Sign In

Simon Barnett

Research Director @ Dimension

Simon Barnett is a highly qualified professional with a Bachelor's Degree in Chemical and Biomolecular Engineering from The Johns Hopkins University.

He has an extensive background in research, life sciences, genomics analysis, finance, risk management, engineering, and education.

Simon has held key positions in reputable organizations such as Dimension, ARK Investment Management LLC, Aditus Holdings, 3 Day Startup, The Johns Hopkins University, Tissue Analytics, Inc., InstaEDU, Nebulab Technologies, Oncopath Laboratory, and START-South Texas Accelerated Research Therapeutics.

Highlights

Nov 12 · twitter

It's 2025. I've heard that "binder design is a solved problem" dozens of times this year.

I don't necessarily agree. Further, I think there are two reasons why this belief has manifested.

Our benchmarks are saturated and our test/train splits aren't rigorous.

IMO - sequence-based T/T splits, lack of time-based cut-offs, and more - all contribute to data leakage and over-stated generalization claims.

I don't really know what to do with RMSE anymore. Biophysically-determined affinity or virtual hit rates feel a bit closer to reality, but not everyone reports these.

I remember when sequencing was this way. Sequence read accuracy was in linear-scale for a long time, at least until Illumina SBS brought forth the Q30 standard, which forced everyone to start thinking about reporting accuracy on a log (Phred) scale.

It took a while, but now everyone does this. Small variations in error mode or quantity are much, much more obvious in log-scale, hence why no one wanted to show it for a while!

Nov 12 · twitter

It has been over five years since DNA foundation models took their perch atop the pantheon of polarizing biology topics.

I’ve been waiting. Watching. Eager to see the first set of models pre-trained on or fine-tuned with cutting-edge sequencing data instead of the shotgun reads of yore.

Have I missed it?

What happens when we replace monochrome, fragmented reads with modern, technicolor data—large phase blocks, 5mC/5hmC, structural variants, tandem repeats, linked regulatory regions, isoform-resolved transcripts…

What secrets can micro-satellites and ancient higher-order repeats tell us?

Is it nothing? Is it something? I am so eager to know.

Simon Barnett
Add to my network

Location

New York City Metropolitan Area