Skip to content

Commit 4f2e953

Browse files
committed
reviewed text
1 parent 73d68b2 commit 4f2e953

1 file changed

Lines changed: 8 additions & 10 deletions

File tree

content/blog/2025-07-biomni.md

Lines changed: 8 additions & 10 deletions
Original file line numberDiff line numberDiff line change
@@ -8,13 +8,13 @@ draft = false
88

99
![][image1]
1010

11-
# scverse × Biomni: Democratizing Single-Cell Omics Analysis
11+
# scverse × Biomni: Agentic Single-Cell Omics Analysis
1212

13-
Single-cell and spatial omics have unlocked unprecedented insights into cellular diversity, tissue architecture, and drug responses. Yet today, most researchers must stitch together fragmented code across multiple packages—writing custom scripts for data loading, normalization, clustering, spatial feature extraction, perturbation modeling, and visualization. This process demands deep programming expertise, detailed knowledge of each library’s API, and often takes days even weeks to build, debug, and reproduce.
13+
Single-cell and spatial omics have unlocked unprecedented insights into cellular diversity, tissue architecture, and drug responses. Despite the remarkable progress in computational tools, the diversity and complexity of analyses can still pose challenges. While the scverse ecosystem provides powerful and interoperable tools such as scanpy, scvi-tools, squidpy, anndata, mudata, and SpatialData, researchers can sometimes face a steep learning curve, particularly when integrating multiple analytical steps or modalities.
1414

15-
**scverse** is a community-driven and open-source initiative behind many of the most widely adopted Python tools in single-cell biology, including scanpy, scvi-tools, squidpy, and more. Built on shared data structures, anndata, mudata, and SpatialData, scverse promotes modular, interoperable, and scalable analysis across modalities, from transcriptomics to spatial and immune profiling.
15+
**scverse** is a community-driven, open-source initiative behind many of the most widely adopted Python tools in single-cell biology, known for promoting modular, interoperable, and scalable analysis across diverse modalities—from transcriptomics to spatial and immune profiling.
1616

17-
We’re excited to announce a partnership with **scverse** to fundamentally change how scientists interact with single-cell and spatial omics tools. Biomni now acts as an **intelligent agent for the scverse ecosystem**, capable of writing, executing, and managing code across ten core scverse packages on your behalf. Researchers can describe their analysis goals in plain English—e.g., *“cluster cells and identify markers”* or *“analyze perturbation effects between treatment groups”*—and Biomni automatically generates and runs the corresponding scanpy, pertpy, squidpy, or scvi-tools code. The agent understands biological context, handles parameters and dependencies, and returns reproducible results—no coding required. Notably, all codes written by agents are documented and available to users for reproduction or modification as a Jupyter notebook. This is an early-stage capability, and manual review of outputs is encouraged to ensure accuracy, and we invite constructive community feedback to improve.
17+
We’re excited to announce a collaboration between **scverse** and **Biomni** to further streamline and enhance single-cell and spatial omics analyses. Biomnis intelligent agentic interface os now capable of using scverse tools, enabling researchers to seamlessly integrate, execute, and manage analyses across ten core scverse packages through natural language prompts. Researchers can describe their analysis goals in plain English—e.g., *“cluster cells and identify markers”* or *“analyze perturbation effects between treatment groups”*—and Biomni automatically generates and runs the corresponding scanpy, pertpy, squidpy, or scvi-tools code. The agent understands biological context, handles parameters and dependencies, and returns reproducible results—no coding required. Notably, all codes written by agents are documented and available to users for reproduction or modification as a Jupyter notebook. This is an early-stage capability, and manual review of outputs is encouraged to ensure accuracy, and we invite constructive community feedback to improve.
1818

1919
---
2020

@@ -32,7 +32,7 @@ To illustrate, here are a few examples of how Biomni equipped with scverse packa
3232

3333
* *“Detect spatially variable genes in this Visium dataset and overlay results on the H\&E image.”*
3434

35-
→ Biomni leverages squidpy and spatialdata for spatial feature detection and plotting.
35+
→ Biomni leverages Squidpy and SpatialData for spatial feature detection and plotting.
3636

3737
---
3838

@@ -65,9 +65,7 @@ Biomni has successfully completed these steps in around 20 minutes. It preproces
6565

6666
### **Explore Single-cell Data to Generate New Hypotheses**
6767

68-
Biomni empowers researchers to go beyond standard analysis, treating single-cell and spatial omics datasets as hypothesis-generating engines. Through open-ended, natural language prompts, scientists can uncover patterns, outliers, and biological structure that spark new lines of inquiry.
69-
70-
Whether you’re examining cellular heterogeneity across disease states or discovering unexpected spatial niches, Biomni supports exploratory analysis guided by curiosity and domain expertise.
68+
Biomni also supports curiosity-driven no-code exploratory and hypothesis-driven analyses. By interacting with datasets through natural language prompts, researchers can easily identify biological patterns and generate meaningful hypotheses for experimental validation.
7169

7270
**Input:**
7371

@@ -111,7 +109,7 @@ By combining data-driven discovery with biological reasoning, Biomni helps resea
111109

112110
---
113111

114-
## **Access scverse through Biomni**
112+
## **Access scverse tools through Biomni**
115113

116114
* **Biomni Web Platform**
117115

@@ -123,4 +121,4 @@ By combining data-driven discovery with biological reasoning, Biomni helps resea
123121

124122
---
125123

126-
By wrapping scanpy, muon, anndata, mudata, scvi-tools, pertpy, decoupler, squidpy, spatialdata, and scirpy into an AI agent, Biomni × scverse removes the coding hurdle from single-cell and spatial omics research. Simply ask your question—and focus on biology, not boilerplate code. While the long-term vision is for agents to autonomously handle full experimental designs, human-in-the-loop remains critical today. We can’t wait to see the discoveries you make.
124+
This integration between Biomni and scverse highlights a powerful synergy, removing coding barriers while maintaining transparency, reproducibility, and scientific rigor. Although agent-based automation enhances convenience, human expertise remains vital for biological interpretation and validation. We look forward to seeing the innovative discoveries our community achieves with this exciting collaboration.

0 commit comments

Comments
 (0)