jmiao24/Paper2Agent

20.	Stojnic, R. & Taylor, R. Papers with Code is joining Facebook AI. Medium https://medium.com/paperswithcode/papers-with-code-is-joining-facebook-ai-90b51055f694 (2019).

23.	Anthropic. Claude Code: Deep coding at terminal velocity. Anthropic https://www.anthropic.com/claude-code (2025).

29.	Bloom, J., Tigges, C., Duong, A. & Chanin, D. SAELens. https://github.com/decoderesearch/SAELens (2024).

The orchestrator agent invokes sub-agents as needed at different stages of the process. As the Paper2Agent workflow progresses, the results are automatically recorded for each step for traceability and reproducibility. The detailed setup an

We applied the Paper2Agent framework to the AlphaGenome paper to generate an AlphaGenome MCP and connected the MCP with Claude Code to create the AlphaGenome agent. The generated AlphaGenome MCP server is remotely hosted on Hugging Face Spa

For the AlphaGenome agent generated by Paper2Agent, each benchmark query was wrapped in a system prompt instructing the agent to use the available AlphaGenome MCP tools, return a structured JSON response containing both the final answer and

To evaluate robustness to upstream code defects, we constructed adversarial variants of three representative repositories: AlphaGenome (Python notebook–based), POP-TOOLS  (Python command-line–based)<sup>45</sup>, and mlearner  (R command-li

44.	Sun, E. Single-cell Spatial Transcriptomics Data with Paired RNAseq for TISSUE spatial gene expression prediction. Zenodo https://doi.org/10.5281/zenodo.8259942 (2024).

10x Genomics single-cell RNA-seq datasets: http://cf.10xgenomics.com/samples/cell-exp/3.0.0/pbmc\_1k\_v2/pbmc\_1k\_v2\_filtered\_feature\_bc\_matrix.h5; http://cf.10xgenomics.com/samples/cell-exp/3.0.0/pbmc\_1k\_v3/pbmc\_1k\_v3\_filtered\_f

Mouse somatosensory cortex spatial transcriptomics data (Dataset 15): [https://doi.org/10.5281/zenodo.8259942](<https://doi.org/10.5281/zenodo.8259942>).

ADHD GWAS summary statistics: [https://www.ebi.ac.uk/gwas/studies/GCST90568440](<https://www.ebi.ac.uk/gwas/studies/GCST90568440>) and [https://www.ebi.ac.uk/gwas/studies/GCST90568441](<https://www.ebi.ac.uk/gwas/studies/GCST90568441>).

GTEx portal: [https://gtexportal.org/home/snp/chr1\_109274968\_G\_T\_b38](<https://gtexportal.org/home/snp/chr1_109274968_G_T_b38>)

CD4+ T-cell perturb-seq data: [https://virtualcellmodels.cziscience.com/dataset/genome-scale-tcell-perturb-seq](<https://virtualcellmodels.cziscience.com/dataset/genome-scale-tcell-perturb-seq>)

MPRA-coupled scCRISPRi data for cis-regulatory element perturbation: [https://static-content.springer.com/esm/art%3A10.1038%2Fs41588-025-02301-3/MediaObjects/41588\_2025\_2301\_MOESM4\_ESM.xlsx](<https://static-content.springer.com/esm/art:

Paper2Agent is publicly available at [https://github.com/jmiao24/Paper2Agent](<https://github.com/jmiao24/Paper2Agent>).

AlphaGenome MCP server: [https://huggingface.co/spaces/Paper2Agent/alphagenome\_mcp](<https://huggingface.co/spaces/Paper2Agent/alphagenome_mcp>).

Scanpy MCP server: [https://huggingface.co/spaces/Paper2Agent/scanpy\_mcp](<https://huggingface.co/spaces/Paper2Agent/scanpy_mcp>).

TISSUE MCP server: [https://huggingface.co/spaces/Paper2Agent/tissue\_mcp](<https://huggingface.co/spaces/Paper2Agent/tissue_mcp>).

Paper2Agent-generated AlphaGenome agent is publicly available at [https://huggingface.co/spaces/Paper2Agent/alphagenome\_agent](<https://huggingface.co/spaces/Paper2Agent/alphagenome_agent>).

Reprints and permissions information is available at http://www.nature.com/reprints

[Credential-related source omitted]

[Credential-related source omitted]

[Credential-related source omitted]

[Credential-related source omitted]

[Credential-related source omitted]

[Credential-related source omitted]

[Credential-related source omitted]

[Credential-related source omitted]

[Credential-related source omitted]

[Credential-related source omitted]

[Credential-related source omitted]

Use AlphaGenome (https://github.com/google-deepmind/alphagenome) to answer the

[Credential-related source omitted]

import feather  gtf = feather.read_dataframe(      'https://storage.googleapis.com/alphagenome/reference/gencode/'

gtf = feather.read_dataframe(      'https://storage.googleapis.com/alphagenome/reference/gencode/'      'hg38/gencode.v46.annotation.gtf.gz.feather'

'https://storage.googleapis.com/alphagenome/reference/gencode/'

- [17] Joseph Bloom, Curt Tigges, Anthony Duong, and David Chanin. Saelens. `https://github.com/decoderesearch/SAELens` , 2024.

