Jonathan Harrison
@juharrison.bsky.social
AI for drug design
Interested in, among other things, Bayesian Optimization and active learning.
Currently at Recursion (formerly Exscientia). Previously postdoc at Warwick and PhD at Oxford.
Interested in, among other things, Bayesian Optimization and active learning.
Currently at Recursion (formerly Exscientia). Previously postdoc at Warwick and PhD at Oxford.
Reposted by Jonathan Harrison
getting in an arms race with LLMs over em-dashes by using dashes so long ——— so very long ——— that they have never appeared in training data
July 7, 2025 at 3:21 AM
getting in an arms race with LLMs over em-dashes by using dashes so long ——— so very long ——— that they have never appeared in training data
www.kaggle.com/competitions...
Kaggle competition on molecular property prediction for polymers
Kaggle competition on molecular property prediction for polymers
NeurIPS - Open Polymer Prediction 2025
Predicting polymer properties with machine learning to accelerate sustainable materials research.
www.kaggle.com
June 17, 2025 at 6:54 PM
www.kaggle.com/competitions...
Kaggle competition on molecular property prediction for polymers
Kaggle competition on molecular property prediction for polymers
Reposted by Jonathan Harrison
Excited to unveil Boltz-2, our new model capable not only of predicting structures but also binding affinities! Boltz-2 is the first AI model to approach the performance of FEP simulations while being more than 1000x faster! All open-sourced under MIT license! A thread… 🤗🚀
June 6, 2025 at 1:46 PM
Excited to unveil Boltz-2, our new model capable not only of predicting structures but also binding affinities! Boltz-2 is the first AI model to approach the performance of FEP simulations while being more than 1000x faster! All open-sourced under MIT license! A thread… 🤗🚀
Reposted by Jonathan Harrison
Are you looking for a postdoc position in #SpatialBiology?
We're advertising roles in both mathematical method development and in spatial data analysis @ox.ac.uk in colorectal precancer spatial biology!
Method Development: tinyurl.com/4vcbdjcy 🧮
Data Interpretation: tinyurl.com/yzsp729s 🧪
We're advertising roles in both mathematical method development and in spatial data analysis @ox.ac.uk in colorectal precancer spatial biology!
Method Development: tinyurl.com/4vcbdjcy 🧮
Data Interpretation: tinyurl.com/yzsp729s 🧪
Job Details
my.corehr.com
April 14, 2025 at 9:55 AM
Are you looking for a postdoc position in #SpatialBiology?
We're advertising roles in both mathematical method development and in spatial data analysis @ox.ac.uk in colorectal precancer spatial biology!
Method Development: tinyurl.com/4vcbdjcy 🧮
Data Interpretation: tinyurl.com/yzsp729s 🧪
We're advertising roles in both mathematical method development and in spatial data analysis @ox.ac.uk in colorectal precancer spatial biology!
Method Development: tinyurl.com/4vcbdjcy 🧮
Data Interpretation: tinyurl.com/yzsp729s 🧪
No foundation without foundations - why semi-mechanistic models are essential for regulatory biology arxiv.org/abs/2501.19178
No Foundations without Foundations -- Why semi-mechanistic models are essential for regulatory biology
Despite substantial efforts, deep learning has not yet delivered a transformative impact on elucidating regulatory biology, particularly in the realm of predicting gene expression profiles. Here, we a...
arxiv.org
February 5, 2025 at 9:29 AM
No foundation without foundations - why semi-mechanistic models are essential for regulatory biology arxiv.org/abs/2501.19178
Reposted by Jonathan Harrison
Expediting hit-to-lead progression in drug discovery through reaction prediction and multi-objective molecular optimization | ChemRxiv - doi.org/10.26434/che... #compchem
Expediting hit-to-lead progression in drug discovery through reaction prediction and multi-objective molecular optimization
The rapid and economical synthesis of novel bioactive compounds remains a significant hurdle in drug discovery efforts. This study demonstrates an integrated medicinal chemistry workflow that effectiv...
doi.org
January 29, 2025 at 8:22 AM
Expediting hit-to-lead progression in drug discovery through reaction prediction and multi-objective molecular optimization | ChemRxiv - doi.org/10.26434/che... #compchem
Reposted by Jonathan Harrison
Active Learning Enables Extrapolation in Molecular Generative Models arxiv.org/abs/2501.02059 #compchem
Active Learning Enables Extrapolation in Molecular Generative Models
Although generative models hold promise for discovering molecules with optimized desired properties, they often fail to suggest synthesizable molecules that improve upon the known molecules seen in tr...
arxiv.org
January 8, 2025 at 9:24 AM
Active Learning Enables Extrapolation in Molecular Generative Models arxiv.org/abs/2501.02059 #compchem
Reposted by Jonathan Harrison
Tomorrow, in the reading group, Johannes Brehmer will present his paper "Does equivariance matter at scale?" arxiv.org/abs/2410.23179
Join us on zoom at 12pm ET / 6pm CET: portal.valencelabs.com/logg
Join us on zoom at 12pm ET / 6pm CET: portal.valencelabs.com/logg
January 5, 2025 at 8:08 PM
Tomorrow, in the reading group, Johannes Brehmer will present his paper "Does equivariance matter at scale?" arxiv.org/abs/2410.23179
Join us on zoom at 12pm ET / 6pm CET: portal.valencelabs.com/logg
Join us on zoom at 12pm ET / 6pm CET: portal.valencelabs.com/logg
Reposted by Jonathan Harrison
Happy 1^3+2^3+3^3+4^3+5^3+6^3+7^3+8^3+9^3, everyone!
January 1, 2025 at 7:00 AM
Happy 1^3+2^3+3^3+4^3+5^3+6^3+7^3+8^3+9^3, everyone!