2025 Hyper Recent •CC0 1.0 Universal

This work is dedicated to the public domain. No rights reserved.

Access Preprint From Server
April 6th, 2025
Version: 1
Genentech, Stanford
bioinformatics
biorxiv

SpatialAgent: An Autonomous AI Agent for Spatial Biology

Wang, H.Open in Google Scholar•He, Y.Open in Google Scholar•Coelho, P. P.Open in Google Scholar•Bucci, M.Open in Google Scholar•Nazir, A.Open in Google Scholar•Chen, B.Open in Google Scholar•Trinh, L.Open in Google Scholar•Zhang, S.Open in Google Scholar•Huang, K.Open in Google Scholar•Chandrasekar, V.Open in Google Scholaret al.

Advances in AI are transforming scientific discovery, yet spatial biology, a field that deciphers the molecular organization within tissues, remains constrained by labor-intensive workflows. Here, we present SpatialAgent, a fully autonomous AI agent dedicated for spatial-biology research. SpatialAgent integrates large language models with dynamic tool execution and adaptive reasoning. SpatialAgent spans the entire research pipeline, from experimental design to multimodal data analysis and hypothesis generation. Tested on multiple datasets comprising two million cells from human brain, heart, and a mouse colon colitis model, SpatialAgent performance surpassed the best computational methods, matched or outperformed human scientists across key tasks, and scaled across tissues and species. By combining autonomy with human collaboration, SpatialAgent establishes a new paradigm for AI-driven discovery in spatial biology

Similar Papers

biorxiv
Mon Apr 07 2025
Integrative Multi-Omics Analysis of Melanoma: Uncovering Pathways Associated with Immunotherapy Outcomes
Purpose: The treatment of advanced malignant melanoma with immune checkpoint blockade (ICB) therapies such as anti-CTLA4 and anti-PD1 has been transformative, yet a significant proportion of patients demonstrate intrinsic resistance or develop severe immune-related adverse events (irAEs), complicating treatment strategies. This study aimed to integrate clinical and molecular data using multi-omics...
MacBeth, M. L.
•
Kaizer, A. M.
•
Wagner, B. D.
•
Couts, K.
biorxiv
Mon Apr 07 2025
Rapid label-free identification of seven bacterial species using microfluidics, single-cell time-lapse phase-contrast microscopy, and deep learning-based image and video classification
For effective treatment of bacterial infections, it is essential to identify the species causing the infection as early as possible. Currently, the main techniques for determining the species require hours of overnight culturing of a bacterial sample and a larger quantity of cells to function effectively. This study uses one-hour phase-contrast time-lapses of bacterial growth collected from a micr...
Hallstrom, E.
•
Kandavalli, V.
•
Wahlby, C.
•
Hast, A.
biorxiv
Mon Apr 07 2025
SAUNA - Simulated Annealing for Unique Nucleosome Arrangements from Cell-Free DNA-Sequencing Data
The analysis of cell-free DNA (cfDNA) offers a non-invasive approach for cancer detection and monitoring. Nucleosome positioning, inferred from cfDNA coverage data, serves as a proxy for gene expression, but existing peak callers fail to resolve overlapping peaks, limiting their utility in identifying precise nucleosome configurations. Here, we introduce SAUNA (Simulated Annealing for Unique Nucle...
Lambrinos, M. R.
•
Rahimi, T.
•
Mehra, T.
•
Schmid, M. P.
...•
Balazs, Z.
biorxiv
Mon Apr 07 2025
ECLARE: multi-teacher contrastive learning via ensemble distillation for diagonal integration of single-cell multi-omic data
Integrating multimodal single-cell data, such as scRNA-seq and scATAC-seq, is key for decoding gene regulatory networks but remains challenging due to issues like feature harmonization and limited quantity of paired data. To address these challenges, we introduce ECLARE, a novel framework combining multi-teacher ensemble knowledge distillation with contrastive learning for diagonal integration of ...
Mann-Krzisnik, D.
•
Li, Y.
biorxiv
Mon Apr 07 2025
Intratumor Heterogeneity Through the Lens of Gene Regulatory Networks
Tumors exhibit highly heterogeneous genomic and transcriptomic landscapes. Characterization of this intratumor heterogeneity (ITH) and its implications for the cancer therapies has been one of the focal points of computational cancer biology in the past years. However, while ITH at the level of genomic changes, such as copy number aberrations, has been well characterized, the downstream impact of ...
Rao, N.
•
Sapoval, N.
•
Zafar, H.
•
Nakhleh, L.
biorxiv
Mon Apr 07 2025
Fast and memory efficient searching of large-scale mass spectrometry data using Tide
Over the past 30 years, software for searching tandem mass spectrometry data against a protein database has improved dramatically in speed and statistical power. However, existing tools can still struggle to analyze truly massive datasets, when either the number of spectra or the number of proteins being analyzed grows too large. Here we describe enhancements to the Tide search engine that allow i...
Kertesz-Farkas, A.
•
Acquaye, F. L. N. A.
•
Ostapenko, V.
•
Locon, R. H.
...•
Noble, W. S.
biorxiv
Sun Apr 06 2025
Unsupervised Whole-Genome Representation Learning Captures Bacterial Phenotypes
Shifting from hand-crafted to learned representations of data has revolutionized fields like natural language processing and computer vision. Despite this, current approaches to bacterial phenotype prediction from the genome rely on training machine learning models on hand-crafted features, often binary indicators or counts of the presence of different conserved genomic elements and protein domain...
Dufault, C.
•
Moses, A. M.
biorxiv
Sun Apr 06 2025
SEHI-PPI: An End-to-End Sampling-Enhanced Human-Influenza Protein-Protein Interaction Prediction Framework with Double-View Learning
Influenza poses a persistent global health challenge, necessitating innovative approaches to predict host-virus interactions and inform antiviral strategies. Despite advancements, machine-learning-based computational methods typically struggle with limited high-quality negative samples and inadequate modeling of complex host-virus interactions, which hinder predictive accuracy and generalization. ...
Yang, Q.
•
Fan, X.
•
Zhao, H.
•
Ma, Z.
...•
Yin, R.
biorxiv
Sun Apr 06 2025
Enhancing variant detection in complex genomes: leveraging linked reads for robust SNP, Indel, and structural variant analysis
Accurate detection of genetic variants, including single nucleotide polymorphisms (SNPs), small insertions and deletions (INDELs), and structural variants (SVs), is critical for comprehensive genomic analysis. While traditional short-read sequencing performs well for SNP and INDEL detection, it struggles to resolve SVs, especially in complex genomic regions, due to inherent read length limitations...
Luo, C.
•
Liu, Y. H.
•
Liu, H.
•
Zhang, Z.
...•
Zhou, X. M.
biorxiv
Sun Apr 06 2025
Improving hifiasm haplotypes for autopolyploid genome assemblies using constraint programming
Actual versions of hifiasm often produce unbalanced haplotype contigs for autopolyploid genome assemblies. Adding a contraint on the distribution of proteins among haplotypes improves haplotype sizes and busco scores balances. Four species were used to benchmark this approach against hifiasm raw results. The script can be downloaded from https://github.com/chklopp/Separate_haplotypes_with_toulbar2
Schiex, T.
•
Klopp, C.
•
Givry, S. d.
•
Durante, V.