Posts

Capsida's Trailblazing Moment: What the Field Owes the Next BBB Program

-
Image
  By Fatma Elzahraa Eid , PhD | TheBioMLClinic Brief disclosure: I am a named inventor on patents and author on publications related to AAV capsid engineering and CNS gene delivery, developed during my time at the Broad Institute. I now operate independently. This post does not represent any prior employer, current advisory client, or collaborator. The mechanistic analysis presented here is my own scientific interpretation of publicly available data. Full disclosures at the bottom. A child died. Brain swelling (cerebral edema) within days of receiving a gene therapy that had cleared every preclinical safety bar the field knows how to set. The IND had FDA clearance. The Good Laboratory Practice (GLP) toxicology study showed no adverse histopathology. The capsid achieved greater than 70% neuronal transduction across critical brain areas in nonhuman primates. By every metric the field currently uses, this program was ready. And yet. This post is not about assigning blame. Cap...
5 comments
Read more

AAV-ML for Experimentalists #6: Working With ML Teams, What AAV Scientists Actually Need to Know

-
Image
TL;Dr Do not set your collaboration to failure from Day 1.  Learn how to speak with ML teams.  The typical AAV-ML collaboration failure scenario goes like this:  The collaboration looked good on paper. An ML team with a strong publication record, an experimental team with three years of capsid library data, a shared interest in CNS targeting. The first few meetings were energizing, the kind where everyone leaves feeling like something real is about to happen. Six months later : a ranked list of 200 variants, a brief methods section explaining that a transformer-based model was trained on the pooled screen data and optimized for predicted transduction scores, and a looming decision about which 20 variants to actually synthesize.  Nobody can clearly explain why variant 47 ranked above variant 12. Nobody wants to be the one to say that out loud. The experimental team is wondering whether to trust the list or quietly fall back on their own instincts. The ML team is wait...
Post a Comment
Read more

Why AlphaFold Won't Engineer Your Next AAV Capsid

-
Image
TL;DR   Structure prediction tools answer a question about shape. AAV capsid engineering requires answers about function. These are not the same question, and confusing them is how engineering programs fail. The conversation that prompted this article A gene therapy program recently asked me to advise them on a potential computational collaboration. The pitch from the computational side was straightforward: they would use AlphaFold-like tools to identify functional AAV capsid variants. Not as a starting point for further screening. Not as a filter to eliminate obvious failures before experimental testing. As the primary discovery engine. Zero-shot. Predict the structure, identify the functional capsid, done. I did not advise them to proceed. Not because the computational team was incompetent. They were not. Not because structure prediction tools are bad. They are not. But because the entire proposal was built on a category error: the assumption that predicting what a capsid looks l...
Post a Comment
Read more

AAV-ML for Experimentalists #5: How to Assess ML Claims in AAV Without Being an ML Expert

-
Image
TL;DR   The scientist who knows when to trust ML results, and when to push back, is the most useful scientist in the room. Be that Scientist.  There is an as ymmetry built into every conversation between an ML practitioner and an experimental AAV scientist, and it works against you. ML practitioners report results in a language that was not designed for experimental decision-making: AUC, precision-recall, held-out accuracy, latent space coverage. These are real concepts with real meaning, but they are also words that land with the weight of rigor even when the underlying work does not earn that weight.  A 90% accuracy sounds like a good thing.   AAV scientists are trained, correctly, to trust quantitative results. You were taught that numbers mean something. That discipline is a virtue. In this context, it is also a vulnerability. The result is a field where genuinely useful ML work gets adopted uncritically because the metrics looked good, and where other genuinely ...
Post a Comment
Read more