Alim430

bioresearch-ancestry-aware-mr

Run cross-ancestry Mendelian randomization with correlated-pleiotropy modeling (CAUSE-like) and mixture-model pleiotropy detection (MRMix-like), plus per-ancestry IVW, cross-ancestry meta-analysis (FE/RE), and portability assessment. Use when the user needs to test whether a causal effect estimated in one ancestry transfers to others, or to distinguish true causation from correlated horizontal pleiotropy. Mock mode validates the MR + pleiotropy pipeline; live mode would use BBJ/FinnGen/TPMI/UKB summary statistics.

bioresearch-foundation-embeddings

Generate and evaluate single-cell foundation model embeddings (scGPT / UCE / scFoundation). Runs mock-mode pipeline validation by default — simulated embeddings that recover planted cell-type structure. Live mode stubs (pip install scgpt, UCE eval_single_anndata.py, scFoundation get_embedding.py) are documented for real deployment. Use when the user wants to embed scRNA-seq data with foundation models, compare embedding quality across models, or benchmark cell-type recovery.

bioresearch-gwas-harmonization

Harmonize GWAS summary statistics across ancestries for cross-ancestry MR. Use when the user needs to align alleles, flip strands, resolve palindromic SNPs, compute allele-frequency divergence (Fst-like), or identify cross-ancestry genome-wide signal overlap. Supports GWAS-SSF schema; mock mode validates the harmonization pipeline with simulated cross-ancestry GWAS including strand confusion and allele flips.

bioresearch-biomarker-discovery

Run reproducible biomarker discovery via the BioResearch Agent biomarker workflow (differential expression, pathway enrichment, candidate ranking, volcano plot) over a GEO dataset or synthetic data. Use when the user asks to find gene biomarkers or candidate genes for a disease.

bioresearch-causal-evidence

Run the causal-evidence chain (GWAS → eQTL → colocalization → TWAS → fine-mapping → MR) via the BioResearch Agent causal-evidence workflow. Use when the user wants to test whether a GWAS risk locus acts through a candidate gene's expression, identify the credible set, and check MR consistency. Synthetic loci by default; methodology validation, not a real etiological claim.

bioresearch-causal-inference

Run two-sample Mendelian randomization via the BioResearch Agent causal workflow (IVW estimation, leave-one-out sensitivity, scatter / funnel plots). Use when the user asks to test a causal effect between an exposure and an outcome (e.g., BMI → Type 2 Diabetes) using genetic instruments.

bioresearch-differential-expression

Perform differential expression (DEG) analysis via the BioResearch Agent biomarker workflow, which computes DEGs (t-test + Bonferroni correction) as its first stage. Use when the user specifically asks for differential expression, DEGs, fold-change, or a volcano plot for a disease or gene-expression dataset.

bioresearch-disease-case-study

Run a reproducible, real-public-data disease case study that composes the literature → biomarker → causal workflows end-to-end and emits a blind benchmark evaluation (known-gene recovery, pathway sanity, reproducibility hash). Use when the user wants to prove the framework on a real disease (e.g. Parkinson's GSE7621) or asks for a "case study", "validation run", or "benchmark" of the biomedical workflows.

bioresearch-literature-analysis

Run a reproducible biomedical literature review via the BioResearch Agent literature workflow (PubMed retrieval, entity co-occurrence knowledge graph, structured review outline). Use when the user asks for a literature review, research-gap analysis, or a synthesized reading of PubMed papers on a biomedical topic.

bioresearch-pathway-enrichment

Run pathway enrichment (hypergeometric KEGG / GO) via the BioResearch Agent biomarker workflow, which performs enrichment after differential expression. Use when the user asks for pathway, GO, or KEGG enrichment analysis for a disease or a gene set.

bioresearch-agent-router

Classify a free-text research intent into the right BioResearch Agent workflow (literature / biomarker / causal / case-study / doctor) using a deterministic, rule-based router. Use when an AI client receives an open-ended biomedical research request and must pick which workflow or skill to invoke — NOT to generate or evaluate hypotheses. No LLM, no network.

bioresearch-environment-check

Verify the BioResearch Agent framework installation, Python/dependency versions, network access to biomedical databases, and output-directory permissions. Use when the user asks to validate the environment, troubleshoot a failed workflow run, or before running any analysis.

bioresearch-introduction

Describe the BioResearch Agent Framework and its capabilities to a user or agent. Use when the user asks "what can BioResearch Agent do", wants an overview of available workflows, or is deciding which skill to invoke for a biomedical research task.