Bradley Woolf
Bradley Woolf

Bradley Woolf

Programmer

Merging computers, software, and engineering principles with biology.

Projects

  1. How to design your own medicine with software, synthetic DNA, and a bioreactor

    The main reason for my monomaniacal obsession with computers + programming in 2018 was due to a synthetic biology class I took. It changed my life. In that…

  2. How to sequence your own DNA at home

    I have now sequenced my own genome 5 times with an Oxford Nanopore Technologies MinION. This means collecting them from a swab, prepping them for…

  3. Designing drugs with a quantum computer

    With every new hardware architecture comes an accompanying software architecture, and when the two meet, there is a cambrian explosion of invention and…

  4. Calculating my Polygenic Risk Scores from the Nucleus Origin Model

    I’ve downloaded my personal whole genome sequence from Nucleus and the Nucleus Origin dataset. In this project, I’ll calculate my own polygenic risk scores…

  5. Sampling energy-based models with Extropic’s thrml

    Extropic recently announced their new ML-based chip and sister library, thrml. They’ve basically built an ASIC for energy-based models, where you are…

  6. GPU-based whole genome analysis

    What information can we actually get from a whole genome once it has been sequenced? There are a lot of categories, but I would personally categorize all…

  7. Building a VAE cell-type annotator

    In this project, we build a model that can annotate cells by their type. Kinda like Yolo8 but for labeling cells with their names. A VAE is just a sub-type…

  8. CPU-based whole-genome analysis

    I had my genome sequenced by Nucleus and received 54 GB of compressed files. This turns into 273GB when you decompress it, which is needed to run any…

  9. Fine-tuning Evo2 to generate novel phage designs

    Today, we are recreating the paper “Generative design of novel bacteriophages with genome language models” from the Arc Institute. I am assuming the reader…

  10. Fine-tuning toolkit for ESM2

    In this project, I fine-tuned ESM2 on the CAFA3 labeled proteins and annotate them with the Gene Ontology labels. The goal is to predict labeled protein…

  11. Gene therapy pipeline (in silico)

    Gene therapy is expensive. But boy when it works, it works. We want to help drop these insane costs. The scAAVengr pipeline is an AAV delivery pipeline that…

  12. Frances: optimizing E. Coli gene edits → output yields

    There are two primary focuses in cell engineering: designing cells to output a specific molecule, and designing cells to respond specific ways in specific…

  13. K-mer counting CUDA kernel

    This is a brief write-up for a brief project. My goal with this was to learn what CUDA kernels are and when you would need to build one. The algorithm I…

  14. scRNA-seq Leukemia cell pipeline with DL

    I attended SynBioBeta in May where I briefly spoke with Daniel Ives, the founder of Shift Biosciences, who was an extreme AI skeptic. He mentioned that…

  15. Ranking proteins with reinforcement learning

    This was a great project, but took way longer than I anticipated. 4 days execution vs 1 day expectation. I read part of Sutton’s book last year and thought…

  16. Automated variant interpretation (FASTQ → Clinic)

    Genetic testing has grown exponentially in clinical settings and variant calling has now become an industry in its own right. Tools like Fabric Genomics…

  17. cfDNA fragmentomics with Python

    This project is essentially a parsing script wrapped in a Python pipeline for BAM files that helps distinguish cancer from normal samples based on fragment…

  18. Automated scraping from UniProt, PubMed, Reactome, and KEGG

    Ok so this project started out as being pretty boring to me, but quickly became interesting and quite useful. One big difference between programmers in…

  19. Bioinformatics code challenges with Python

    This script splits a DNA sequence (e.g. a plasmid) into 3 PCR fragments that are: 400-800 bp long, connected by 20 bp overlaps for Gibson Assembly…

  20. A soil microbiome pipeline with Nextflow, Kraken2, and MultiQC

    Soil is alive, with billions of microbes driving plant health, carbon cycling, and ecosystem resilience. But raw sequencing data doesn’t tell that story…

Writing

  1. Do performance-based biotech bompanies outperform traditional biotech?

    Biotechnology has traditionally been an industry where companies raise large amounts of capital, spend years developing drugs, and hope that a small number…

  2. What can I do with my DNA?

    The most common question I have seen from people is “what can I do with my DNA?” As dimenwarper and Bryan Johnson put it on X:

  3. The $0 genome

    For years, we have been hearing about the $100 genome. Two systems in particular, the BGI Genomics DNBSeq and Element Biosciences VITARI both claim the…

  4. Representing the human body as a state of trajectories

    One of the most interesting things I have noticed in biology is that the biologists is essentially a classifier. They name, identity, categorize, and move on.

