aristoteleo

Drive the user's Atrium desktop: open files in installed viewer apps (Viv bioimages, Vitessce/Spatial 3D omics, Mol* structures, IGV/Gosling genomics, Volume 3D, Cytoscape, MSA, PhyloTree, RDKit), read and steer ANY window — including ones the user opened —…

Skills for querying and downloading data from genomic, transcriptomic, 3D-genome, and cancer-genomics databases. Covers programmatic access to public repositories, gene annotation, sequence retrieval, processed functional-genomics tracks, Hi-C / Micro-C…

Skills for single-cell and spatial omics data analysis. Best practices, code snippets, and workflows for the scverse ecosystem.

Obtain and predict protein 3D structures — fetch AlphaFold predicted models from the AlphaFold DB, experimental structures from the RCSB PDB, or predict a novel sequence with ColabFold — and visualise them in the Mol* desktop app.

Query the Virtual Embryo knowledge graph (mouse/human developmental biology: genes, anatomy, Theiler/Carnegie stages, gene expression, diseases, papers) and its 3D atlas catalog (anatomical OPT/light-sheet volumes + 3D spatial- transcriptomics datasets), and…

Panel design for DEVELOPING / dynamic systems (embryonic organs, differentiation, regeneration). The target experiment is usually a LATE / terminal stage, but the biology is a trajectory: terminal cell types are end-products of earlier lineage programs. A…

General-purpose skills for data analysis infrastructure: workspace file organization, environment management, parallel computing, and performance.

Aesthetic guidelines and output-type recipes for scientific figure production. Supports lightweight default-agent use through SKILL.md + one outputType recipe, with optional venue-specific style guides when requested.

Perform in silico gene perturbation on a dynamo vector-field AnnData to predict cell fate diversion after single or multi-gene activation or suppression, then visualize the results with streamline or quiver plots. Use when running dyn.pd.perturbation,…

Compute least action paths (LAP) between hematopoietic or general cell types in a dynamo vector-field AnnData, then rank transcription factors by MSD along each path and evaluate predictions via ROC analysis. Use when running…

Run or adapt a conventional spliced/unspliced RNA velocity workflow in `dynamo`, including `Preprocessor` preprocessing, `dynamics`, low-dimensional `cell_velocities`, `VectorField`, topology / potential analysis, confidence-based correction, fate prediction,…

Run downstream differential-geometry analysis on a `dynamo` vector-field `AnnData`, including velocity, acceleration, curvature, Jacobian, regulatory-network, ddhodge pseudotime, and state-graph branches. Use when adapting the…

Convert Ensembl-style gene IDs to gene symbols in `dynamo` with `dynamo.preprocessing.convert2gene_symbol` or `dynamo.preprocessing.convert2symbol`, including human and zebrafish IDs, version-suffix stripping, `AnnData.var_names` updates, and optional…

Compare lineage appearance timing and its regulators on a precomputed `dynamo` vector-field `AnnData` using topography, graph potentials, Jacobian, and vector-calculus outputs. Use when checking whether one lineage appears earlier than its peers, curating…

Run or adapt a one-shot total RNA velocity workflow in `dynamo` for metabolic-labeling or scNT-seq `AnnData`, including monocle preprocessing with an optional curated gene list, grouped moments by labeling time, Model-2 `dynamics`,…

Run or adapt dynamo preprocessing with `dynamo.preprocessing.Preprocessor`, including the `recipe` branches `monocle`, `seurat`, `sctransform`, `pearson_residuals`, and `monocle_pearson_residuals`. Use when converting or reproducing…

Create or update a Codex skill that packages reusable workflows, references, scripts, and assets for repeated tasks. Use when turning notebooks, tutorials, analyses, or domain procedures into a triggerable local skill for other agents, or when deciding…

Evaluate the quality of a Codex skill from multiple dimensions and produce a structured score, verdict, and revision guidance. Use when reviewing a `SKILL.md`, scoring a skill folder, comparing two skills, performing cross-review on a newly created skill, or…

Maintainer-side operations for the Chiron platform and its Tabula trainer. Covers backend environment setup, running a worker locally, BioEngine app upload and deployment, and federated training session orchestration.

Single entry point for an AI agent working on the Chiron platform. Covers exploring published Tabula models on the Hub, setting up a Chiron worker on your hardware, launching and monitoring federated training, and adding a new foundation-model trainer beyond…