alterlab-geo — independently scanned and version-tracked by SaferSkills.
SaferSkills independently audited alterlab-geo (Agent Skill) and scored it 100/100 (green). The audit ran 55 deterministic rules across Security, Supply Chain, Maintenance, Transparency, and Community; it found 0 high-severity and 0 lower-severity findings. The full rule-by-rule trace and per-finding evidence are below. Free, methodology-open.
Findings & checks · 0 flagged
Every scanned point with the score it earned and what moved between them.
First recorded scan — no prior version to compare against.
The primary manifest — the file an agent reads to learn what this artifact does.
The Gene Expression Omnibus (GEO) is NCBI's public repository for high-throughput gene expression and functional genomics data. It holds hundreds of thousands of studies (Series) and millions of samples from both array-based and sequence-based experiments. For current totals see the GEO browser at https://www.ncbi.nlm.nih.gov/geo/.
Use this skill when searching for gene expression datasets, retrieving experimental data, downloading raw and processed files, querying expression profiles, or integrating GEO data into computational analysis workflows.
Bio.Entrez) → see references/searching.md.GEOparse.get_GEO(geo="GSE...", destdir="./data") → see references/geoparse_usage.md.gse.pivot_samples('VALUE') (genes × samples).references/expression_analysis.md.For bulk downloads, skip GEOparse and pull files directly over FTP → see references/data_access.md.
GEO organizes data hierarchically using different accession types:
GSE123456).Largest organizational unit. Holds experimental design, samples, study info.
(e.g. GSM987654). Linked to platforms and series.
GPL570,Affymetrix Human Genome U133 Plus 2.0 Array). Shared across experiments.
GDS5678)processed for differential analysis. Ideal for quick comparative analyses, but the curated GDS set is legacy and no longer growing — most recent studies exist only as GSE Series, so prefer GSE for new work.
queryable by gene name, cross-referenced to Entrez Gene.
metadata, extract expression matrices, supplementary files, filter samples. → references/geoparse_usage.md
gds, geoprofiles. Always set Entrez.email. → references/searching.md
and supplementary files; no rate limits. → references/data_access.md
https://www.ncbi.nlm.nih.gov/geo/geo2r/?acc=GSExxxxx; generates reproducible R scripts. Useful for exploratory analysis before downloading.
uv pip install GEOparse # primary GEO access (recommended)
uv pip install biopython # E-utilities / programmatic NCBI access
uv pip install pandas numpy scipy statsmodels scikit-learn # analysis
uv pip install matplotlib seaborn # visualizationConfigure NCBI E-utilities access (email is required by NCBI; an API key raises the rate limit from 3 to 10 requests/second):
from Bio import Entrez
Entrez.email = "[email protected]" # required
# Optional API key — https://www.ncbi.nlm.nih.gov/account/
Entrez.api_key = "your_api_key_here"metadata and data tables. Easily parsed by GEOparse.
access and data exchange.
genes/probes as rows). Fastest format for getting expression data.
submissions.
check platform and processing methods.
biological interpretation.
across studies, identify DEGs, run meta-analyses.
response biomarkers, build sensitivity models.
patient subgroup/stage comparisons, expression–outcome correlation.
cohorts, integrate with clinical data.
delays: time.sleep(0.34) (no key) or time.sleep(0.1) (with key).
wget -r for wholedirectories).
destdir; subsequent calls reuse them.batch metadata, FTP for direct/bulk file pulls. Cache locally; always set Entrez.email with Biopython.
platform annotation and processing methods, verify metadata and design, watch for batch effects, and consider reprocessing raw data.
CEL) can be very large. Plan disk space; download incrementally.
database (Barrett et al. 2013, Nucleic Acids Research). Check per-dataset usage restrictions and follow NCBI guidelines.
mapping); values may be raw/normalized/log-transformed (check metadata); sample metadata is inconsistently formatted; older submissions may lack series matrix files; platform annotations may be outdated.
references/searching.md — E-utilities search (datasets, profiles, advancedqueries, search→summary→fetch workflow, batch metadata).
references/geoparse_usage.md — GEOparse install/usage, expression matrices,supplementary files, sample filtering, caching.
references/data_access.md — direct FTP access (ftplib, wget, curl) and theaccession-to-path rule.
references/expression_analysis.md — QC/preprocessing, differential expression,correlation/clustering, batch processing, cross-study meta-analysis.
references/geo_reference.md — in-depth technical reference: E-utilities APIendpoints, SOFT/MINiML format docs, FTP directory structure, normalization pipelines, platform quirks, and error-handling troubleshooting.
scripts/query_geo.py — runnable helper for GEO DataSets via NCBI E-utilities (no key; NCBI_API_KEY lifts the rate limit). Has a PEP 723 inline-dependency header, so uv run installs requests automatically:
uv run scripts/query_geo.py search "breast cancer AND Homo sapiens[ORGN]" --retmax 5
uv run scripts/query_geo.py summary 200000001,200000002~30 seconds. Free. No account. Every finding cites a rule and a line of evidence.