REFERENCE = "https://github.com/<owner>/<repo>/blob/<commit>/colabs/batch_variant_scoring.ipynb"

The tool verifier in `scripts/verify_mcp_server.py` checks tools only. Add separate client checks for resources/prompts and record their results; listing them does not validate their content. See the official [resources](https://gofastmcp.c

renv's [restore API](https://rstudio.github.io/renv/reference/restore.html) accepts explicit project, library, and lockfile locations. Its [path settings](https://rstudio.github.io/renv/reference/paths.html) allow project-owned caches and l

- [FastMCP installation CLI](https://gofastmcp.com/cli/install-mcp)

- [FastMCP migration requirements](https://gofastmcp.com/getting-started/upgrading/from-fastmcp-3)

- [pytest configuration](https://docs.pytest.org/en/stable/reference/customize.html)

- [uv environments](https://docs.astral.sh/uv/pip/environments/)

- [uv locking](https://docs.astral.sh/uv/pip/compile/)

- [IPython kernels](https://ipython.readthedocs.io/en/stable/install/kernel_install.html)

"schema_version": 1,   "route": "r",   "repository": {"url": "https://github.com/owner/project", "commit": "<full lowercase Git commit>"},

"route": "r",   "repository": {"url": "https://github.com/owner/project", "commit": "<full lowercase Git commit>"},   "tutorials": [

"repository": {"url": "https://github.com/owner/project", "commit": "<full lowercase Git commit>"},

#!/usr/bin/env python3

else base64.b64decode("".join(value.split()), validate=True))

#!/usr/bin/env python3

#!/usr/bin/env python3

[Credential-related source omitted]

#!/usr/bin/env python3

#!/usr/bin/env python3

NS = {"s": "http://schemas.openxmlformats.org/spreadsheetml/2006/main"}

REL = "http://schemas.openxmlformats.org/officeDocument/2006/relationships"

workbook = ET.fromstring(archive.read("xl/workbook.xml"))         ns = {"s": workbook.tag.split("}")[0].lstrip("{")}         if ns["s"] not in {NS["s"], "http://purl.oclc.org/ooxml/spreadsheetml/main"}:

ns = {"s": workbook.tag.split("}")[0].lstrip("{")}         if ns["s"] not in {NS["s"], "http://purl.oclc.org/ooxml/spreadsheetml/main"}:             raise ValueError("Unsupported workbook XML namespace")

if ns["s"] not in {NS["s"], "http://purl.oclc.org/ooxml/spreadsheetml/main"}:

#!/usr/bin/env python3

[Credential-related source omitted]

Executable file

#!/usr/bin/env python3

def make_workbook(path):     namespace = 'xmlns="http://schemas.openxmlformats.org/spreadsheetml/2006/main"'

namespace = 'xmlns="http://schemas.openxmlformats.org/spreadsheetml/2006/main"'

<sheet name="Empty" sheetId="3" r:id="rId3"/>         </sheets></workbook>''',         "xl/_rels/workbook.xml.rels": '''<Relationships xmlns="http://schemas.openxmlformats.org/package/2006/relationships">

</sheets></workbook>''',         "xl/_rels/workbook.xml.rels": '''<Relationships xmlns="http://schemas.openxmlformats.org/package/2006/relationships">           <Relationship Id="rId1" Target="worksheets/sheet1.xml"/>

"xl/_rels/workbook.xml.rels": '''<Relationships xmlns="http://schemas.openxmlformats.org/package/2006/relationships">

with zipfile.ZipFile(regular) as old, zipfile.ZipFile(strict, "w") as new:             for name in old.namelist():                 data = old.read(name).replace(b"http://schemas.openxmlformats.org/spreadsheetml/2006/main", b"http://purl.ocl

for name in old.namelist():                 data = old.read(name).replace(b"http://schemas.openxmlformats.org/spreadsheetml/2006/main", b"http://purl.oclc.org/ooxml/spreadsheetml/main")                 data = data.replace(bundle.REL.encode(

data = old.read(name).replace(b"http://schemas.openxmlformats.org/spreadsheetml/2006/main", b"http://purl.oclc.org/ooxml/spreadsheetml/main")

data = data.replace(bundle.REL.encode(), b"http://purl.oclc.org/ooxml/officeDocument/relationships")