  5. Sids Osteosarcoma data as mathematical functors

    One theme I keep thinking about throughout my molecular biology experiments is that when we run an assay (some molecular measurement technique), we end up…

  6. Where is all the dreamy sci-fi for biotech?

    I’ve been consuming electronics my whole life, thanks mainly to my dad. Despite being a real estate salesperson, he was a fervent believer in technology,…

  7. Technological progress

    Across many domains, the history of technology looks like repeated cycles of:

  8. Cost structures

    The way PayPal/X grew to millions of users is they gave $10 to each new user. This is effective marketing, especially if you make a fee on when they spend…

  9. Genes from the sci-fi book Upgrade

    In the book, there are tons of genes that are discussed. I list the ones mentioned below, more for my own reference, but I figured I would share. Most of…

  10. How each programming language thinks
  11. Shaping a neural network like marble

    Building a neural network has a lot of components, and it can become overwhelming when trying to decide what to use when, where, how, and why. I struggled…

  12. Regulations for biotech agents

    Note: a bigger fix to the problems these seek to address could actually be fixed with federated architectures/homomorphic encryption. It’s coming but my god…

  13. The core primitives all computer programs are built on

    After building a lot of software for the past few months, it’s gotten to the point where it is predictable. Sure, I need to fix a lot and go deeper on…

  14. Who can I raise money from for my biotech startup?

    I am creating a new compute-driven biotech company. I need to raise $50M. This $50M raise might look like:

  15. How do I leverage LLMs to program better?

    9/21/25

  16. Cost modeling DL vs linear regression paper

    In this article, I calculate the approximate costs of this paper. In the paper, they benchmark the performance of scGPT, scFoundation, scBERT, Geneformer,…

  17. Using AI game engines to simulate synthetic biology

    A lot of the work I have done over the past few months has somehow led me to game engines. Specifically, rendering engines using AI architectures.

  18. Can deep learning solve bottlenecks in gene therapy?

    (aka a fully vertically integrated, end to end, capture → modeling → product, system)

  19. BioRAG

    I came across this great repo that builds an entire RAG ecosystem from scratch. I wondered if we could do the same for multimodal biological datasets,…

  20. Gene therapy: can we perceive the body as an RL agent’s environment?

    Humans, individually, are essentially a neural network. We are each a foundation model. Most of the data, in my opinion, is inherited through our genome. We…

  21. GPU programming and custom kernels

    A brief post on how GPUs work, examples of when to use them for the layperson, and CUDA kernels.

  22. Core crypto algorithms

    What: Algorithms that generate proofs that something happened. Can take minutes to hours. Can leverage GPUs for parallelizable arithmetic, FPGAs for custom…

  23. Core genomic algorithms

    What: Algorithms that identify differences (mutations) in DNA compared to a reference genome.

  24. The Quantification of biology

    The quantification of biology, in my opinion, creates one of the largest opportunities for social good in human history. If you look at many of the…

  25. The technical bottlenecks that high frequency trading overcame

    Every great industrial revolution faced bottlenecks for achieving its eventual success. An area of technical accomplishment that is oftentimes overlooked,…

  26. Making illiquid data liquid

    Science is the collection of information, application of models to that information, and deduction of patterns from that collection. We then use these…

  27. A brief history of quantifying biology

    There is a lot written on the history of genetic reading, editing, and synthesis, but not much has been written on the quantitative modeling of biology.…

  28. Filtering nature through the dominant technology of the day

    Throughout history, humans have consistently used the dominant technology of the day to describe processes in nature. These are limiting but also very…

  29. Commoditizing health

    Due to my goal of improving the performance of some biotech algorithms 1000x, I have come to the conclusion that we live in the stone age for health. I…

  30. History of Technology

    Goal of the study:

  31. History of Health

    Health is one thing:

  32. High Level Global Statistics

    Conclusions: The internet is almost fully saturated(70%), save India, some of China, and Africa. This means that people now have online identities and can…

  33. Scientific Databases