Skip to content
View musharna's full-sized avatar

Block or report musharna

Block user

Prevent this user from interacting with your repositories and sending you notifications. Learn more about blocking users.

You must be logged in to block users.

Content in all repositories owned by your account will be closed.
Maximum 250 characters. Please don’t include any personal information such as legal names or email addresses. Markdown is supported. This note will only be visible to you.
Report abuse

Contact GitHub support about this user’s behavior. Learn more about reporting abuse.

Report abuse
musharna/README.md

Hi, I'm Jaret πŸ‘‹

PhD student in Genetics, Bioinformatics & Computational Biology at Virginia Tech. I build open-source tools at the intersection of plant genomics, machine learning, and LLM tooling.

πŸ”§ What I'm building

data-aggregator-mcp PyPI β€” a unified research-data MCP: search & fetch datasets across Zenodo, DataCite, NCBI omics (GEO/SRA), and literature (PubMed/OpenAIRE) behind one normalized model. uvx data-aggregator-mcp

plant-genomics-mcp PyPI β€” 50 genomics tools across 23 backends (Ensembl Plants, Phytozome, UniProt, AlphaFold DB, InterPro, JASPAR, AraGWAS, KEGG, NCBI BLAST, …) with cross-source synthesis.

plantcv-mcp PyPI β€” PlantCV as a measurement instrument: plant trait numbers and the segmentation overlay they were computed from, and it refuses to return numbers when the segmentation is degenerate.

breedsim-mcp PyPI β€” breeding-scheme simulation over AlphaSimR, built on one rule: a single stochastic run is not a result. Returns per-cycle mean, sd and CI, with no flag that collapses it to a point estimate.

jobd PyPI β€” a self-hostable, GPU-aware job broker for your own machines: capability-tagged routing, and the queue exposed natively to agents over MCP.

ghostcite PyPI β€” deterministic, no-LLM detection of ghost citations: catches author/year that don't match the DOI's CrossRef record, plus retraction flags. Ships as a pre-commit hook and a GitHub Action.

llm-panel PyPI β€” ask several LLMs independently and read every answer in full: parallel judges, an anonymized rebuttal round, and cost accounting.

phylokit-mcp PyPI β€” phylogenetic inference over IQ-TREE 2 as an MCP server; never returns a topology without its bootstrap support.

ldraw-mcp PyPI β€” renders LDraw/LEGO models to images with real part geometry via headless Blender, for vision-capable models.

tmodloader-mcp PyPI β€” drive a running tModLoader instance from an agent: launch it, query it, screenshot it, read its state back as structured data.

🌱 Also working on

Long-tail CLIP for orchid ID (5,000+ species) Β· StyleGAN2-ADA for Cattleya Β· RNA-seq differential expression Β· auxin / ARF genomics.

🧰 Stack

Python Β· R Β· PyTorch Β· MCP / LLM tooling Β· Bioconductor

πŸ“« Find me

musharna.github.io Β· Hugging Face Β· ORCID 0009-0003-4055-5238

Pinned Loading

  1. data-aggregator-mcp data-aggregator-mcp Public

    Unified research-data acquisition MCP β€” search & fetch datasets across Zenodo, DataCite, NCBI omics (GEO/SRA/BioProject), and literature (PubMed/OpenAIRE) behind one normalized model.

    Python 2

  2. plant-genomics-mcp plant-genomics-mcp Public

    Plant genomics MCP server β€” 50 tools across 23 backends (Ensembl Plants, Phytozome, UniProt, AlphaFold DB, PDBe, InterPro, JASPAR, PANTHER, OrthoDB, AraGWAS, NCBI BLAST, Gramene, KEGG, STRING-DB, A…

    Python 4

  3. jobd jobd Public

    Self-hostable GPU-aware job broker for your own machines, with native MCP/agent integration

    Python 4

  4. ghostcite ghostcite Public

    Deterministic, no-LLM CLI that catches ghost citations β€” when the author/year you cited doesn't match the DOI's CrossRef record (+ retraction flags).

    Python 1

  5. breedsim-mcp breedsim-mcp Public

    MCP server for breeding-scheme simulation (AlphaSimR) β€” returns distributions, never a single stochastic run

    Python

  6. plantcv-mcp plantcv-mcp Public

    MCP server exposing PlantCV as a measurement instrument β€” plant traits plus the segmentation overlay they were measured from

    Python