignment for Spatial Transcriptomics via Canonical Expression Fields

Source: [INST-Align: Implicit Neural Alignment for Spatial Transcriptomics via Canonical Expression Fields](<https://www.sota2.com/research/paper/69df1b9f7d0172140ec7b9a1>)

Collection index: [llms.txt](<https://www.sota2.com/research/llms.txt>)

Data updated: 2026-04-16T05:00:04.684Z

Authors: Bonian Han, Cong Qi, Przemyslaw Musialski, Zhi Wei

Published: 2026-04-13T21:44:18.000Z

## Abstract

Spatial transcriptomics (ST) measures mRNA expression while preserving spatial organization, but multi-slice analysis faces two coupled difficulties: large non-rigid deformations across slices and inter-slice batch effects when alignment and integration are treated independently. We present INST-Align, an unsupervised pairwise framework that couples a coordinate-based deformation network with a shared Canonical Expression Field, an implicit neural representation mapping spatial coordinates to expression embeddings, for joint alignment and reconstruction. A two-phase training strategy first establishes a stable canonical embedding space and then jointly optimizes deformation and spatial-feature matching, enabling mutually constrained alignment and representation learning. Cross-slice parameter sharing of the canonical field regularizes ambiguous correspondences and absorbs batch variation. Across nine datasets, INST-Align achieves state-of-the-art mean OT Accuracy (0.702), NN Accuracy (0.719), and Chamfer distance, with Chamfer reductions of up to 94.9\\% on large-deformation sections relative to the strongest baseline. The framework also yields biologically meaningful spatial embeddings and coherent 3D tissue reconstruction. The code will be released after review phase.

## Recorded benchmark results

A research result applies to its stated dataset, split, metric and evaluation configuration. Compare like-for-like settings, preserve the metric's unit and direction, and distinguish paper-reported results from independent reproduction. The best result recorded here is not a claim of exhaustive, current global coverage.

- [Spatial alignment on DLPFC D3](<https://www.sota2.com/research/sota/spatial-alignment-on-dlpfc-d3.md>): Spatial Alignment; DLPFC D3; OT Accuracy: 83.9.

- [Spatial alignment on MERFISH Brain MB11](<https://www.sota2.com/research/sota/spatial-alignment-on-merfish-brain-mb11.md>): Spatial Alignment; MERFISH Brain MB11; OT Acc: 0.357.

- [Spatial alignment on MERFISH Hypothalamus (MH)](<https://www.sota2.com/research/sota/spatial-alignment-on-merfish-hypothalamus-mh.md>): Spatial Alignment; MERFISH Hypothalamus (MH); OT Accuracy: 69.

- [Spatial alignment on MouseEmbryo (ME)](<https://www.sota2.com/research/sota/spatial-alignment-on-mouseembryo-me.md>): Spatial Alignment; MouseEmbryo (ME); OT Acc: 74.6.

- [Clustering on MouseEmbryo](<https://www.sota2.com/research/sota/clustering-on-mouseembryo.md>): Clustering; MouseEmbryo; ARI: 0.341.

- [Spatial alignment on MERFISH Brain MB2](<https://www.sota2.com/research/sota/spatial-alignment-on-merfish-brain-mb2.md>): Spatial Alignment; MERFISH Brain MB2; OT Acc: 31.8.

- [Spatial alignment on MERFISH Brain MB7](<https://www.sota2.com/research/sota/spatial-alignment-on-merfish-brain-mb7.md>): Spatial Alignment; MERFISH Brain MB7; OT Accuracy: 74.6.

- [Clustering on DLPFC Sample 1](<https://www.sota2.com/research/sota/clustering-on-dlpfc-sample-1.md>): Clustering; DLPFC Sample 1; ARI: 39.7.

- [Clustering on DLPFC Sample 2](<https://www.sota2.com/research/sota/clustering-on-dlpfc-sample-2.md>): Clustering; DLPFC Sample 2; ARI: 0.463.

- [Spatial alignment on DLPFC D1](<https://www.sota2.com/research/sota/spatial-alignment-on-dlpfc-d1.md>): Spatial Alignment; DLPFC D1; OT Accuracy: 86.3.

- [Spatial alignment on DLPFC D2](<https://www.sota2.com/research/sota/spatial-alignment-on-dlpfc-d2.md>): Spatial Alignment; DLPFC D2; OT Acc: 89.9.

- [Clustering on DLPFC Sample 3](<https://www.sota2.com/research/sota/clustering-on-dlpfc-sample-3.md>): Clustering; DLPFC Sample 3; ARI: 24.6.

- [Spatial alignment on STARMap ST](<https://www.sota2.com/research/sota/spatial-alignment-on-starmap-st.md>): Spatial Alignment; STARMap ST; OT Acc: 85.7.

## Original sources and implementation

[Original paper](<https://arxiv.org/abs/2604.12084>)

[PDF](<https://arxiv.org/pdf/2604.12084.pdf>)
