Skip to content
View parshawn's full-sized avatar

Block or report parshawn

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.

Maximum 250 characters. Please don't include any personal information such as legal names or email addresses. Markdown supported. This note will be visible to only you.
Report abuse

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

Report abuse
parshawn/README.md

Hi, I'm Parshawn Gerafian

Bioengineering + Computer Science @ UC Berkeley
I work on turning biological data into systems we can model, understand, and use.


🧬 About Me

  • Research

    • Stanford: Single-cell foundation models, cross-species alignment, and interpretability
    • Ian Holmes Lab (UC Berkeley): Sequence modeling for transposons using transformer-based approaches
    • NASA GeneLab (ADBR AWG): Multi-omics pipelines for studying brain resilience and neurodegeneration
    • MD Anderson Cancer Center: Deep learning for clinical imaging and cancer treatment analysis
  • Engineering

    • Building reproducible pipelines for large-scale biological data
    • Working across cloud and HPC environments
    • Focused on clean systems, evaluation, and usability
  • Interest

    • AI in biology and healthcare
    • Foundation models + multi-omics data
    • Making biological research more computational and scalable

⚙️ Skills

Languages
Python · Java · SQL · R

Machine Learning
PyTorch · TensorFlow · JAX · scikit-learn · Hugging Face · Weights & Biases

Bioinformatics
Scanpy · AnnData · Biopython · ViennaRNA · Bowtie2 · FastQC

Data & Infrastructure
Pandas · NumPy · Cloud · HPC (Slurm) · CUDA · Git · FastAPI


🚀 Current Focus

  • Fine-tuning and evaluating single-cell foundation models
  • Cross-species representation learning
  • Building scalable pipelines for multi-omics data
  • Improving model interpretability in biological systems

🎯 Goals

Build tools that make biological data easier to work with
Bridge machine learning and real-world biomedical applications


🔗 Connect

Popular repositories Loading

  1. datasciencecoursera datasciencecoursera Public

    First R repository

    Rebol

  2. ihh.github.io ihh.github.io Public

    Forked from ihh/ihh.github.io

    Holmes lab website

    HTML

  3. bilby_encoder bilby_encoder Public

    Forked from ianholmeslab/bilby_encoder

    BED-FASTA-BAM one-hot data encoder method utilizing state transition tuples.

    Python

  4. bioe131-finalproject bioe131-finalproject Public

    Shell

  5. bioe131-finalproject-web bioe131-finalproject-web Public

    JavaScript

  6. parshawn parshawn Public

    Config files for my GitHub profile.