Skip to content
View sashachernenkoff's full-sized avatar
1 follower · 2 following
  • canada

Achievements

Achievement: Pull Shark

Achievements

Achievement: Pull Shark

Block or report sashachernenkoff

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
sashachernenkoff/README.md

Hi, I'm Sasha 👋

I'm a machine learning scientist working on transfer learning, representation learning and model interpretation for genomic data. I fine-tune large pretrained models for new prediction tasks and use in silico perturbation to study what they learn. I also build models that learn compact representations of high-dimensional genomic data. My work is in PyTorch, with distributed training (DDP), parameter-efficient fine-tuning (LoRA) and custom loss functions. A background in clinical diagnostics informs the problems I work on.

🌱 Currently learning: cloud infrastructure for ML training and deployment (AWS)

📫 sjchernenkoff@gmail.com · sashachernenkoff.com · LinkedIn

Pinned Loading

  1. pastl pastl Public

    Foundation model empowered transfer learning (PyTorch, custom biologically-gated attention head) to predict cis-regulated protein abundance. Analysis suite for interpretability and validation inclu…

    Python 1

  2. herl herl Public

    Deep generative modelling (PyTorch, categorical VAE) and representation learning of high-dimensional genotype data for heritability estimation

    Python

  3. nf-germline-variant-calling nf-germline-variant-calling Public

    GATK best-practices germline variant-calling pipeline built in Nextflow. Originally validated on CPTAC LUAD (~230 samples); public version runs on GIAB reference data. Containerized, resumable, por…

    Nextflow 1

  4. variant-prior-pwas variant-prior-pwas Public

    A generalized pipeline for evaluating variant effect predictors by using their scores as biological priors in TWAS/PWAS

    Python 1

  5. DL_cancer_drug_response DL_cancer_drug_response Public

    Forked from davidenoma/DL_cancer_drug_response

    Predicting patient drug responses using gene expresions, drugs SMILES and advanced deep learning techniques. My contributions: LUAD-specific data pipeline and a multimodal CNN + LSTM architecture (…

    Jupyter Notebook

  6. MDGE612project MDGE612project Public

    Predicting phenotype from genotype: a comparison of Lasso and Ridge regression

    Python 1